Hacker News new | comments | show | ask | jobs | submit login
Mozilla Acquires Pocket (blog.mozilla.org)
1187 points by qdot76367 204 days ago | hide | past | web | 445 comments | favorite



Mozilla is growing, experimenting more, and can acquire startups.

Mozilla doesn't have the resources to continue with Thunderbird.

I am increasingly baffled by their decisions and how they relate to the strategic plans [0] they've been producing for a while. Despite the worthy words in their plan they seem to have no sense of direction. That saddens me.

That said I'm happier having Pocket as an open source part of Mozilla/Firefox than a surprise integration of a commercial app.

[0] https://wiki.mozilla.org/MoFo_2020


Thunderbird doesn't make money. Pocket makes money.

That Mozilla is a Corp owned by a not-for-profit isn't the real thing anymore (ie not what people may believe, not that they have any obligation beside following the rule of law) - it just frees the Corp from having to report to random share holders - instead, the Corp and the Org self-select a Board of directors that they report to.

At the end of the day, it's otherwise now similar to many corporations.


Mozilla needs products that are relevant in the marketplace, to exist. Revenue is part but not all of this.

I don't think Mozilla can make meaningful impact in a world where it doesn't make products that internet users want to use. It would be a largely toothless advocacy organization at that point.


Millions of people want to use Thunderbird! I would even pay some amount for an "officially supported edition" of Thunderbird.


Would you be willing to pay that amount for an open source project without the official backing of Mozilla? Because I was under the impression that the source code to Thunderbird was, in fact, available under an open source license.


Yeah, I might be, actually. It would depend on who was involved and how much I trusted those devs, but at least conceptually, sure.


Why not postbox ?


never heard of it, looked at "Download", saw a big apple and a big window, no penguin. fail. also closed source AFAICT. thunderbird would be an incredible email app if only it had better maintenance.


Well it does the job already, working just fine for the 31-years-old IMAP protocol, the 33-years-old POP protocol, the 18-years-old RSS protocol, etc.. I might like an "incredible" email app but all I really need is a "sufficiently solid" one.. =)


You can see that just by looking at their job postings. The experimental ads team was hiring and then stopped and then the team was absorbed into other teams. Now they have more postings for project managers. They're basically introducing more hierarchy and traditional corp. stuff from the sounds of it.


Having more project managers is how things get done. I see it everyday at my company...


I had a coworker that used to say that when you have more managers than toilets, something is clearly wrong. It was a rule of thumb in his experience..


That is beautiful, but I need more data.

1. Do project managers, product managers, etc. count, or only real managers with teams reporting to them?

2. Do urinals count, or only real sit-down toilets?


Urinals MUST count. Assuming that women's bathrooms have more sit-down toilets than men's, we are biased. It is unfair and sexist to allow female dominated companies more than their fair share of project managers!

/s


Do women's bathrooms have more sit down toilets than men's?


Generally, though less toilets in total because you can cram more urinals into the same space. A common small office toilet space has room for three stalls so in the female room that is what you'll find. In the male variant the space of one of the stalls will instead be given to two individual urinals or a small "urinal wall" that comfortably houses two people or you can get three into if people are desperate. Larger rooms offer greater flexibility, but ultimately urinals take less space than stalls so you get more of them.

Of course that doesn't always mean more people going at the same time as some people don't like urinating so close to someone else so the second spot goes unused, and some of us prefer to sit for a no.1 anyway unless in a rush.


Usually, yes, in the space occupied by urinals in the men's toilets. But you can fit more urinals in that space.


I don't think I've ever been somewhere with more toilets than managers, that probably supports the assertion though.


I can't tell if you're for real or just being sarcastic.


More meetings help too :)



Clearly sarcasm.


you might have not had more than one manager then :)


Product managers and hierarchical reporting structures are nothing new to Mozilla, quite the contrary.


how does pocket make money? they do have pocket premium: https://getpocket.com/premium?ep=2 but I never saw them advertising it and never thought about using it and I'm using pocket heavily past several months. Is "pocket premium" their main source of income? Is it bringing real dollars?


They slot 'sponsered content' blocks into your stream of saved items, it's done very well IMO. These have to make some decent $$$ surely.

However, I just checked my queue and couldn't see any.


> Pocket makes money.

Just pocket money though.


Do you have a source? Revenue and profit are completely different things, and I am curious if Pocket was making any profit. Otherwise it's losing money.


I suspect it's about the longer term profits, like any other corporation purchase. Generally, you can't buy/afford it when it's already super successful, but you can when it's on the verge of being that (but needs your funding/image/resources).


Normally companies raise venture capital in the cases you describe.

Acquisitions happen when the stakeholders (founders, employees, and investors) don't think they will get the level of return on investment they will be satisfied with.

Even if they DO decide to sell, the case you describe (on the verge of being successful but requires money to get to that point) sell at a disclosed amount. Undisclosed amount mostly means the investors and founders are not proud of the result so they would rather just keep it private.


Just a question on top of my head, could Mozilla make a SaaS, like a email services ( example only ), and have those money go into continuing develop Firefox.

And as a non-profit entity, could they retain some profit they have as investment or cash somewhere. In case some day they run into negatives they could use those as reserves?

But at the end of the day, Mozilla need some very serious management restructuring, starting from the very top end. Having been a Netscape user from over two decade ago, I finally left Firefox. And if you follow their decision, mailing list etc, you can feel how bureaucratic they are. Slow response to market changes, completely lack of ( actionable )vision, often having completely unrealistic target. Extreme inefficiency in development of its product. And these aren't my thoughts or comment only, their are many ex mozillian employees who felt the same. We all thought Mozilla should have been a great company achieving far then then they are today.


If the NFL can call themselves a nonprofit then anyone else should be able to do the same.


The NFL is a nonprofit by any reasonable definition, even without the explicit exception in 501(c)(6). Or rather, it was, until it voluntarily gave up its tax-exempt status in 2015 supposedly over public opinion about their status.

I think people simply aren't aware that all the major revenue associated with the NFL goes through specific teams, which are normal taxable corporations. Humorously, people also aren't aware that they gave up their tax exempt status, so I'm not sure that decision was worth it.


I surely wasn't aware that they gave up their tax exempt status. Huge props to them.


> Thunderbird doesn't make money. Pocket makes money.

That's not the case anymore with this merger though, right? Pocket will now cease to make actual money.


From the announcement: "Pocket will become a wholly owned subsidiary of Mozilla Corporation"

Mozilla Corporation is a for-profit, taxable entity. Here are their articles of incorporation: https://www.mozilla.org/en-US/foundation/documents/articles-.... It is itself a wholly owned subsidiary of Mozilla Foundation, which is the non-profit thing that people call "Mozilla". The Foundation is really just a holding company. The thing that people call "Mozilla" that does all of the work is The Corporation.


The foundation exists as a separate entity because laws prohibit charitable donations from being used for all but a small number of categories, and software development generally does not fall under such a category. The funds received by the foundation pretty much have to go towards education (or things that can be justified as education, e.g. sponsoring educational conferences or sponsoring attendees to such conferences). Charitable donations can't be used to fund Firefox development.


I really wish there was an easier way to do this, ideally something like a corporation with no shareholders.


That's exactly what a nonprofit is. Nonprofits are corporations like any other, with a few key differences:

- Instead of shareholders, nonprofits have members. Members don't own the corporation (no one does) but control it insofar that they hire-and-fire its directors.

- Memberships normally can't be sold, and nonprofits have no equivalent to dividends.

- There are stronger conflict-of-interest protections, preventing officers from signing cheques to themselves or otherwise using the corporation's resources for their personal benefit - at least without someone else's approval.

A typical setup for nonprofits is to tie memberships to being a director, so that current board members choose their own successors. This is known as a "self-perpetuating board".


IIRC non-profits can't make and keep profits. Corps can, hence Mozilla Corp. They can keep profits around in order to.. like, purchase Pocket, or simply pay employees later when Mozilla makes less profit, or whatever else.

These things are hard or maybe impossible in a non-profit (IANAL/correct me if I'm wrong)


> IIRC non-profits can't make and keep profits.

Non-profits can make surplus revenues over costs and retain them; they can't return profits to shareholders or other particular beneficiaries.

Certain classes of nonprofits are restricted from certain business activities, or limited to certain activities. E.g., charities (501c3 nonprofits) must be organized and operated exclusively for purposes on an list of charitable purposes.


Yes, nonprofits can charge money and hold onto it. What they can't do is distribute it to shareholders. Many universities are nonprofits, and they hold on to millions or billions of dollars. Nonprofits can also acquire companies, etc.


The terms "nonprofit" and "not-for-profit" are unfortunate misnomers. Like any other company, nonprofits can offer gainful employment and will eventually bankrupt if they never turn a profit.


Yes, something like that but without the need to create a separate foundation and corporation.


AFAIK, Mozilla has the structure they do because the IRS frowns upon charities (which is shorthand for "tax-exempt corporation") engaging in business-like activities but is fine with charities owning for-profit companies as a source of income (and like any shareholder can do, direct the for-profit company to benefit its owner).

Regional differences may be relevant here: in Canada, you can incorporate a nonprofit corporation and later apply for charity status (which restricts the company's possible activities in exchange for a 0% income tax rate) but it's completely optional.

In the US, on the other hand, to the best of my knowledge, one has to establish a for-profit entity first and then apply to the IRS to recognize it as a charity (which is a lengthy process). The words "charity" and "nonprofit" are interchangeable in the US for that reason.

If the concept of a non-charitable nonprofit exists in your jurisdiction, you can get the benefits of a nonprofit structure (ie. no owners) without restricting the company's activities, minus tax exemption and being able to issue tax receipts.

For context: I founded a non-charitable nonprofit in Canada; I'm not super familiar with the US side of things so I'm happy to be corrected.


> in Canada, you can incorporate a nonprofit corporation and later apply for charity status

The US works similarly. You incorporate as a nonprofit, but you have to separately apply for tax-exempt status afterwards. If you don't, you're a nonprofit without the benefits of being tax-exempt.


A corporation is a legal entity, so to do this I imagine you must alter the country's legal code first?


The people that complain about mozilla stopping development on thunderbird seem to be the same people that complain about things like firefox moving to WebExtensions, and Firefox changing their UI, and Firefox integrating Pocket.

I don't understand why people are disappointed that Mozilla has stopped actively developing thunderbird - it means it's not going to change. There is an email client that you like, as is, and want to keep using, and you're annoyed because mozilla has promised not to change it?


Does the lack of changes include no security updates? Then yes, one should be annoyed and Thunderbird usage should be discontinued.


No, security updates are still being made. Here's the release notes for recent security fixes in thunderbird:

https://www.mozilla.org/en-US/security/known-vulnerabilities...


Are the security updates done by Mozilla itself or by the volunteers that continue Thunderbird development?


The security patches are largely from Mozilla's Gecko/SpiderMonkey/Toolkit work. The people who create the patches tend to be Mozilla employees. Looking over the 45.x Thunderbird releases I don't see a single Thunderbird specific security fix. [1]

The team that qualifies and does the release work for Thunderbird are volunteers. I believe that the builds are done on Mozilla hosted hardware.

[1] https://www.mozilla.org/en-US/security/advisories/


why do you care where the patches come from?


One might care where the patches come from as a volunteer community could be less reliable over time than the paid Mozilla staff. I know I certainly feel that way.


But if you prefer security updates from paid employees of an organization rather than volunteers, surely a commercial product rather than a free one would be more suitable for your needs?


Not necessarily. I appreciate the values for which Mozilla stands, and those values are entrenched in their products.


So you're looking for an organisation who will use non-free employees to give you a product for free?


I really don't see your point.


It doesn't feel that way anymore.


In what way? I'm wary of some of Mozilla's decisions, but their allegiance to their mission & manifesto doesn't seem at stake.


> One might care where the patches come from as a volunteer community could be less reliable over time than the paid Mozilla staff. I know I certainly feel that way.

The people who are making these complaints are disproportionately those who have no problem using Debian or Arch, so I don't think that's the crux of the issue for them.


Spoken like a manager.

Something is either being actively developed or it's dieing, there is no in between with software, there is no finished state.


Does being maintained count as being actively developed? I think it should. As long as Thunderbird gets security patches, and maybe even the occasional bug fix, I'm happy with it.

I wish more pieces of software would get to the finished state and would slip into maintenance mode.


Probably not. Developers are fickle and few like to do maintenance only work. So even if it's still getting security patches they are likely to be losing the institutional knowledge to continue that in future, not to mention keeping up with things like API changes for gmail integration.


Which Gmail APIs? Has there been an update to IMAP that I'm not aware of?


GMail has been pushing for its custom XOAUTH2 authentication for a while, and password-based authentication can result in security warnings. Password-based authentication also doesn't support 2FA.

By the way - the IETF standardised an OAuth2 SASL mechanism (RFC 7628) a couple of years ago, no idea why GMail doesn't support it.



I'm asking which Gmail APIs would Thunderbird have to keep up with. As far as I know, it never used that API, hence it doesn't have to keep up with it.


Sorry, didn't understand you right.

Basically, the only thing that is kind-of Gmail specific is the auth support (https://bugzilla.mozilla.org/show_bug.cgi?id=849540). If Gmail sometimes in future switches from OAuth2 to some other protocol, the client will have to be updated.


I haven't used thunderbird in a while, have they not added an easy setup for people using the popular online mail servers yet?


It seems that they have. I just started using Thunderbird again after a number of years and the setup to connect to my gmail account has been simplified to confirming the settings on one page with three auto-filled fields.


You don't need an API for that, just a few hardcoded settings that get auto-filled based on the domain.


That certainly sounds nice but it's not true. Or at the very least depends on how you define software. If software is built upon an evolving platform (such as the web) and has a self-supplying revenue stream if it needs one at all, then it can be done and will last.


