Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: Which abandoned proprietary software would you resurrect?
478 points by geff82 on Dec 8, 2018 | hide | past | favorite | 807 comments
Hi all! Sometimes the best days seem to be behind us, even in software development.

What closed-source, proprietary software that you once loved is not being developed/enhanced any more? What more features would you like to have it in the future? Would you pay for it to be resurrected?

* Windows Phone. We need a third viable option in the mobile space. This week we are concerned about the loss of a browser engine, recognizing this is unhealthy for the market. Windows Phone was terrific in many ways and recent concepts by fans such as @boxnwhisker (Harry Dohyun Kim) [1] show how beautiful the Metro design could be today.

* Microsoft Image Composer. A little known sprite-oriented graphic arts program. Combining its sprite model with the necessary several years of modernization it would have enjoyed had it not been abandoned would be impressive and easy to use.

* High-performance lightweight desktop email clients. I was especially fond of one named AK-Mail in the late 1990s and early 2000s. Blisteringly fast on PCs of its era, a modern incarnation would seem incomprehensibly fast compared to today's bloated apps that have difficulty keeping up with keystrokes on 8-core 3.5 GHz monsters. Of course, back in the day, desktop apps were built in systems languages and not in JavaScript.

* As a broader concept, a return to prevailing use of on-device processing and computation. Today the too-facile argument that data must be shipped off-network to a third-party cloud in order to be processed efficiently means this happens all the time without users paying it any attention.

[1] https://twitter.com/boxnwhisker/ (scroll down to see a bunch of examples)

Windyws Phone's tile setup is still the best home screen setup for mobile. As I peep at my Android home with pretty wallpaper with acres of bhank space and uninformative icons, I pine for the informative, intelligent WP tiles.

Ms Image Composer was unbeatable for digital and web stuff back then. Photoshop was rather clumsy in comparison.

As for email clients, I do keep wondering how Thunderbird can be so slow, how its rendering system can be so shite.

With respect to windows phone, I really miss the tiles, the consistency of the metro style gui and other little details, like automatic do not disturb mode while driving or face recognition well before it became mainstream.

I completely agree and I used to work at Microsoft.

Microsoft lost because they didn't have the users or the apps. It's an instructive story of how an inferior product can win. I'm not saying that happens all the time or that it's inevitable, but I think it's safe to say that's sort of what happened here.

> Microsoft lost because they didn't have the users or the apps

I wonder how much of it was because they had a sorry arse mobile browser? Not supporting ontouchstart etc events must have been a killer. MSPointerEvent was awfully buggy in the initial release, and what dev wants to rewrite their web app/page to use PointerEvent?

Why wouldn't you make sure Hybrid apps (Cordova etc) work well? Why not just copy WKWebView (20/20 hindsight now!).

Their equivalent of WKWebView is some sort of zombie IE11 version (not edge, with different features and flaws than IE11 desktop). Hideous.

Disclosure: we developed a Hybrid app for Windows Phone (based on Xamarin).

Not just the browser, the whole platform was crap to develop for. My company ported a bunch of apps by a major app publisher to various versions of Windows Phone and the documentation was shitty, the APIs were shitty, everything shifted with each new version and everything was buggy. We kept asking our local Microsoft branch for help (backed by the publisher we were porting for) and more often than not they were powerless to help.

A far cry from developers, developers, developers.

Even Microsoft themselves dropped the ball, when it took over a year for them to release Skype - which they owned - for their own mobile platform (if I'm remembering correctly, i had a Windows Phone and rather liked it).

A platform without users or apps and years late is not one I'd call superior. Good ideas are not nearly enough, as they often say around HN.

I thought it was pretty clear from the poster's message he was talking about product sans network effects.

An inferior product can win because this inferior product you talked about had already won by the time this superior product came along. Microsoft just didn't want to admit it.

Dark mode was nice as well. To toggle from light/dark and have alnost every app follow suit was luxurious. On Android it's such a hodgepodge of light and dark, an affront to the eyes at times.

Also being able to select text from most anywhere was useful.

Typing on Windows Phone was a delight, for a touchscreen-based keyboard. They had their typing game on point. Now with an Android device, I find myself switching keyboard software often.

I'm cursed with thumbs that are not compatible with touch screens. Not because of my thumbs themselves, and obviously touch screen keyboards work for most people so it can't be the keyboards. I don't have abnormally fat thumbs, or any physical deformities.. touch screen keyboards just hate me, to the point where it's a running joke among family and friends. Every post I've ever made is a relentless struggle against typos that even the best autocorrect can't seem to help with. I've tried every major keyboard on Android, and have owned iPhones and every other device imaginable.

Windows Phone though? Absolute pleasure to type on. I've never used any device that was easier to type on than my old Lumia 920.

> As for email clients, I do keep wondering how Thunderbird can be so slow, how its rendering system can be so shite.

It's effectively single-threaded, with all the I/O running on the same thread as the GUI (oh import is so fun). And you need to load the entire metadata database for a folder into memory at once.

I have looked into fixing it before, but it does require redoing pretty much the entire API, which means its several man-years of developer work. And Mozilla has never invested enough in Thunderbird to let that sort of work get done.

>As for email clients, I do keep wondering how Thunderbird can be so slow, how its rendering system can be so shite.

I leave Thunderbird open in the background and come back to a notification that some component or other crashed. Happens all the time, no clue why. Email is common enough that I'd expect there to be a half-decent client for it, but I don't know of one...

(IRC is also common, at least among people who write code, but I don't know of a free client that isn't worse than running irssi in cygwin.)

Google killed the need for a decent mail client. The entire PIM space on Linux has gone through a major regression since 2012, when I shut down my XMPP/IMAP/SyncML/CalDAV/CardDAV services and started using google. It seems like a lot of folks jumped ship to web based tools about that time.

> I do keep wondering how Thunderbird can be so slow

This is a frequently heard complaint, but I can't imagine under what circumstances this would happen. I have hundreds of thousands of mails in tbird, in various accounts and folders. It's never struck me as slow. And its search (one of the features i use most) is simply fast.

> This is a frequently heard complaint, but I can't imagine under what circumstances this would happen.

I see it both fast and slow. Ironically it's absurdly slow on my 8-core desktop and just fine on my older dual core laptop (both with SSDs and same accounts, etc). Definitely weird, but this behaviour has peristed for years, so much that I had to disable TB