Even for embedded software though, if something hasn't been looked at for years then the original developers probably don't remember it and may not still be around. If a client want's a feature or tweak then the institutional knowledge is gone and the best bet would be starting from scratch.


Maybe it's something that is of value to a lot of users and a non-profit (http://www.computerworld.com/article/3011418/web-browsers/mo...) might help move that important software in the future. Like a competitive open-source browser, a competitive mail client is a good thing. Not everyone want to use the cloud.


Well, at some point OS compatibility will break, no?


Thunderbird is still under active development, I'm one of their users, being the best email client for the desktop. Unfortunately Thunderbird is also a dead end, being incompatible with Firefox's direction.

Yes, XUL is dead and Firefox is evolving into a Chrome substitute and while people have serious concerns about it, fact of the matter is that Chrome was designed for web apps and almost 9 years after its release Chrome is still the best browser available for web apps. Not sure how many people here remember Chrome's original goals, beautifully illustrated by their comic book announcement, but here's a refresher and try to count the problems that Firefox still has in 2017: https://www.google.com/googlebooks/chrome/index.html

In other words, Firefox is dying because users have been trained to expect security, stability, an interface free of crappy toolbars and extensions you can't uninstall, not to mention running long lived web apps. And while people on this forum will cry out for Vimperator or other poweruser extensions that 99.9% of the world have never heard about, it's undeniable that without evasive action Firefox will become a footnote in only a couple of years.

I'm also a Firefox user and I understand their move away from Thunderbird. For work I use Gmail, personal email is Fastmail, both have really good web interfaces. And while I try using email clients like Thunderbird, the web interfaces are simply good enough and always there. These are web apps btw and guess which browser is best for web apps? It's not Firefox ;-)

And for all I care Mozilla can kill every project they have, if that means Firefox continues to improve. Because it's still the only browser that cares about my freedoms.


As a data point, my webapp (PartsBox.io) works best/fastest in Safari, Chrome is second, Firefox a distant and slow third. So, I'd take issue with the statement that "Chrome is the best browser available for web apps", at least on MacOS, and at least for some web apps (I know WebGL apps work better in Chrome).


But gmail and fastmail don't work offline.


That is only because they haven't yet caught up with the technologies available to them. Are there are email web apps that have adopted Service Workers? Although to be fair that's not going to work well for keeping a large offline archive.


You can have GMail offline by installing the Chrome App.


The only time I find myself offline is when I'm flying. Not that big of a deal.


So you don't need or require backups of your emails offline?


For personal email I don't care that much for backups, I trust Fastmail to do it better than me.

I do use Thunderbird for offline backups, more because of the thought of losing important work email. But without combining it with an actual backup system, that's not actual backup, because IMAP does synchronization and messages deleted on the server translate to messages deleted on the client side as well. So I combine it with Arq Backup, which does daily uploads of everything on my laptop.

I do wonder whether using Thunderbird for backups is wise. I never recovered from a Thunderbird archive. Sometimes I'm thinking that a solution like OfflineIMAP or mbsync/isync might be better.


I could be wrong here, but the thunderbird thing is basically that it was way too much work for little gain -- the way Thunderbird is designed was that it was a kind-of-fork of Firefox, and too many resources were being wasted just keeping the "fork" up to date from changes to internal Firefox APIs.

I'm also not sure of this, but that vision thing you link to is focused on the goals of the Foundation proper (excluding MoCo), which is not all of Mozilla, and many of the decisions fall out of that scope.


> the way Thunderbird is designed was that it was a kind-of-fork of Firefox, and too many resources were being wasted just keeping the "fork" up to date from changes to internal Firefox APIs.

No. Thunderbird was built against something that the Mozilla foundation used to tout as the future of building software in general: XUL. And there also was a XULRunner which was to Gecko/Firefox what Electron is to Blink/Chrome: a way to develop native apps using the browser engine as a UI toolkit. The difference with electron is that it was built with more native-apps facilities, like spawning up Wizard dialogs: https://developer.mozilla.org/en-US/docs/Mozilla/Tech/XUL/Tu... All written in their XML.

It wasn't developed specifically for Thunderbird. Firefox is also built on top of the same technologies. XULRunner versions were released concurrently with new Firefox versions.

The reason why Thunderbird is a dead is because XUL is dead, or rather, will die soon. They are rewriting everything related to UI in Firefox and also terminating the traditional addon API which.. required XUL, in favor of Chrome's addon API.

Thunderbird is not the only notable app written in XUL beside Firefox. There's also Miro, Songbird, Google's adwords editor ( https://support.google.com/adwords/editor/answer/106323?hl=e... ) and many others.

Most software written in XUL has either been abandoned or going to die, anyway.


XUL was the future of old-school, heavy desktop applications. But people generally aren't interested in those anymore. At least HTML and the rest of the web stack work OK on mobile platforms.


> XUL was the future of old-school, heavy desktop applications.

The old-school, heavy desktop apps seem to be so unbelievably light in comparison to current electron-based craziness. But I guess we deserve all the bloat for not being able to come up with sensible common API for desktop development :(


Thunderbird didn't lose to an Electron-based app, it lost to Gmail.


Yes, but we're talking about XUL the platform. That platform was built with the same use-case in mind that Electron has: allowing people to build cross-platform desktop applications with web technologies (html, js etc).

Unfortunately Mozilla failed to execute on that vision, leaving the field open for Electron and friends to emerge. Part of that failure was due to bad technological choices (RDF-XML was terrible), part to the unwieldy Mozilla legacy (the build system was notoriously byzantine) and part to them de-prioritizing anything that Firefox did not need. They had a working general-purpose JS-based UI runtime more than 15 years ago, and still the rise of nodeJS and html-based toolkits passed them by pretty spectacularly - because they had eyes only for Firefox and vanity projects like FFOS.


But lots of people did write apps in XUL. My first IRC client and my first FTP client were both XUL apps. I'm not convinced that Electron today is any more popular than XUL was back then (aside from Atom (or VS Code, based on Atom), I can't name any programs that use Electron).

Additionally, XUL uses web tech, but it is not a standard and was never intended to be a standard (not even an informal specification exists AFAIK), so it's a stretch to say that Mozilla failed to execute on pushing XUL. What would the web gain by pushing it?


> I'm not convinced that Electron today is any more popular than XUL was back then

comex already named two highly popular ones (slack in particular can't be overlooked) but there are more. https://github.com/sindresorhus/awesome-electron

https://electron.atom.io/apps/

Outside of Slack, WhatsApp and Discord (see a pattern? pretty much all new chat apps are using it), I'm not sure if there's any other truly popular (among users) electron app, but among developers, electron definitely is popular and far more often used than XUL was. It's arguably more popular now for new apps than even toolkits like Qt and WxWidgets. People are writing, not one, but multiple competing implementations of things like.. unix terminal emulators in electron. These are not just webapps contained in chrome, they definitely need native access to local APIs. There's a frenzy among devs.

> so it's a stretch to say that Mozilla failed to execute on pushing XUL. What would the web gain by pushing it?

They didn't fail to "push it to the web". They failed to push it to app devs. They were rather enthusiastic at some point:

https://developer.mozilla.org/en-US/docs/Mozilla/Tech/XUL/Th...

> Whether you need to migrate an existing web application to the desktop, are looking for a technology that will enable you to easily port your applications to multiple platforms, or want to integrate your own cool features into the browser, XUL warrants serious consideration.

And now they are arguably failing to push it to themselves as they're entirely abandoning it for Firefox and will rewrite the UI and extensions APIs to get rid of XUL.


> Outside of Slack

Slack has a standalone app? I've only ever used the web version. What more does it do?

> Discord

Likewise, I never even realized Discord had a non-browser implementation.

> WhatsApp

I never even realized that WhatsApp had a non-mobile implementation.

> they definitely need native access to local APIs

Which ones?

> And now they are arguably failing to push it to themselves

Er, the whole point of Mozilla is to champion standards-based web tech. They're rewriting their UI using CSS/HTML/JS because this is feasible today, whereas it wasn't in 2005. So I ask again, why would Mozilla want to retain XUL instead of making web technologies more powerful?


> Slack has a standalone app? [...] I never even realized Discord [...] I never even realized that WhatsApp [...]

But that's exactly how Electron and friends got popular: they allow for web-first development. Desktop support is a bonus for people who need it.

When Mozilla started working on XUL, the future of the web looked a lot like XML and fat clients. Applications would be built by desktop developers targeting the desktop in a way that would allow for the web to integrate in the background. They wanted to bring the web to the desktop, woo-hoo!

As it turns out, the opposite workflow won: applications are now built by web developers targeting the web, in a way that allows for desktops to come onboard where required. In this scenario, XUL is just too complicated for people who simply need "a web view with a couple of extra APIs". This is still "bringing the web to the desktop", the fundamental role of the runtime is exactly the same; but the paradigm is opposite to what Mozilla had envisioned.

> They're rewriting their UI using CSS/HTML/JS because this is feasible today

XULRunner/Gecko was already built on CSS and JS, but it lived in a desktop paradigm of pseudo-native widgets - because HTML interfaces were terrible back then. It wasn't even a performance issue, but rather that everything was a page refresh because nobody knew any better. AJAX had not been formalized, and XMLHttpRequest was used (guess what) to pull actual XML. HTML4 was considered a hack that was supposed to eventually go away. It wasn't a "this is not possible today" problem, but rather "why would anyone want that?". Already there were lots of arguments on whether JS was "good enough" for desktop development, whether CSS was too verbose, and so on...

> why would Mozilla want to retain XUL

XUL is battle-tested to a different level, having survived almost 20 years, and it's a better option for pure-desktop development than web views. If Mozilla had paid more attention to its evolution, rather than going on quixotic adventures like FFOS, maybe they could have competed with Electron-style apps. Alas, now it's probably too late.


> Electron and friends got popular

Maybe I'm just an old fuddy-duddy (I'm in my 20s, jfc), but I don't see how Electron's popularity is considered a given. I know plenty of people who use the given services, but none who use the desktop interfaces. VS Code is the only Electron app that I've ever seen used in the wild. Am I just hanging in the wrong crowds? Which ones should I be looking at, if so?

> It wasn't a "this is not possible today" problem

Hence "feasible" rather than "possible". :P (And conspicuously not "easy" rather than "feasible", either...)

> maybe they could have competed with Electron-style apps

I sound like a broken record by now, but I still don't see what people think Mozilla has to gain by offering a native GUI framework that people have already proven they don't want. And frankly, I think FirefoxOS, while ill-fated, was at least audacious. :P


> VS Code is the only Electron app that I've ever seen used in the wild. Am I just hanging in the wrong crowds?

Just to note: Atom is the original Electron app. I assume you've seen more people use that than Code.


> As it turns out, the opposite workflow won

Maybe for you, definitely not for me. Most of the applications I use daily are desktop ones and I'll refuse to install an electron one unless I have to.


His point was about the fad amongst developers (well, "developers"...), not users.


> Slack has a standalone app? I've only ever used the web version. What more does it do?

Chew up a bunch of RAM and glitch display compositing if my experience is typical.


My slack has been running for days now, and it's humming along at 50mb.


Electron is multi-process. If looking in Windows task manager, there will be the initial process at 50 Mb and then e.g. 7 more background processes.


That puts it at 250mb. That is quite a lot for a chat app.


> Likewise, I never even realized Discord had a non-browser implementation.

Can't give an answer on slack, as it's not something I use (and I wouldn't use any of the three services if I had a choice, but you don't get to make choices for entire communities as an individual and chat apps are about communities..), but I can say something about discord.

Custom keybindings, ability to do things like push to talk while you're in game, general integration with your OS desktop features (systemtray, native OS notifications), GUI that can overlay over your full-screen window ( like this: https://youtu.be/aVQyk_GX7aE?t=98 ) and probably other features I haven't paid attention to.

> I never even realized that WhatsApp had a non-mobile implementation.

http://www.theverge.com/2016/5/10/11653606/whatsapp-mac-wind...

Released in May, 2016.. and the webapp launched in 2015.

> Which ones?

Are you seriously asking that of terminal emulators that are supposed to be drop-in replacement for iTerm or Gnome Terminal and whatever else? how do you think your bash scripts can operate without breaking on your local data in a webapp without added native calls? I feel like I'm getting trolled here.

> Er, the whole point of Mozilla is to champion standards-based web tech. They're rewriting their UI using CSS/HTML/JS because this is feasible today, whereas it wasn't in 2005. So I ask again, why would Mozilla want to retain XUL instead of making web technologies more powerful?

So a great deal of selective quoting up till here while acting sassy and not paying attention to what was quoted to support the point yet still asking the question?

Why would they retain a competitive edge when other platforms like electron are attracting all the developers to enhance their application with native access goodness, why indeed.

I'll quote it again.

> Whether you need to migrate an existing web application to the desktop [...] or want to integrate your own cool features into the browser, XUL warrants serious consideration.

Is what Electron and Chrome actually do for people now, successfully, in droves and was an aspiration they had before. As for what is feasible now that wasn't before.. electron is not going away any time soon and webapps haven't replaced native. Even more so on mobile platforms, as 90% websites are begging you to install the app with full screen popups.


> still asking the question?

Yes, because the question isn't "how is this useful", it's "how is it in Mozilla's best interest to invest in this?" Mozilla came up with XUL because they wanted a cross-platform way to develop a large graphical application, and given that they're an open source organization they figured they might as well try to popularize the framework to hopefully get some volunteer contributions to the codebase. But XUL was never an end, only a means to an end, and the technology itself no longer has any strategic advantage.

As for "failing to execute" on taking over the world of desktop apps with XUL, I don't think Mozilla will lose any sleep over it. The secret sauce to Electron isn't Chromium, it's Node.js. We can say that Mozilla failed to execute the opportunity to popularize server-side Javascript first... but that's Rhino was for, and really, nobody in 2005 would have foreseen that developers would eventually want to write their backends in Javascript. :P


> Why would they retain a competitive edge when other platforms like electron are attracting all the developers to enhance their application with native access goodness, why indeed.

Be careful with generalizations, I am surely not part of that list and know quite a few others that want their computers Electron-free.


Slack and Riot.im use Electron; Signal's desktop app uses the soon-to-be-deprecated Chrome app platform and will probably end up on Electron (AFAIK). I refuse to use any of them, having been spoiled by IRC clients that can actually handle large message volumes without flopping over.


VS Code is not based on Atom.


It's built on Electron, which was created as the heart of Atom.


In an amusingly ironic twist though, Electron apps with their Javascript frontends also have byzantine and terrible build processes.


FWIW, and this is not make a point either way, Google Mail downloads about 14 MB, two thirds of which is JS. Most of that can be cached, though. It also uses 120 MB RAM after initial load, more as I click around. Figures as reported by the Firefox dev tools.


Because Google usually does a pretty good job as far as performance is concerned.

ie: They have a simple 404 page instead of some novelty page with funny text and a random webm playing in the background


Maybe different Google departments have different opinions about such things. Chrome's error pages include a game http://www.omgchrome.com/chrome-easter-egg-trex-game-offline


But performance-wise, the error pages still load instantly and don't waste CPU when the game isn't activated, so who cares?

...though I don't know why anyone would care about slow 404 pages either, given that they should be rarely visited and quickly left.


I know this might sound insane to everyone, just hear me out.

I always thought Mozilla should have built a Linux distribution (or rather, forked debian like everyone else ha!) and implemented their Firefox UI kit (Gecko right? I'm not as familiar.) and made a Front end for it that was smooth and easy to build apps for using web technologies, but could still run native *nix apps via debion packages.

I believe that now as much as I did when I was younger, but the itme for desktop OSs may have passed.

It would have been a hit I think. Well designed, well maintained, well documented. At least in my starry eyed version of this universe where unicorns exist


They tried that for mobile devices and failed. I doubt they'd have done any better on the desktop.


in my opinion, the last iteration of FFOS looked pretty good. Granted, apps were not there, no shocker. I thought they did a great job with it design wise though, and it had some neat features. I think if it was done 5 years ago it'd been a game changer for linux on the desktop.


I think you might be describing something fairly close to ChromeOS.


Yeah, except ChromeOS is shipped with ubuntu 15.04 i think, and its heavily modified, and you don't have regular access to the greater file system, and the front end isn't really rendered out from blink/chrome, i don't believe.


What definition of heavy are you using here? Because despite often being 'thin clients', most of the programs I run that embed HTML engines for the UI end up being stupendously resource-intensive compared to their native equivalents.


I think that your parent commentator is noting that people pretty much just stopped writing desktop apps altogether, at least relative to webapps. In my childhood I downloaded desktop apps left and right for everything, but nowadays I only download apps in two categories: developer tools and Steam games. E.g. something like http://hirnsohle.de/test/fractalLab/ would have invariably been a desktop app only a few years ago.


Aside from technical issues, native mail clients are shrinking in use over time, not growing.

It does not seem like the right place to invest to continue having relevance in the marketplace.


> native mail clients are shrinking in use over time, not growing.

I see that with my colleagues, who have Gmail (Gapps for domains) in one tab and OWA in another... instead of using a mail client with unified mailbox and not having tp look after which account has the message.

Yes, Apple Mail.app user here, who is not going to switch to webapp mail anytime soon.


Mozilla doesn't have to worry about the marketplace.


I disagree. Mozilla doesn't have a table at standards bodies or any sort of lever to make positive change on the internet without a healthy marketshare for its products.


(Context: I'm a submodule owner on Thunderbird, but haven't had much time to work on it lately. It should go without saying, but the comments below are my own view on the matter and shouldn't be considered authoritative.)

tl;dr: I think Mozilla has actually treated Thunderbird fairly well (but not quite as well as I'd like).

Thunderbird has been difficult for Mozilla to support for a while. In many ways, Thunderbird is only in as good a spot as it is because of people at Mozilla who still care about Thunderbird. There are lots of Gecko features that are used by Firefox and Thunderbird, and often, Gecko devs who make breaking changes to something Thunderbird uses will - if not fix the bug - at least alert the Thunderbird team of the breakage and point them in the right direction to fix it. They're also willing to support (some) stuff that only benefits Thunderbird, especially in the text editor. This can be hard for both teams though, like when Mozilla made huge changes to the Firefox build system; since Thunderbird imports the entire Firefox source tree, Thunderbird had to keep up or be unable to build.

As you can imagine, these difficulties will only be magnified by the (slow) move away from XUL. Worse, Thunderbird's codebase is pretty crufty and full of decisions that last seemed good over a decade ago (e.g. libmime's decision to create its own C-based object system instead of using C++ like the rest of Gecko[1]). It would take a lot of work to make the codebase not suck, and I'm not sure Thunderbird has ever had enough paid staff to do that in a reasonable timeframe. These architectural issues are the reason that message tabs in Thunderbird aren't actually multiplexed: every time you open a message tab, it actually just rearranges the elements from the single instance of the 3-pane layout to make it look right. (That's why scroll position in message tabs isn't preserved when you switch tabs.)

Thunderbird has also has significant difficulties finding a way to make money. Those who remember the Mozilla Messaging days (and the subsequent merge back into MoCo) might recall how that's when the team pushed for features like "Get a New Mail Account" (which has partnered mail providers who could sell users a fancy email address), "FileLink" (which has partners providing file hosting), and even a planned effort to do like Firefox and make money from search referrals (this fell through for some fairly strange partnership issues that I'm not sure I can talk about).

However, despite their general desire to discontinue financial support for Thunderbird, Mozilla (the Foundation, I believe) has opened up donations specifically for Thunderbird[2] which, as I understand it, gave the Thunderbird Council the ability to hire the first paid engineer on Thunderbird in several years[3]. In addition, Mozilla has worked to help Thunderbird find a new long-term home and has provided infrastructure support when they really don't need to do either.

While I think Mozilla should place a greater focus on email, I'm not sure Thunderbird is the best way to go about that, especially as Mozilla moves away from XUL. With the rise of Electron/Positron, something like glodastrophe[4] (based on the ill-fated Firefox OS email app) might have a better long-term outlook. However, trying to replicate all (or even most) of Thunderbird's features in a new application would be a huge effort and it's unclear whether the app would be popular enough to make all that effort worthwhile; I expect Mozilla would want to feel confident that they could use such an app to influence email/messaging standards.

I don't know what's going to happen in the long run, but as a Mozilla employee who started out working on Thunderbird, I still have a special place in my heart for it. I hope that one day Mozilla will be willing to revisit email, but it might take a while.

[1] https://dxr.mozilla.org/comm-central/source/mailnews/mime/sr... [2] https://donate.mozilla.org/en-US/thunderbird/ [3] https://mail.mozilla.org/pipermail/tb-planning/2016-December... [4] https://github.com/asutherland/glodastrophe


We're getting closer and closer with Nylas Mail, although still have a long way to go!

https://github.com/nylas/nylas-mail'

(I work at Nylas)


Thanks for taking the time to provide your insight and especially for the link so that I can contribute directly towards continued Thunderbird development!


Gotta start contributing to TB now. Thanks for all the insight.


I cannot understand the need for a thing like Pocket. Much less why Mozilla prefers to keep it close, and even less for money.

I mean, what is wrong with bookmarks and sharing links? Firefox can even send tabs to other computers I got linked via ff sync.

Iterate on a bad idea, give it a social spin, and voilà! You have a sellable startup. Except you should not buy that.


Bookmarks-management sucks.

Pocket provides multi-platform access. And multi-browser access -- doesn't matter if I'm on Firefox, Chrome, Safari, IE, whatevs, I can pull up Pocket.

Pocket strips all of today's utterly fucked, utterly useless, utterly counterproductive Web design. I cannot emphasize too much: Web design isn't the solution, Web design is the problem.

Offline reading. Pocket articles (via app) can be read when disconnected.

Tags. You can create, and cross-reference, tags between articles.

Scale. I've got ~3k - 4k articles on Pocket. Possibly more. (I'll get to that.) They're related to ongoing research, and the ability to find, classify, and relocate large amounts of material is useful.

Whilst Pocket doesn't have a ratings or workflow-related management, I've somewhat created these. A set of tags that relates to quality (rated 0 - 5, low to high, where 0 indicates negative information, 5 indicates a foundational document -- most content is a 2 or 3, and that's rating high. Books and scientific papers generally earn a '4': definitive or clarifying, though generally not foundational.

For workflow, I'm largely working with "readme" (a reminder to go back and read closely), and a set of tags related to specific writing projects. I need to come up with a more useful system.

Pocket also ... sort of and/or sometimes ... allows for full-text search within documents. This may have been a promotional-only feature.

Given that the Web as a whole is approaching uselessness under scope of search, the ability to look through a set of documents I've already at least partially vetted is tremendously useful.

There's a whole slew of things Pocket doesn't do, which I'll address in a top-level comment. I've shared these with the firm, though I've seen little or no progress on any of them in ages.


The "I'll get to that" point: * Pocket reports absolutely no user stats. * I've got no counts of tags, articles, read articles, time read, most-read, etc., etc., etc.

This desperately needs adding to the UI and tools.


Two big things: 1) it's a lot faster to send to Pocket with the button > mark as read than managing bookmarks IMO but the real killer feature for me is 2) the offline saving of articles. Pocket is killer when I'm commuting or without Wi-Fi because it means I can keep reading things without worrying about finding a hotspot or wasting my (very limited) data.


> Pocket is killer when I'm commuting or without Wi-Fi

... and yet this thread and every other Mozilla apologist thread is always filled with folks who suggest that it's no big deal to assume an always-on, always-available network connection.


Yeah, the problem with all these web apps is the lack of functionality when offline. Most just don't work, and some have a half working offline mode, but non-perfect connectivity or no connectivity really limits productivity today.


I'm a very heavy Pocket user. Here's how I use it. When I browse the web and see something I'd like to read, I hit the Pocket button in Firefox.

When I get home, I take out my Kobo eReader, which has Pocket built-in. It's very nice to read articles on an eInk display, without the distractions of an iPad or a computer.


They try to stay independent. They try to manage their resources. So they can still exist to be Mozilla in the future.

It's HARD.

Don't blame them from not being the perfect champion we used to idealize. They are doing everything they can.


To be fair Thunderbird is awful. Every single time I try to find a better client I see someone suggesting "oh Thunderbird has gotten way better; give it a shot! I love it!" and every time I fall for it and it's the same slow, horrible application that I've tried again and again.

I'm all for them killing it.


My 1K+ folders and millions of archived emails disagree with you. I've yet to find a client that works as well as Thunderbird with my massive email archive.

Why would it be slow? I can only type so fast and sending/receiving mail is mostly dependent on the connectivity, not on the software.


I recently downloaded TB to move some offline mbox emails of large quantity. TB became increasingly slow processing few thousand. I downloaded Opera Mail to do the same exact work. I was done within few hours.

I have a need for more people to repeat the same process. There's zero chance I'll be asking them to download TB.


TB by default saves 1x file per folder (e.g. MBox like) -

This means depending on your storage configuration and how you use folders, potentially writing 100MB, 1G, 10G, etc. back to disk on folder content change.

For batch processing though, I use the CLI tools - thunderbird is view/compose only (procmail/formail/archivemail/offlineimap,etc)

I was hitting performance probs on bigger folders, and setup some more strict archiving to archive older mails to subfolders/zap offline using an 'archivemail' job, etc, and the problem went away, with TB being quite snappy.

as I understand it there was some effort to support 1-file-per-email, which might be an option now, but this has it's problems too.

Outlook has it's PST corruption hell...

Basically, in my opinion, email is a pain, each client shows it differently, though I am admittedly not up to speed on Opera Mail.


Thunderbird sets up a full-text index to allow you to search your emails. This eats up a bit of time while importing large mailboxes but is quite handy once it is done.


I tried that (I googled to see what am I missing) didn't help.


How does KMail perform for that? I never had trouble with its performance, though I don't think I ever quite made it into the millions of emails.


Thunderbird searches can be quite slow.


> My 1K+ folders and millions of archived emails disagree with you.

You seem to be attempting to address just the slowness of my complain and not its overall awfulness. Start-up time, loading each email; I've never seen it work very fast it always has that android-like lag throughout the entire thing. It also doesn't feel native to whatever OS you're using it in almost like using a Swing app.

The UX is just awful.


Well, maybe I'm just used to it. I tried many other email clients and other than text mode ones it is the best I've found so far (Linux, Debian).


Have you tried Sylpheed? It's a lightweight, open source email client for Windows and Mac OS. I am using the Windows version. The installer is a mere 8mb in file size. I switched from Thunderbird to Sylpheed and am very happy with it so far.

http://sylpheed.sraoss.jp/en/


Question, does Sylpheed have a calendar like Outlook or Thunderbird+Lightning?


The problem is that there aren't exactly a lot of other great FOSS standalone email applications these days. Everything's been neglected because of webmail, and corporations' dogged insistence on Outlook.

So if you think Thunderbird should be killed, what do you suggest as a replacement? It needs to be free, and run on Windows for my company to use it.


Not free, but Postbox is pretty great. It is a continuation of Thunderbird. By former Thunderbird engineers if I am not mistaken.

https://www.postbox-inc.com/


Might be pretty great but it's Windows/macOS only, so not sure how much of a continuation of thunderbird it can be considered.


Claws Mail is a good option.


If you're using Linux, try Geary.


The only thing that forces me to keep Thunderbird around are mailing lists. I haven't found anything else that doesn't break quoting or messes with word/line wrapping in plaintext mails. To be fair, even Thunderbird requires quite a bit of about:config tuning to get it to behave sanely.

I suppose I'll have to learn to live with mutt eventually.


IMO, mutt is the best mail client there is when it comes to lots of high-volume mailing lists. I can catch up on all my lists so much quicker than with any other mail client.


Seconded!

Mutt is the only way I can keep up with my mail volume. The only time the UI doesn't react instantly is when it loads huge mailboxes. And it still does that faster than I've seen graphical clients load the same mailbox.

Add to that pretty much infinite customization, and it does indeed suck less.

If you have to deal with large amounts of mail for pretty much whatever reason, you can do it faster with mutt than with any other client I've ever used. Yes, there is a learning curve. It isn't a vim-style learning curve, but it is there. But if you're considering a command line MUA at all, you can handle it.


Totally unrelated, but I keep falling for this same cycle of disappointment based on testimonials for The Simpsons.


I just hope you realize when people say the Simpsons are good, they mean seasons 1-{8-12}.


Every single time anyone complains about Thunderbird I ask myself wtf they are doing with this mail client.


I use Thunderbird on MacOs and on Linux. On Linux its a fine email client, no complaints at all. On Mac it's a pile of shit that randomly hangs for 10+ seconds (and eats my battery) because I have a combined inbox.

There's a bunch of config tweaks that are supposed to help but so far they haven't.


I wasn't even aware it had a Mac version but since you and me had no problems with it on X and I have no problems with it on Win, are all those complainers Apple users and should that undermine the otherwise good product? I don't think so.


My experience using thunderbird on linux is that it's a decent mail client, but I've found it corrupts its search index every few months. Rebuilding the index with my mailbox takes several days.

Also, sometimes it'll just not find certain messages. That also means it's time to clear out and rebuild indexes.

This has been my experience every time I use thunderbird, over many years. I'm not sure what it is about my mailbox that makes me see these issues when others don't.

It's a good program but could use some maintenance loving.


Thunderbird is awful, but I have yet to find anything better.


So did you find a better client?


Slow? Hmm...are you sure it isn't your server that's slow?


The speed of a desktop UI client shouldn't be related to how quick the server is.


I imagine he was talking about IMAP?


I was recently looking for desktop email clients and there is literally not a single free alternative? Cant be that awful if it's the only one.


Mailbird is pretty awesome. Been using for a year now.

If you like thunderbird theres postbox developed from thunderbird


I use opera mail client. It's quite decent. I switched from thunderbird.


Ddbdh


It would be great to have a new modern open source email app on iOS, Android and desktop OS. I want a GMail app clone (conversation view), but open source (= doesn't phone home to the creator) that works with IMAP.

Vivaldi browser (unofficial Opera successor) is working on an HTML5 powered email app, though will it be open source?

On Android there is open source K9Mail app but it's pretty old fashion, no conversation view.

I envision a new lean Mozilla browser (like Firefox was to Mozilla Suite back then) based on Rust powered Servo-HTML-engine and a HTML based UI. Additionally an email app addon would fit in my vision.


Ive thought the same thing. This is pure speculation on my part, but I wonder if they...how do you say this nicely...I wonder if they are holding on to what they were? In their approach I mean. Like they think they are still fighting with Google? So they make decisions trying to get back to that "leader position" rather than really focusing on who and what they are...which is an awesome, creative, international open-source community with great ideas...that can out maneuver any large corporation because they are smaller and faster. And cooler. Then again, maybe this rebranding is evidence of exactly that realization.


If Thunderbird is going to stop being developed, what should I use instead? I like having the add-ons, and use the Calendar with Office 365 Enterprise stuff, and Google Calendar.


I was pretty shocked when I learned they had more than 1000 employees.


Oh, what a twist!

In a story that began two years ago with Pocket's integration by Mozilla [1] in Firefox [2], large segments of the userbase spoke out with scathing criticism.

This, at first blush, appears unrelated: Mozilla previously announced its Context Graph initiative, which was a bold undertaking to be built partially upon a new and emerging set of W3C standards to take back some of the control over linkage, metadata, and the consumption and annotation of web content [3] from big incumbent providers who run content portals, content silos, or content aggregators (largely the usual suspects, including Google, Facebook [4], Apple, Microsoft, and Yahoo [5]).

To understand this play, temporarily forget about Mozilla the Foundation, and think about Mozilla as a strategic competitor to the above. In the case of Pocket, a hard-to-deny side effect is that Pocket's presence in Firefox, despite the exact nature of the integration, is likely here to stay. While this is bound to frustrate many, Mozilla's competitors routinely ship software or entire platforms with tight captive integrations, against which competition has proven difficult to mount solely on the merits of values and philosophical purity.

[1] https://hn.algolia.com/?query=firefox%20pocket [2] https://hn.algolia.com/?query=mozilla%20pocket [3] https://news.ycombinator.com/item?id=13729525#13740110 [4] https://news.ycombinator.com/item?id=13375451#13375917 [5] https://news.ycombinator.com/item?id=12863565#12867493


My only problem with Pocket was that it was proprietary and it was integrated with Firefox. If it's released under and open source license, then I'll be glad to have it integrated.

Pocket makes Mozilla competitive on mobile because Firefox's market share on mobile is very low.


The client-side portion of Pocket has always been open-source.

If you mean the server-side then fair enough, maybe this acquisition will help with that.


It looks like their Firefox addon has (https://github.com/Pocket/pocket-ff-addon), and I guess by definition you can get the source of their Chrome addon and getpocket.com, but I don't think they meet the definition of open source unless they're somewhere other than Github. Also their mobile apps seem to be closed source too.


The built-in Pocket feature is an addon that is pre-installed into Firefox:

https://dxr.mozilla.org/mozilla-central/source/browser/exten...


Yeah, but there's for example also a Pocket Android app and iOS app. Those should get open-sourced now, too, and I heavily doubt that they were before.


I agree!


Submitting links to Pocket is open source and has always been well documented, but other client-side portions of Pocket have always been closed, such as the Article View API. [1]

It would be great to have access to article view to build other clients, such as a Linux client.

1: https://getpocket.com/developer/docs/v3/article-view


Pocket will be open sourced.


Is there any official source for this?


I don't think this counts as official, but a Mozilla employee replied to a Reddit comment stating that it'll be open source too.

https://www.reddit.com/r/firefox/comments/5wio45/mozilla_acq...

[edit] also nnethercote is a Mozilla employee :)


Does that mean I can run my own pocket server at home? That'd be neat and useful, I don't need any more profiling as is.


Wow, that will include a Linux client because Pocket also runs on the Kobo eReader. I'm curious if they open source that one. It should be doable to port to desktop Linux.


"If it's released under and open source license, then I'll be glad to have it integrated."

Not me. I don't want a bloated Firefox; even if the bloat is well intentioned and fully open sourced.


One person's bloat is another person's critical feature. Are you thinking of install size, runtime cost, all of the above?

Firefox is open-source and (fairly) easy to build - if you really want the utmost control then I'd suggest this is the best way to do it. You can easily do a build that has no Pocket integration.

There's a cost to the level of configuration that I assume you're advocating here. Not saying we can't get better here, but there's definitely an impact on code complexity, QA test combinations, and stability.


I don't like bloat in web browsers because, often, if just adds features best left as an add-on. Also, in something as important as a web browser, I feel a smaller codebase is wise from a security point-of-view.


I agree in principle, although in practice allowing Firefox add-ons to be fully equivalent to other features has been a detriment to security, stability and performance - hence the switch to WebExtensions (which rely on stable APIs purposely exposed by Firefox).

Having the codebase be as small as possible is a laudable goal - with browsers it is difficult since Web standards are fairly complex on their own, and being cross-platform brings along a lot of weirdness.

An important aspect of security is compartmentalization - for instance, using separate sandboxed processes for web content vs. the main UI (which runs with full user privileges).

Sandboxing is a good example here since it improves perf/stability/security but also adds to the size and complexity of the code.


I agree in principle, although in practice allowing Firefox add-ons to be fully equivalent to other features has been a detriment to security, stability and performance

Only in an abstract "we might have been able to implement security/performance improvements faster if we wouldn't have had to worry about breaking addons" sense. Any other effects are restricted to those people actually using the addon and don't affect everyone else.

A powerful extension interface can also be used to improve performance/stability: while e.g. adblocking certainly doesn't come for free, its cost should be more than offset by not running all that crappy code pulled from ad networks.


This was actually one of the arguments for feature shredding things like browser customization (see Classic Theme Restorer) and tab groups (see uhm Tab Groups).


> One person's bloat is another person's critical feature.

That's what extensions are for. Those who want it, can have it, those who don't, they won't. Everybody wins, no need to raise emotions.


I'm so confused with the HN comments that bemoan the death of XUL, alongside comments that complain about bloat in the browser. It's interesting to see how different people use different parts of Firefox differently.


I know! I often see comments that imply HN is homogenous (with the exception of the poster). There's a lot of different people that make up the HN community, which is one of its strengths, and one of the reasons I value it.


"large segments of the userbase spoke out with scathing criticism"

Actually .. no. It was a very vocal minority.


You're right, I misspoke; I meant to say several audiences out of a larger pie of distinct cohorts; instead of using wording that implied a large absolute amount of people.

In the past, I've written about [1][2] my views of Mozilla's audiences, and soon after, Mozilla's own internal audiences [3] were brought to my attention. Several of their 'user types' are either noted as averse to change or loss of control.

[1] https://news.ycombinator.com/item?id=13425956 [2] https://news.ycombinator.com/item?id=12192509#12194161 [3] https://blog.mozilla.org/ux/2013/08/firefox-user-types-in-no...


Something I've always wished Pocket would do is download an offline copy of videos I add to the list (using, say, youtube-dl [1]), so I can watch them on the subway, etc. But they never did this, presumably either for legal reasons, or because it was a giant waste of bandwidth considering a huge percentage of Pocketed articles never get read.

Now that Mozilla is promising to open-source this, I eagerly await adding this feature to my own fork :)

-- [1]: https://rg3.github.io/youtube-dl/


If you're on Android, I've found a set of tricks which work for this pretty well.

Ingredients:

1. Termux App.

2. Termux API App.

3. Termux CLI utilities. apt install termux-api

4. Python: apt install python

5. youtube-dl: pip install youtube-dl

6. VLC App. Install from app store.

7. A set of shell functions for clipboard and application access:

    alias xc='termux-clipboard-set'
    alias xp='termux-clipboard-get'
    open () 
    { 
        termux-share "$@"
    }
I'll generally download videos to the Download folder, which I've symlinked from the Termux home directory:

   lrwxrwxrwx 1 u0_a204 u0_a204 29 Apr 26  2016 Download -> /storage/emulated/0/Download/
Oh, and how did I get those functions and aliases?

    (alias | egrep x[cp]; typeset -f open) | sed s/^/ / | xc
(Pre-formatted for HN's code block.)

With all of that, you can download and play, in one swell foop, a video (assuming you know the output format you prefer -- something you can set as a youtube-dl preference):

    open $( youtube-dl -f 18  "$(xp)" | awk -F':' '/Destination/ {print $2}' )
Or you can fetcha whole slew of content and queue it up on VLC.

I'd still prefer to be able to use a commandline audio player with controls rather than the GUI client. There's a very simple "play-audio" command, but it has no controls -- <ctrl>-C to end is about it, even backgrounding doesn't actually stop playback.

I've messed with the sox "play" command, but it doesn't have an assigned audio device, and I've been unable to convince it that such a thing exists (see StackExchange posts regarding Android for similar issues/discussions).


Seconding this for macOS. If you haven't installed Homebrew, do so: https://brew.sh/

Then install youtube-dl as follows:

  $ brew install youtube-dl
And download videos as follows:

  $ youtube-dl "https://www.youtube.com/watch?v=dQw4w9WgXcQ"


The lack of a true, independent, freestanding multimedia application which can handle this sort of thing -- building and managing queues, interrupting, deleting, sorting, etc. -- puzzles me.

The browser is such an obviously poor place to stick A/V. Other than, of course, the fact that advertisers slobber over cramming every more crap down some "channel" or other.


Well, Google has a history of preferring HTML above all other interfaces. Only when someone else comes along and offers a better native interface, they jump in.

Thank god there's a native iOS/Android client.


It's not the transport that's the problem, it's the app.

Meredith L. Patterson has addressed the point of Port 80 (or 443) adequately. For the time being, we're locked into that.

My point is that handing of A/V material to a dedicated AV app is, all else considered, probably optimal. It's the 4th browser replacement mentioned in my "Tabbed Browsing" rant, elsewhere in this thread.


I've played a bit with terminals on Android, but I only had real success using ssh. This seems very convenient, I'll play around with termux, thanks!

BTW, I'm unable to access /storage/emulated/0, it seems Termux doesnt have permission (and the app doesnt list it in its permissions). Is your phone rooted?


I've attempted rooting the phone, but don't believe I've succeeded. One of my major gripes about the whole Android ecosystem (and a massively significant one). Samsung Tab A.

I've just confirmed though that I can create and delete files on /storage/emulated/0/ YMMV, is all I can say.


It also kind of sucks with javascript when it downloads the page offline. Pages that use Mathjax or something for LaTeX never display correctly so I can enver read them offline :/


This problem got solved when YouTube started allowing offline videos. Idk if it is available in your country, but in India you don't need to be a paying customer to save offline videos.


India usually get those offline features much more faster than other countries for some reason.

I wish we had these options in Canada, because mobile bandwidth is stupidly expensive over here.


> India usually get those offline features

This is because Internet in India sucks


... plus with a population of ~1.25 billion people, it's an incredibly large market to overlook.


That's only on mobile afaik. How do you deal with it on the desktop?


On macOS, do something like this:

  $ brew install youtube-dl


On macOS, the following works as well:

    $ pip install youtube-dl
even if you don't use brew. Python is there out of the box.


There are many apps that do that. A Pocket competitor, Spool, used to that. It was then acqui-hired by Facebook and shutdown. At least they had the decency of sending us our data by email.


You could script this with a bookmarking service like pinboard.in (anything with an API) and a cron job.


See Youtube red.


Youtube Red is not available in most of the world:

> available in the United States, Australia, Mexico, New Zealand and South Korea.


I guess they'll have to wait.


My conscience is assuaged by the fact that I am already paying for Google Play Music, which would include YouTube Red at no additional cost, if I lived a few miles away on the other side of the Canada-USA border.

And anyway, despite the name, youtube-dl supports a whole lot more than just YouTube.


That's a paid service. Youtube-DL is free and open source.


It is YouTube's prerogative to offer that feature for their paid service.


And it's the user's prerogative to build and use tools to circumvent artificial[1] limitations.

[1]: reminder: technically, streaming isn't different from download, what your computer does when watching a video in streaming is downloading. Forcing users to delete the video they have downloaded (which is exactly what's happening with streaming) so they pay the premium service for off-line use, is not a legitimate way to do business. They can do it of course, if people are willing to pay for something they can have for free it's fine, but people who don't want to pay aren't doing anything wrong.


> is not a legitimate way to do business.

You're not the arbiter of this.

> but people who don't want to pay aren't doing anything wrong.

Not wanting to pay isn't wrong. Circumventing the Youtube's restrictions is though.


> Circumventing the Youtube's restrictions is though.

That's exactly my point: it's not.

It may be bad for Youtube's business, but it's not morally wrong. We owe nothing to Google.


https://www.youtube.com/static?template=terms

Look at 5b. Downloading Youtube videos, except with Youtube Red, is a breach of contract.

Breaking contract is wrong.


> Downloading Youtube videos, except with Youtube Red, is a breach of contract.

Too bad, then every single user of Youtube is violating this rule: the browser needs to download the video to play it, streaming is just the act of automatically removing the video chunk after it's played. I don't seen written anywhere that the user is forbidden to prevent the deletion of the video chunk he downloaded.

Even if Youtube updated its rules to fill this loophole, I would still be allowed to do it. In France people are protected by «l'exception de copie privée» which allow them to have their own copy of video or music which their are legally allowed to see/listen.[1] That's why we are allowed to record TV broadcast. I guess there must be a similar law in the US but I'm not really aware of the regulation there.

[1] sharing the video with somebody outside of your family is forbidden though.


What restrictions?


It sure is their prerogative. It's unfortunate their paid service is not available to most of the world.


So is the user's prerogative to use their downloaded content in any way they want, in this case not consuming immediately but saving for later.


> So is the user's prerogative to use their downloaded content in any way they want, in this case not consuming immediately but saving for later.

It's not their (the user's) content.


Pocket, for me, is a weird product/feature. I've tried to use it, but it ends up just being like a clumsy bookmarking feature.

Allegedly, you can read things offline, but that feature never works for me. It seems the feature for offline reading doesn't exist at all in Firefox (which I'd assumed was the point of its integration, but my assumption was wrong), and I've tried the Chrome desktop app and the Android mobile app ..nothing I save to pocket is ever readable without a data connection. It must work for some content for some people because I see people talking about it like it does. But, without that feature I see literally no utility in Pocket...it's just a clunky bolted-on bookmark manager, and so I end up going back to bookmarks with tags. I use Sync and I have Firefox on all of my devices. So, my bookmarks go with me.

So, I guess it's good that it's going open source. I wasn't comfortable with the way integration was presented back when they added it...it wasn't at all apparent, to me, that Pocket was a third party for-profit entity when the "setup your pocket" process launched in Firefox, since I'd never heard of it before that. But, it still seems like a solution looking for a problem. I'll give it another look. Maybe I was just unlucky with my choices of what to save...but it seems like if offline reading is not going to work on a page (for whatever reason) it should warn you. It'd suck to get on a train/plane or get lost in the woods, with plans to catch up on some reading, only to find there's nothing there.

But, maybe that's not even the primary purpose of Pocket? I dunno, it's still pretty fuzzy to me wtf it's for, if not that.


> Allegedly, you can read things offline, but that feature never works for me.

Strange, because that is the main use case I use in Pocket. I save article there, and it stores it for offline reading, striping all formatting. It isn't a bookmark manager, it stores stuff and doesn't need internet to read it.


My experience has been GP's: Pocket will claim to save something, but the moment I go offline, it craps out and tells me I need a connection to read the content I've saved.

This is much more common on larger PDFs, but I've had it happen on so many sites that there was a point when I literally resorted to PDF'ing the pages I wanted to save - which is a damn shame, because Pocket's reader view really does do a good job.


Pocket doesn't store videos or pdf's offline. Only HTML articles with the formatting stripped.


I used to use Poki on windows phone and never experienced any issues. But since moving to iOS I often add stuff to pocket only to realise it hasn't auto synced or that I forgot to download before getting on the plane. The experience isn't that great IMO :(

If I remember to open pocket on my phone before flying I don't have issues.


The Kobo ereaders have had Pocket integration for a while now, and it works great offline. My usual routine is to find interesting articles on Hacker News, and if they are longer than a page or two, click the Pocket button and read them at home on the ereader. A few don't render well (the article on Arrival/Story of your Life not being a time-travel story was a recent fail), but generally it works great.


I love this feature. My only concern was being tied into Pocket and wondering what might happen to the company. I imagine that being acquired by Mozilla makes that future a bit more secure.


the article on Arrival/Story of your Life not being a time-travel story was a recent fail

Loved the book / movie but haven't read that article. Do you have the link? Thanks.



It isn't all that obvious how the offline function works - which is a shame. I've been caught without offline availability for some articles I saved.

The main thing with the work-around is that you need to start Pocket on your phone while you have a data connection so that the articles are on the phone. There may be an option to automatically download the articles, but I've never found it. After the one-time data connection, it comes up offline.

This is the case when I save from the computer and view them on my phone (or different device, I guess, but I've not checked a different computer). Your luck may vary.


Turn on "Instant Sync" and new additions will be added to your app asap.

Also check your download settings. If set to "download only on wifi" then the app won't download if you're on mobile networks. You can also set the "refresh when opened" option to sync everything when the app is first loaded.


> The main thing with the work-around is that you need to start Pocket on your phone while you have a data connection so that the articles are on the phone.

Hmm, I always found this to be obvious, and not a work-around. How else would they entered the phone?


I would not expect to have to open the app. I can set my podcast manager to download episodes automatically -- I don't need to remember to open the app while I'm on data.


I've always liked Pocket, but felt queasy about its integration with Firefox.

I consider this a great move for a better bookmarking experience in Firefox as well as a better pocket service.


Couldn't agree with you more, I absolutely love both of the brands and use their products regularly, but the Pocket integration into Firefox itself is something I always considered a bad move from the principles perspective. At least including it now makes much more sense.


It is ridiculous though. Mozilla already had a sync infrastructure, it didn't need another service just for bookmarks, "only" a decent additional UI.

To me, the whole Pocket thing always smelled of cronyism. Someone is friend with someone else, and lo, Pocket appears. People get all angry, Pocket's hockey-stick fails to materialise, and lo, Mozilla bail them out...


Not only that, but the original Read-It-Later extension did exactly that: it stored links directly in Firefox's bookmarks system in a "Read-It-Later" folder. Only later did it begin using its own, separate database.


Or in a less negative spin, acqui-hire a group of people you already know you enjoy working together with.

I don't think that there would have been that many other suitors for pocket, so depending on how close they were to just closing the price might be extremely low (or are any numbers making the rounds?). Maybe there is not much acqui in the hire, could have been more like a deal between founders and investors, "allow me to take the brand with me to my future employer".


This is far more powerful than just bookmarks. Pocket succeeded in getting clients everywhere, even on eReaders.


I agree. For example, you can't really see a list of what you pocketed unless you go to their webpage. Or if you want to be able to read them offline, you need to download the Pocket application. I wish all was integrated in Firefox. It's 2017 and a browser needs this and tabs on the side by default.


You can have tabs on the side via an addon. Here: https://testpilot.firefox.com/experiments/tab-center


It kinda sucks compared to tree style tabs though :/


The old discontinued extension did exactly this. It had oflline, integrated into sidebar and displayed if you already have added link to pocket.


"We believe that the discovery and accessibility of high quality web content is key to keeping the internet healthy by fighting against the rising tide of centralization and walled gardens" that alone is a reason to welcome this aquisition. Facebook has become for many, me included the entry point to the web much more so than the browser and I really don't like that. I can imagine Mozilla suggest content to users based in what they have saved in the past on pocket and integration with Firefox. Nice move.


I suppose then that the strategy will be "the browser as a platform" vs "FB as a platform", with the technical difference that all FF tools will be open source?

That actually sounds promising. Next logical content vertical would be video then (judging from FB's content importance)


Pocket's post about the acquisition, referring to the main value their product brings, says, “a platform where high-quality, thoughtful content and free speech can rise above the rest.”

Pocket used to have a feature aimed at surfacing high-quality content from the collection of things a user had already pocketed. They removed that feature. I was paying for their premium service before that change. After they removed it, I stopped the premium subscription because I wasn't sure what pocket's value to me was anymore.

Now they have a recommender that recommends things you haven't pocketed yet. But that just encourages the user to accumulate an ever larger collection of pocketed things, not surface the best things in that collection.

Ever since, I've reflexively kept saving things to pocket, hoping they would bring that feature back. But the only practical thing I have done with it is pull up something I just saved recently, because it will be near the top of the list.

Pocket's CEO also writes, “In fact, we have a few major updates up our sleeves that we are really excited to get into your hands in the coming months.” I hope they will bring that feature, or something with a similar aim, back.


I also was/am a frequent Pocket user. But the removal of features (also pointed out by others here) and on the other hand the addition of the recommendation/feed thingy left me wondering where Pocket was headed.

Pocket for me was my archive of things for which I would love to have a great diversity of searching, clustering and organising tools. The addition of recommendation of things I may like is nice, but not a core feature for me - my content is.

I also tried Wallabag [0] which looks promising but hasn't the ecosystem around it yet. So I'm exited to see where this open sourcing of Pocket will lead us. ;-)

[0]: https://wallabag.org/en


I have similar concerns. I really used Pocket a lot. Two times I got an e-mail at the end of the year that I'm in the top 1% of their users. That was nice!

I want distraction free reading which means no social recommendations, sharing and other stuff. Things I would really like is pdf support, x days of reading left, auto scrolling and detailed statistics about my reading. Also cool would be some garbage collection, e.g., news articles are useless to read in 2 weeks but a blog post may be still useful in a few years.


OK, so it's not just me imagining things.

My view is pretty much as yours: POCKET IS MY VETTED WEB. If I've saved stuff there, it's because that is shit I've already determined is worth a second look, and I want to be able to find it again.

Increasingly, when I hit my Pocket archive (3k-4k+ items, top 1% user), I find that I cannot find what I'm looking for. This negates approximately 80% of the purpose of having Pocket in the first place, and I've been giving strong thoughts to simply using it as a link dump until I can figure out how to:

1. Download the content myself.

2. Download the tags, which are the most important element.

3. Get the simplified-Web view. Web design isn't the solution, Web design is the problem.

4. Be able to search and filter specifically what I've already fetched. And filter by more options: date, website, author, tags (positive and negative filters), etc.

None of that currently exists. And it is increasingly frustrating.


+1 for searching and filtering. I use Bookmark OS for this. It's more desktop oriented but the sorting and visual nature really helps me browse what I've already bookmarked. https://bookmarkos.com


Interesting.

I've just started speccing out some thoughts on how document references (files, URLs, URIs, ...?) might be improved. A sort of api / multiple-attribute interface strikes me as useful:

1. Meaningful titles / filenames.

2. A content hash. Not exactly human-friendly... I'd prefer something shorter to longer, though there's the collisions problem. Still thinking on this. Could tie into VCS (git, Hg, etc.)

3. Various standard metadata: author(s), editor(s), publisher(s), and their relation(s) to the doc. See various bibliographic or MARC 21 formats.

4. Dates: Initiated, published, modified, accessed, read date(s).

5. Classification schema. Here with time I find that working with extant rather than de novo or ad hoc schema is most likely preferable. Dewey Decimal is apparently more logical than LoCCS, but the Library of Congress classification is nonproprietary. That may well be what makes it win out.

6. Workflow indicators. Unread, read, deep-archive, to-process, etc.

7. Reputation and ratings. Associated with works, authors, other contributors (editors, fact-checkers, etc.), publishers. Some form of distributed assessment here would be useful.

The question is how to make these visible. My thought is to treat the parameters as, essentially, search. It doesn't matter what name you use, so long as the set characteristics is unique or determinable. E.g., returning 3 items would allow me to manually determine which I'm interested in, returning 300 would make that difficult, returning 3,000 would probably require some programmatic determination (or further specification).

A URL scheme, say, URI:/au=clarke&dat=lt:1980&ti:=imperial&ti=earth ... could, say, return Arthur C. Clarke's Imperial Earth.

A filesystem / virtual filesystem approach might work somewhat similarly locally, either from a CLI or graphical approach. I'm thinking of how that might work....

Very early stages.


Not only they removed that feature, but they also removed a bunch of other automatic-categories that I used to love (quick reads!)

And, as I've described in a blog post I did about Pocket, the recommendation system they created fails in one huge aspect: it is recommending me things that I've seen elsewhere but just decided not to pocket...


This is weird. Has Mozilla acquired any startups in the past? Feels weird to me for a browser non-profit to own a web app product.


There was a resistance to using Pocket, because it's a closed proprietary service. Now Mozilla plan to open source it, for which they had to buy it first. So it's a good outcome IMHO.


Now all they need to do is make it self-hostable.


I'm with you!

I know there's wallabag, but i always felt it was too hefty for self-hosting on, say, a small digital ocean vps. I've been using Shaarli [https://github.com/shaarli/Shaarli] because its so lean, and easy/quick to set up...though admittedly it is also lean on features (which i'm cool with, but others might want more). My hope is that pocket - once made to be self-hostable - is easy and lean to self-host on a small vps.

...Though even if its not as lean and easy to set up, the fact that pocket will now/soon be added to the open source world is always a win!


I ran wallabag with nginx, fastcgi and sqlite on servers shittier than cheapest DO instances and haven't had any problems. Sure it may be slightly heavier than scp'ing and running a single go binary but compared to the likes of rails or java apps this setup as light as it gets.


wallabag is an alternative that's self-hostable


It is weird - the press release itself notes that this is the first such acquisition Mozilla has ever made.

In their defense, it won't be the first time they've gotten into operating a service -- there's Sync, for instance. So if they see expanding FF's bookmarking abilities as a key strategic differentiator, buying an existing read-it-later service they're already familiar with could be a simpler/cheaper way of getting there than building one from scratch.


First acquisition but not first investment. For example see https://blog.mozilla.org/press-uk/2016/08/23/mozilla-makes-s...


Mozilla also joined Everything.me's Series C funding round:

https://blog.mozilla.org/blog/2012/11/28/mozilla-invests-in-...


Too bad sync kind of sucks... Just yesterday I was looking at a recipe on my desktop and then I opened Firefox on my phone and tried to get at the link without emailing or messaging myself and it only showed sync'd tabs from over a week ago.


There is an option in `about:config` for that -- `services.sync.sendTabToDevice.enabled`, disabled by default. Suggest you to enable it and try it out.


This is Mozilla corporation, which l, while being a subsidy of the foundation, is a corporation regardless.


Right, I thought pocket was actually one of the Mozilla sponsors, given the default installation in Firefox.


It was stated several times that Mozilla didn't get any money from Pocket for the integration.



Wow, if that article is true then not only were Mozilla [indirectly] taking money for the integrations but they also lied to us all about it.

Is there corroboration that Mozilla had/has revenue sharing with Pocket and Telefonica? I've always trusted Mozilla, this suggests that trust has been badly misplaced.

I can't help but feel that I'm being suckered somewhere here - who from Mozilla is benefiting personally out of this acquisition and why are they being allowed to steer the ship?

It's 14 months since that blog post, I realise it's not news to some, but it's really disappointing to me.


Mozilla releases their financial reports about two years after any given financial year, so we should soon know for sure and I also just really can't imagine them having lied as a result of that.


IMO the simple reality is they need to diversify their income sources to avoid putting all their eggs in one basket, and they decided Pocket makes sense. I'm still a huge Mozilla fan, the whole Pocket episode is not a big deal to me.


Where did Mozilla say that they didn't have revenue sharing with those?


In that article the Mozilla spokesman says that they had no financial interest in the integration of Pocket. If they get revenue from customers introduced through the integration that seems like a barefaced lie.

>Chad Weiner, Mozilla's director of project management told PC World in an email back then that "there [was] no monetary benefit to Mozilla from the integration" and that Pocket "didn't pay for placement in the browser". (ibid) //

vs.

>Although the company emphasizes that Pocket and Telefonica didn’t pay for placement in the Firefox browser, Mozilla Corp. chief legal and business officer Denelle Dixon-Thayer told WIRED that Mozilla has revenue sharing arrangements with both companies. (Wired, as quoted ibid) //

They seem directly conflicting statements to me barring a monumental coincidence. Even if it were considered not contradictory, per se, then it's highly misleading. If I say "I wasn't paid to write this car review" and it turns out I get a cut of sales of the cars ... pretty fraud-y.


At the time, this was the most baffling aspect to me; it would've been more understandable if Pocket were paying Mozilla for the preinstall, but Mozilla opting to corrupt their own ideals of an open web without any monetary gain was mind-boggling.

Now it makes at least some sense.


They invested a million or so in Everything.me a few years ago for Firefox OS. I don't think they've actually acquired another company, though, or at least not any company of note.


I have been working on a prototype for a competitor to Pocket and other save-for-later services with an emphasis on minimalism, privacy and ease of use.

https://slushi.es

Very early app, features are rolling out daily. My end goal is to build a recommendation engine out of user data, while keeping things anonymous.

Although social bookmarking has been done plenty times before, I think execution has been sub-par in previous solutions. My aim is to make things streamlined and fast for the user to go from 'what was that link?' to 'there it is' (sidenote: check out slushi.es opensearch, and how it behaves when there's one result (only works for signed-in users)).

A complaint I hear often from Pocket users (and other save-for-later users alike) is that they will save 100s of links without ever going back to reading them. I, too, shared the same sentiment until one day I experienced a 'what was that link again?' moment. It was rare and fleeting at the time, but since using slushi.es more and more, those moments have appeared with increasing frequency.

I believe the best save-for-later app will transform the regular web browsing individual's habits of reading an article and forgetting about it, to reading an article and remembering it later; either a day later, a month later, or perhaps years. The thing is remembering. I think a good save-for-later app works as a memory reinforcer; something that augments your ability to build and recall knowledge.


It doesn't sound like you are a competitor for Pocket. It sounds like you are a competitor to Diigo. You might want to re-evaluate who your competitors are and do a SWOT analysis of them.


thanks for the tip. diigo is definitely a competitor (was noticed by our analysis a long time ago but attention to it has fell by the wayside). pocket seems like the biggest whale, so naturally have pinned it as one of our competitors. i think any save-for-later app that introduces discoverability is a competitor


I'd be curious to see an organization scheme other than just tags, as I suspect that it leads to difficult to trim (potentially always climbing up) bloat and cognitive load.


interesting point. tags have been a feature we have not added yet partly because they seem slightly superfluous. the user is able to search across titles, descriptions, URLs, and can curate links with lists. not sure if tags can provide any extra value


If your goal is to create a service that's main feature is content discoverability, I would think that tags are essential. It would provide you with great metrics for knowing what the user prefers. For instance, "They tend to tag lots of funny stories on Saturday. The tag tech news on Monday and Tuesday. They tag movie news on Thursday." Your recommendation engine would then be able to suggest articles to them. "Other users have tagged these stories as tech news and it is Monday. Here's your Monday feed!" I'm not saying you can't make a working product without them. I'm simply saying you probably can't make a competitive product without them.


Lists and tags are the same thing, where with "tags" URLs can be on multiple lists.

Tags are essential. Think about this: a user saves this link to a list of stories about Pocket. He also collects links about Mozilla's management, so he puts it on a list of stories about Mozilla. Then he remembers that he wants to talk about this link with a friend, so he adds it to a list of links he wants to share with his friend.

According to "lists," he has to choose between one of those three lists, while with tags he can use all three. Search does not replace tags, and neither does having to manually type keywords into the description field.

Leaving out tags is a fundamental oversight. I would never consider using any service that didn't have them.


sorry alphapapa, it may not be obvious on the site as it is currently, but you are able to save 1 link to multiple lists


Okay, so how is that different than tags? A rose by any other name...?


Read it later apps are dead. Why do you think Pocket just got absorbed by Mozilla ?


What would you say killed them?

Better optimisation for mobile viewing has given me less use for Pocket, but I still use it a lot, so I'm interested in your opinion here.


I signed up and tried to use it but I don't even understand how to add links. The search bar hangs itself (chrome 56) and there seems no way to manually add a link? Am i missing something?


bosie, so far there aren't any guides/helpful messaging or placeholders, so my apologies it's not clear.

the functionality as it stands is, pasting a link and hitting ENTER will add the link.

it needs to be pre-pended with http in order for it to add a link, otherwise it will perform a search.

this will be changed to add a link (http prefixed or not) in any case when there are no search results


I have used Pocket for the last five years or so. 99.5% of the articles I bookmark every day I never actually read. Probably should make me think why I use it at all...


I read A LOT on pocket (they sent me an email that I was one of their top 1% readers, with having read 1.7 million words).

For me it's a great time saver, I see some interesting article during work, but instead of wasting my time on it during work, I save it for later. Granted I get no signal during the subway where Pocket becomes a really useful app, but it results in me being a lot more efficient with reading.`


Same here.


I see Pocket as the place where I put incredibly interesting things that I will never see again. I absolutely love Pocket and I have So Many interesting things in there. But I never look.

I started putting things in Google Inbox, and I found that I like the reminders to read things - but I don't look at those either. I just keep pushing them back, and eventually, during a less busy week, move them over to Pocket, where I can truly stop reading them.

One day, years from now, I'll read everything I've put into pocket over the years and finally understand the universe.


You are not alone, but I just use bookmarks for that. Is using Pocket any different from that, e.g. does it save pages internally so if the original site goes down the page will still be available in Pocket?


Yes, Pocket saves a snapshot of the original, as well as a slimmed down version that's ideal for reading (akin to a "readability" version - maybe exactly that).

The mobile app also syncs the slimmed down versions so you can read them while offline. If you're using it for that, be sure to open the pocket app once in a while as it will stop syncing if you haven't opened the app for some [undetermined] amount of time.


I am in the same spot. I bookmark a TON of stuff and I never visit it again.

I do sync pocket right before a flight but find myself engaging with it rarely.

I find my use of bookmarks akin to liking something on facebook (or even saving it for later, which I never go back to).


All the latest e-readers from Kobo come with Pocket integration (hopefully also going forward) and I can testify that it helps a lot if you get the reading material off the LCD display and off the general internet where I would otherwise also just browse around and bookmark stuff without actually reading anything.


Neither do I, but it's invaluable when I remember something in that 0.5% and need to search for it.

I use Pinboard for organized reference bookmarks, and Pocket as a dumping ground for anything that seems interesting. Some of it gets moved to Pinboard or deleted, but otherwise I like keeping it there as a record in case my brain decides something from the past is important again.


First: I'd love to have stats on material indicated within Pocket.

1. Total articles saved. Archived. Rread.

2. Counts by tag.

3. A "times read" count. There are hot articles I hit a lot.

4. Numbering of items on page.

The lack of ANY user-focused stats or reporting on virtually any Web-based tools is ... kind of creepy, actually, given the extent user data is sliced, diced, marinated, and skewered on the provider and advertiser ends.

Back to the matter of what's read: Pocket is effectively my "vetted web". Content I've found and found at least worth a second look. The biggest problem(s) I have presently are that I cannot effectively mine that archive. Pocket's tools for searching, ordering (or unordering -- there's a strong argument for randomly accessing a stratified sample) are a hugely increasing friction point.


That's me as well. I've used Pocket since it was called Read It Later and probably have opened my saved list only a handful of times.

What I've started using some time ago, though, is Reread.io – it sends you emails containing random articles from your saved list, on specified intervals. I have it set up to send me a couple each week, that way it doesn't feel "forced", and better than having them forever forgotten.


My Pocket usage increased significantly if I'm out of the house often. I use very little mobile internet to keep my phone costs down - using pocket offline works out great.

I don't read all of the articles I have on there, of course, but that's OK. It still works for bathroom or on-the-road reading. Admittedly, games get played during toilet visits more often than articles getting read.


This is why I created a IFTTT action to add the item to todoist if I have given it a tag ('read', in my case).

This way I can go back and read the articles that I really care about, while still saving low priority stuff that I only go through when I'm bored.


I did the same thing. Then I bought a kindle and use calibre to download my reading. That's made all the difference. Much more comfortable to read and I can catch up on an article or two before bed or when I've got a few minutes.


I Pocket a lot of stuff that I don't read, but in the end, I end up reading more stuff thanks to Pocket anyway.


The Firefox bug to open source Pocket: https://bugzil.la/open-pocket


As a former del.ici.ous engineer and user, I'm happy to see that Pocket will be in good hands for a while :)


Is there any chance you know what's happening with delicious now? Their last blog post is dated "April 24, 2016" [1], and they are silent on Facebook since May 2016 too [2] despite loads of comments.

[1] http://blog.del.icio.us/?p=1193

[2] https://www.facebook.com/delicious/posts/10153614657337205


a year or two they went down for about a week. That was the final straw for me and pushed me to pay for pinboard.in. I haven't looked back since. it just works and I don't worry about it going down or going away.


Pocket is one of my favorite products to use. Congratulations to the team!


So much for my hopes of them removing it from Firefox. Uhg.


In case you didn't know, you can remove it from the toolbar via "Customize", or completely disable it by toggling `extensions.pocket.enabled` in `about:config`.

That being said, I started using it extensively after being a big fan of Instapaper. The browser integration (especially the easy tagging) has been really helpful in keeping my tab count down.


I use a simple bookmarklet for Pocket. This consumes zero resources and works just fine.

https://pastebin.mozilla.org/8980504


That's how Instapaper did it back when I used it (may have a full-fledged add-on now, but I don't know for sure).

The only drawback, for me, was the added bulk of the Bookmarks bar. I like to slim it down, so having the single button has been great. (Also, being able to add the tags, but maybe the bookmarklet is capable of that?)


I remove the bookmark's name so only the Pocket icon shows. It's very slim, but you do need the bookmark bar for that. I suppose it would work from the bookmarks menu too.

When you click the bookmarklet a little dropdown banner shows and you can add tags from there.


You might like the "Custom Buttons Bookmarklet Maker" I use it to have a button for adding videos from a page to my Plex queue, removes the need to have the bookmark bar on screen. I found it rather arcane to set up however.

https://addons.mozilla.org/en-US/firefox/addon/custom-button... ... ah, and now I see it's unmaintained and recent version isn't working for some, carry on, never mind (works for me though).


When people say that they want to "remove Pocket from Firefox", they don't mean just hiding a toolbar button or disabling it using about:config.

They don't want it there at all, in any way. They don't want it using any of their computing resources, no matter how minute this usage might be.

They tend to see it as something that should be available as an add-on that's installed on-demand by those who actually want the functionality.


I disagree. People wanted to get rid of it, because they viewed it as an untrustworthy component, which it now is not anymore. The few kB that it takes up are really not a problem that anyone could rationally complain about.


You can have hope for a better integration than that odd toolbar button and the silly login page that asks you to make Yet Another Password instead of using the Sync/Mozilla account one.


Aren't you able to log into Pocket using your Mozilla account for a while now?


No, it still requires you to create additional passwords for Pocket.


What is the resistance to having it in at this point? Every browser has a cloud-backed page snapshot feature.


My own reason for disliking it was that it's a closed-source piece of an otherwise-FOSS browser. Hopefully this acquisition means a change for the better on that front.


I always thought that complaint was kind of odd. It's essentially the same as their Search input. Google/Yahoo/etc. are proprietary but Firefox lets you talk to them directly from their browser. Similarly, the code for the Pocket button (component) for where it handles the user clicking the button, along with what it does to talk to the Pocket server, was all in the open and if you build from source you could just remove that entirely. I was more against it for the same reasons I'm against any UI 'upgrades' when I upgrade the browser. The "share this page" button that suddenly appeared annoys me equally as much. Fortunately I can just remove it from the menu but I'd rather Mozilla didn't mess with my UI customization. (I don't know what I'm going to do when I can't use treestyle tabs anymore. I guess move to Pale Moon.)


Mozilla replaced Reading List (which was client-side encrypted and theoretically self-hostable) with Pocket (which is neither). Search engines need search terms to function; Pocket uses user data to support a particular business model.


Reading list was seriously underfeatured.

Pocket is now Free Software, which might make those elements from RL you liked possible again.


Reading List was still in development when it was canceled. Anyone who preferred Pocket was free to install it.

Continuing the current revenue model and supporting Context Graph seem at odds with improving user privacy. Pocket isn't now Free Software, but hopefully it will be feasible to self-host whenever they do release the source.


"I don't know what I'm going to do when I can't use treestyle tabs anymore. I guess move to Pale Moon."

I hear ya, being in a similar boat. My hope is that Netsurf ends up being a decent contender here by the time Mozilla kills XUL. Else, hopefully TST will indeed be made possible with Firefox-specific WebExtension APIs.


TST will be possible with WebExtensions. For hard evidence, even the Add-Ons Engineering Manager is working on his own, personal vertical tabs webextension to suss out what APIs we need: https://github.com/andymckay/sidebar-tabs

We may not land everything we need by the time Firefox 57 releases, but we'll get there. If we miss, you have options: you can decamp to Firefox ESR for a few months, where TST will keep working, or someone could theoretically convert TST to a WebExtension Experiment which would work on DevEdition and Nightly builds of Firefox.


That's very good news. TST is one of the few things that makes web browsing even vaguely viable these days.

I've noticed (on the rare instances I'm on desktop, mostly Mac these days) that Vimperator is also borked (only parts of it seem to be functioning). That's another lifesaver to me, and whatever it takes for it to come back to life and/or an alternative to provide effectively similar functionality ... would be peachy.

If I can lure you to another bit of developer / user advocacy, a bit from about 3 years ago on the content management problem:

https://www.reddit.com/r/dredmorbius/comments/256lxu/tabbed_...


The search plugin feature is an open standard. Anyone can create their own search engine and plugin for it.


Pocket will become FOSS, so yes. Follow [0] for status updates.

[0] https://bugzilla.mozilla.org/show_bug.cgi?id=open-pocket


So FOSS should be pure?


Ideally, yes. I can understand some exceptions there to get, say, driver/firmware support for FOSS-unfriendly hardware, but integrating Pocket - at the time - was not such a situation.

Now there's a different, much better story.


You want them to remove one of the most useful feature of Firefox?

You can remove it easily, but I'd advise you to try using it.

It personally helps me go through interesting news and "pocket" it if I don't have the time to read it now, or want to keep my time for more important articles but still think it's an important piece of news.

I have both the desktop Pocket application and the mobile Pocket application. They download the web view offline, like that I can read articles wherever I am, when I have some down time.

It's been a huge time saver when it comes to only reading what is important during the day. Having less tabs. And being able to see what I deemed important to read after a while.


"You want them to remove one of the most useful feature of Firefox?"

Completely subjective. Also pointless as it could easily just be an addon. It's just bloat however and my own subjective POV is it's useless. I've never once used it, have no desire to use and don't even know anyone who uses it.

So yeah, there ya go


The existing bookmark system does pretty much all of that already (with the sole exception of offline viewing, but Pocket's implementation is reportedly buggy anyway), and then some. It also syncs across one's devices.


the offline viewing is the only reason I use Pocket.


I use pinboard.in I never had an issue with Pocket because I never used it and just uninstalled it.


As I do a fair amount of offline reading, I'm a fan of pocket.


Dear Mozilla, I'd love to help out with Pocket.

As a concerned user, I even wrote this a few months ago http://constantbetasoftware.com/2016/09/02/pocket.html

How can I apply?


Pocket is a wholly owned subsidiary of the Mozilla Corporation, so Pocket retains a great deal of autonomy within their organization. Since Mozilla acquired the entire team, I'd start by reaching out to the individuals listed on https://getpocket.com/about/.

Longer term, my understanding is that we want to use Pocket's expertise and data to help build a recommendation engine for the open Web (Context Graph: https://wiki.mozilla.org/Context_Graph). This speaks directly to the problem of sites like Instagram which restrict links to external sites: users can enter via URLs, but they can't navigate away. (More reading: https://www.theguardian.com/technology/2015/dec/29/irans-blo...). If anything could help users route themselves out of these silos, it'd be a huge win for the openness of the Web.


Thanks, I'll try to get in touch with them.

I'd love to help on the recommendation engine. I'm (for two years now) at the top 1% of users.

I also happen to live in Brazil, to be a native portuguese speaker and the recommendation system absolutely sucks. I get a lot of articles regarding startups and Trump, but absolutely nothing in portuguese and/or about Brazil.


Well there's a Mozilla employee posting in this discussion thread, so you might start by reaching out to him. He'll likely have some suggestions.


I was shocked to read the news since afaik Mozilla had never made acquisitions like this in the past (which they mention in the post).

I'm excited though, since the reason I hadn't used Pocket in the past was because I didn't want to be part of yet another walled garden (YAWG?).

A quick look at Crunchbase shows that they'd raised $14.5M from investors. Given Mozilla's ~$300M of annual revenue, I wonder whether this leans towards an acquihire or a technology/product acquisition (definitely not a business acquisition). Difficult for me to assess the significance of 10M users for a company like Mozilla.


My general rule of thumb is that if they don't announce a price it's an acquihire.


I'll keep saying that Mozilla should acquire Fastmail.

Now that would be a strategic acquisition.


Please, no. I think part of the reason Fastmail has been able to focus on the user experience is because they are independent. I don't want them acquired, disbanded, and dumped. It's the best paid service I've ever used.


There's a joke at Mozilla that "Opera did it first". Pretty much every new feature in Firefox, Chrome, etc has appeared first in Opera.

In this case: https://techcrunch.com/2010/04/30/opera-fastmail/


Funny how far Opera has fallen these past few years. Years ago it was setting trends with tabs. Now it's yet another Chromium-based browser.


It's fairly early stage but I've been actively building an alternative to Pocket called Filltray (https://filltray.com). I found myself being overwhelmed by products like Pocket/Instapaper so wanted to build something different. A place to keep things for "short" periods of time but without the baggage of having 100s of items permanently on your never-ending reading list. A "do-later" place that may or might not come around. A couple months ago I put up a public roadmap (https://medium.com/filltray/public-product-roadmap-3964c3fa4...) and I'm continuing to work hard to produce a product which meets mine — and I think others — needs for a rethink on read-later. Please feel free to follow @filltray to keep up to date with my progress :)!


The first thing I do with a new Firefox install is remove the Pocket button, but I still welcome the prospect of seeing the Pocket service becoming open source and part of Firefox Sync. There are definitely some interesting ideas around the Context Graph, and hopefully the Pocket team will be able help move the project forward.


My impression is that most things that Mozilla starts, they shut down.

I was following their experiments, and they shut all of them down while some of them were already working.

I used to use Thunderbird years ago, but they completely stopped working on it.

I just hope they don't kill Pocket, since I use it every day to save things to read later.


That's interesting. I see they plan to open source it. I might even use it after that.


Does this mark a strategy shift away from search revenue as the primary funding source? A quick crunchbase search for other aquicisions by Mozilla yeilded nothing, so I'm curious if there are other examples of them buying up for-profits.


The only other example is the hiring of three co-founders of Humanized, https://techcrunch.com/2008/01/15/breaking-mozilla-buying-hu...


Mozilla is actively working to diversify its sources of funding; the greatest step in that was when we switched from a global default of Google search to smaller, regional deals with Yahoo, Google, Yandex, and Baidu toward the end of 2014.


This is Mozilla's first acquisition.


This is an interesting development. I am curious to see how they take this integration forward. Specifically, I have questions whether it will still have a Premium component or will it be made completely open-source.

For those interested in an alternative, I am currently building a simpler bookmarking tool [1] that works pretty much exactly like Pocket or Instapaper. Email This will extract useful content from a page and send it to you via email instead of you having to download an additional app or login to another service to access your bookmarks.

[1] https://www.emailthis.me


I think this is great. I'm weary of add-ons/extensions so only run a couple (which are from providers I trust)

I didn't really trust my saved article data to a random startup. Being under Mozilla means now I'll actually use it.


I use Pocket differently than most people here. I never save articles to Pocket — I just use their Recommended engine to find interesting articles to read. Some of these come from sources I regularly read (NYT), but others come from obscure sources or big sources that I tend not to read for whatever reason. In a small way, it helps me get news from outside my bubble, and sometimes I'll end up reading regularly from a source I discover through Pocket.

My only complaint is that a large portion of the Recommended content is junk from Business Insider. Hopefully this will wane as Mozilla's priorities are implemented.


The ability to blacklist sources would be fucking awesome.

I'm mostly limited to firewalling domains at my router. Doesn't stop them from turning up on, e.g., Web search.


Can Pocket please finally support Facebook links to non-Facebook context? And just strip the Facebook URL junk?

Otherwise, you have to click on a FB link to get the unmangled URL, thus counting against your number of free views every month.


Incognito mode.

Or Archive.is


Huge fan of the Pocket App. Any guesses for how much the deal would be worth?


I use pocket because it is automatically synced with my Kobo e-reader. One thing I wish they would offer is folders. Before I bought the kobo I stuck with instapaper just because of the cleaner interface and the option to use folders.

I get that pocket offers the option to tag, but different people have different preferences and folders would be nice.

Also a bunch of large squares doesn't seem to be the best way to organize information consisting of web blogs and pictures.

Hopefully mozilla will make some UX changes to pocket.


It sounds like a good plan, making Firefox a platform for organizing information while web browsing, research, etc.

A question: what advantages privacy-wise does Firefox have over Safari on iPad and macOS? I frequently delete all cookies on Safari and set strong privacy options. I ask because I have transitioned in the last year or so to working on my iPad and MacBook, largely letting my Linux boxes collect dust. On Linux, using Firefox was an easy decision. On iPad and MacBook, Safari is more convenient.


I never used pocket, but I use instapaper.

I wonder if Pocket will be completely free now or if mozilla counts on this to fund more of its efforts. Beyond that. I also wonder if this means a read it later service will finally properly support RSS. Forgive me I don't know the features of pocket but as i understand it is does not correct?

I been using the built in firefox RSS feeder for years. with pocket integration it'd be a lot more useful.


If you're interested in using Pocket in Chrome/Chromium but aren't ready to hand over the insane amount of permissions required for their official extension, you can run the Add to Pocket (mini) extension.

https://johnrockefeller.net/use-pocket-without-being-tracked...


Seems that the user data/behavioral patterns is really always the most important thing for this type of services huh? The aim was mostly to get acquired and it seems that the acquirer always has an eye on the vast amount of data etc.


I LOVE POCKET! I'm only very happy to hear that it's Mozilla who has bought them. Warm feelings of security and privacy.


I was wishing Pocket would offer AWS Polly playback for articles, so I started making my own app to do it. Now that it is going open source I might just piggy back their parser as its a good bit better than mine.

Hope they build in some Context Graph projects into Pocket, as there is some great ideas their.


After this, why should I donate to Mozilla? I really don't get what's the difference between donating to Mozilla and donating to Facebook. Facebook open-sources some of their stuff. Facebook buys companies to generate revenues... I guess Mozilla doesn't need help anymore.


Donations to "Mozilla" go to the non-profit Mozilla Foundation, which does exceptional, meaningful work. Firefox is developed by the Foundation's self-funding, for-profit subsidary, the Mozilla Corporation (MoCo). MoCo acquired Pocket, using its own funds.

For legal / tax reasons, MoCo can only push a tiny fraction of its revenue up into the Foundation. This means that donations are necessary to sustain the Foundation's initiatives.


I feel like sufficiently large numbers of people are confused about the arrangement of the company and the foundation (I definitely don't know the details of the money flow between the two), that this warrants some kind of easily referable page on Mozilla's website.

(appreciate you clarifying this here)


Mozilla open sources everything, and has an open Internet as its mission? I'd trust them to do that with Pocket far more than Facebook with... Facebook.


I don't understand your point. Are you upset they bought Pocket?


Mozilla's donation page says: We are proudly non-profit, non-corporate.

I don't know the ins and out of Mozilla dual identity as a foundation/corporation. But it'd be unsettling to learn that part of your donations to an open-source project is financing a corporate acquisition.


I've used Pocket in the past but when I started using Evernote I discovered that I do not need Pocket anymore. Is there any killer feature that I do not know about that Pocket offers and Evernote doesn't?


When they add ML clustering to one's pages, so that I stop using bookmarks, and find useful stuff automatically grouped together, plus recommendations...then perhaps I'll start using it


I hope they'll remove some of the "growth-hacker" stuff from the chrome extension. Modifying the new tab page and injecting links on pages by default is super annoying.


Yeah! At last I can change Feedly! Any advice before migrating?


I've been lately dissatisfied with pocket and related apps for lacking some admittedly fringe features.

The possibility of adding them myself without rebuilding the whole app is exciting!


what will happen to pocket premium?


It will fund Firefox development :P


Source?


I think, he was mostly joking, but it might actually be the case. Sounds like they want to keep the Pocket company intact as it is, just as a wholly-owned subsidiary of the Mozilla Corporation and with everything open-sourced. So, Pocket as a company should continue to try to make a revenue and if they do, that revenue goes upwards into the Mozilla Corporation (and from there upwards into the Mozilla Foundation, if the non-profit nature allows it).

There's also Callahad commenting in this thread, who, as far as I'm aware, is a Mozilla employee. And well, he sort of confirmed that this is a move to diversify Mozilla's income streams:

>bearcobra 12 hours ago [-]

>Does this mark a strategy shift away from search revenue as the primary funding source? A quick crunchbase search for other aquicisions by Mozilla yeilded nothing, so I'm curious if there are other examples of them buying up for-profits.

>>callahad 12 hours ago [-]

>>Mozilla is actively working to diversify its sources of funding; the greatest step in that was when we switched from a global default of Google search to smaller, regional deals with Yahoo, Google, Yandex, and Baidu toward the end of 2014.


There were references elsewhere in the thread to comments by Mozilla staff directly, so it appears this is true.


did anyone buy pocket premium? i always perceived it as a sales page that got no sales.


I got a year of it through the humble productivity bundle (or something similar) a couple of years ago. It's one of the few things in the bundle that I continued to subscribe to (along with lastpass) after the bundle ran out. It's cheap, and I enjoy the service. I don't really use the premium features so much that their absolutely indispensable, but they are nice. I keep a huge library of stuff in pocket and the unlimited storage space and easier searching is nicer than having to re-google for things.


I did. Mostly because I like/use the tool and I wanted to support their work.


I did for the same reasons and also in the hope of having advanced search.


I never saw any value to buying a premium. Their free features did the trick, and the only reason I may even consider purchasing a premium for are the ads, which will probably disappear after Mozilla takes it over.


I did. I'm not sure if there is any features of value, but I pay for all the small apps I use so that the developers don't go out of business.


this is really great news. i like pocket integration on kobo aura one. i hope they keep making that integration improvements further.


https://blog.mozilla.org/blog/2017/02/27/mozilla-acquires-po... explanation for this move is very laconic IMO. What are benefits? We already know majority of user were against even basic integration with pocket.


I think it would be great if they looked into merging the existing bookmarks system with Pocket!


I hope this is good news.

I've been using Pocket for a couple of years, and am a top-1% user. It's been useful, though also highly frustrating, and is at best only about a 50% solution.

I've provided considerable feedback to Pocket over issues and features. Over the course of a couple of years, few if any of them appear to be addressed, which has been highly disheartening. (Any Pocket folk: I'm "dredmorbius" at gmail, though not in my account, which is otherwise attributed, because, you know, privacy, Kristalnacht, Snowden, etc., etc.)

I'd previously used Readability, which was nice in fixing a primary problem of online content: Web design isn't the solution, Web design is the problem. But that service shut down, on 30 days notice (though with loud signs for the previous couple of years) last September. I spent several days largely manually transferring information out of Readability (404 links meant that a straight export wasn't useful -- I had to track down alternative references where possible). I'm hoping I don't have to go through that again.

I've looked at other alternatives: Instapaper (no compelling use case), Pinboard.in, by our very own @idlewords (his comment below was ... not particularly useful). Upshot: without a very specific walkthrough of the product, I can't tell if it's worth my time, and my time would be a week or month of trying to reconstruct the structure I already have in Pocket, in Pinboard. I've visited the "Tour" link multiple times, and no, it isn't clear that this offers me anything useful.

I'm thinking that RMS's model of having an email-based web-requesting systsem might actually be more useful than anything else. A Mutt (or Emacs mail-mode) searchable, taggable archive, date-sorted, threadable/groupable, would be a huge win over a lot of other alternatives. I've been thinking through other ways in which I might create my own self-hosted archive repository.

Zotero is another option, though again, I can't quite penetrate the use model / workflow.

In both cases, the prospect of being stuck with as-published layouts rather than as-useful layouts (have I mentioned: Web design isn't the solution, Web design is the problem) is disheartening.

I'd really like to have an offline / commandline tool which could manage webpage decrufting. I'm aware that it's not a fully deterministic process -- the problem with the Web, for better or worse (I've been suspecting the latter for some time) is that there are no publishing standards, and content follows no consistent form. That said, much content can be divided into two general categories: 1) known states of fucked up and 2) unknown states of fucked up. Growing the size of category one, and taking the hopeless cases and simply stripping all markup from it and starting from scratch (something I do myself, manually, far more often than I care to think) is a semi-reasonable approach.

I've written a few times on what's plaguing the Web (as have others, Maciej's rants are particulalry recommended).

Tabbed browsing: a band-aid: https://www.reddit.com/r/dredmorbius/comments/256lxu/tabbed_...

The Reference Management Problem: https://www.reddit.com/r/dredmorbius/comments/50o1jv/the_ref...

And some general feedback (wrapped around some specific feedback) to Pocket: https://ello.co/dredmorbius/post/3eucmp_s0tumjuoxmfclbw

(Yes, it's heavy on the caps and asterisks -- it's difficult otherwise to indicate important points in email. SoKookMeHarder...)

Borrowing from that last:

Don't put solving your own problems ahead of solving those of your users.

The features I do use, heavily, and would like to see improved, are article view (many websites remain glitchy, I'm well aware this is a whackamole problem, but it does need to be continually addressed), import (I've been working through a backlog of 2,000+ Readability articles migrating to Pocket, slowly, one-at-a-time, there's no automated tool), bulk-action tools, sub-corpus tools (working on a specific set of articles at a time), and tags tags tags tags tags tags tags tags tags tags tags tags tags tags tags tags tags tags tags (did I say tags) tags tags tags. And more tags.

A principle, perhaps the principle problem with the Web as it stands is organisation particularly as concerns quality and vetted content.

There are various sources who can vet content. Some are good. There's one I especially trust, and whose judgement over what I like is strong: ME.

I've commented before, AND CONTINUE TO HUGELY APPRECIATE several features of Pocket, especially the ability to peform full-text searches of my current and archived articles. It's HUGE.

One of my (many) interests is COGNITIVE LOAD. Pocket is a tool for REDUCING COGNITIVE LOAD when reading online media. I see several mechanisms:

1. A standardised presentation of Web content. I can look at an article, and identify what is the primary content (virtually all of it), what's metadata (usually the top -- you do a good job but I'd love to see this improved as well), and where the controls are (standard Pocket bits).

2. Elimination of online distractions. Sidebars. Ads. Animations. Videos. Spurious links. Shitty page layout. Shitty font choices. Shitty colour choices. Shitty CSS choices.

3. Format optimised for my device and preferences. I love you to tears for not having to fuck with zoom, font choices, background/foreground colour, etc., etc., etc., on Every Fucking Webpage. ("Web design isn't the solution, Web design is the problem.") Huge. Or is that "Hyuuuuge!" now?

4. My own curation. Anything I've saved to Pocket I've seen at least once. There's a familiarity in returning to material which makes it more valuable.

5. Tags. Again, my assigned classifications for material. (And often picks selected from Pocket's tag suggestions, which are quite good.)


I use Instapaper on a daily basis, how does Pocket compare?


These are the most salient differences in my experience with both:

- Apart from saving the parsed, "readable" version of an article, it can also save the web as-it-is. This is very useful for those times when the formatting is important (as it is in articles that include code; you loose syntax-highlighting with the readable view. Pocket can be set to try and choose the best view for each article.

- It does a great job with videos.

- It uses tags instead of folders.

- Instapaper has been less reliable for me (specially the android app), although in fairness I've used it more, so I'd notice it's unreliability more often.

I'm not that interested in recommendations, but I gotta admit that the few times I've looked at them, Pocket's recommendations were fairly good.

I've been using Instapaper for around two years now, keeping Pocket as a bag of less-important links, not-really-reading-material, and very often video. I was thinking of going back to Pocket as my main read-it-later service, specially for the first feature I mentioned.

It being Mozilla-owned and some day also OSS will surely be a further advantage in my mind.


Nowadays, I add an article to both Instapaper and Pocket. I used to use Pocket a lot but sticking on Instapaper more.

Things I love about Pocket are

- Curated (and from other people I followed) "recommendations",

- tags "Best Of", "Quick Reads".

- beautiful android app

Instapaper has less features than Pocket but I find that

- Instapaper article parser is better,

- UI is cleaner, highlights ( I love these!),

- ability to switch style (type faces, article background)

- speed read

- they also have "Browse" (something similar to Pocket's recommendation?) but in Android app, it's a bit hidden in the drawer menu so I don't use it often.


Pocket has clients on more exotic platforms, such as the Kobo eReaders.


Why would someone use pocket over zotero?


Fair question, and one I'm asking myself increasingly, though:

1. Zotero doesn't have a full-service (or any?) Android app. Which is my primary reading platform. So it's simply not an option.

2. More straightforward onramp. It's pretty easy to figure out what the fuck to do with Pocket, though after a bit, it gets utterly unusable with scale (I've got ~3-4k articles, possibly more, the fact that I can't even extract a !@#$%^&() count is a major PITA, and referencing them Simply Doesn't Work.

3. Simplified and uniform presentation. Web design isn't the solution. Web design is the problem. I prefer Pocket's styling to ANYTHING* any website can throw at me, with exceptions I can count on the fingers of one hand. Most of which I've made myself.

Frankly, I'm utterly frustrated with use of the Web as a research and archival tool, and none of the options I've seen yet are satisfactory.


I just searched zotero in the App Store and got nothing, so because every morning I save stuff from Feedly into Pocket and I have no idea how I'd do that with zotero.


Did you try googling that?


Yea, I found their website : https://www.zotero.org/ Available for Windows, Mac and Linux. So no mobile apps at all, Pocket apps are great so that's a good reason to choose them over Zotero.


In my case, because it syncs with my Kobo Aura HD ebook reader. That way I can read "my internet" offline :)


Will it become firefox exclusive?


It's more likely to get fully open sourced, I bet, which would make that rather hard?


"save to pocket" :D


How does Mozilla earn money?


Search engine deals, and now pocket


They're losing their Yahoo search money in the future and need to make it up somehow.


Two apps I no longer use because they became bloated and unfocused, they deserve each other


I will never again donate to the Mozilla Foundation.


Why's that? This acquisition was carried out by the self-funding Mozilla Corporation. The Mozilla Foundation is a separate entity.


Ahahahahaha


Mozilla can spend money on this, but not Thunderbird or XUL-based extensions?


It's more a question of "won't" rather than "can't".

Why do you think XUL and Thunderbird are good investments vs. things like Servo or the context graph initiative (which the Pocket team is going to accelerate according to the blog post)?

Do you think a non-standard framework for making fat native apps and a mail client built on this framework are going to make Mozilla more relevant?


The reason I bring up Servo is that parts of it are already starting to land in Gecko and will ship with Firefox soon - do you think making Servo support XUL is a good use of anyone's time?


It seems we have have been taken to a bizzaro parallel earth where people enjoy built it bloat that should be an addon in their formerly highly customizeable browser.


The Pocket integration in Firefox is very simple and is in fact a built-in add-on.

I think it's a failure of imagination to assume that this is going to be the extent of what Pocket will be able to contribute... they have a lot of users and popular mobile apps for instance which is useful on its own.

As the blog post points out, the hope is that Pocket will accelerate the context graph initiative.

So, I disagree that it's about "bloat" or customization. In any case, the Pocket add-on is like ~700k and it doesn't do anything if you take it off the toolbar...


I have no idea what Pocket is, but as a Firefox stalwart and a person looking to learn Rust, the fact that Mozilla is actually acquiring stuff gives me confidence on its financial stability. Very important issue in the decision making process on future technologies to back.


Mozilla has the resources to buy companies. But not the resources to make videos play smoothly. 2017 and Firefox has no hardware accelerated video playback on Linux.

Every time I read about something Mozilla does, I am reminded of this xkcd comic from 2013:

https://xkcd.com/619/


Wow, I thought it would be the other way around!


Synergy: One product that nobody uses is now bundled with another one.


> Mozilla is growing, experimenting more, and doubling down on our mission to keep the internet healthy, as a global public resource that’s open and accessible to all.

...by buying a crappy, proprietary app?


It has a huge user-base, so we can disagree on crappy. And obviously the latter part is fixable.

Pocket was originally a Firefox add-on.



... and making it open-source?


On the plus side, now it may still be crappy but at least it won't be proprietary anymore!


> ...by buying a crappy

citation needed

> proprietary app?

fixable


This will give them plenty of data for their ad network, which is a nice strategic move. They always claimed to be privacy conscious so this is perhaps a nicer way to get interest and behavior information.


I'm not sure which ad network you mean - there was an effort to serve ads on the new-tab "tiles" page but that was shut down a little over a year ago: https://blog.mozilla.org/advancingcontent/2015/12/04/advanci...


Interesting that people think Mozilla either isn't or is against being an ad network... every single attempt to do adblocking/protection ends up being an ad network eventually (if there's a corporation behind it).




Applications are open for YC Winter 2018

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact

Search: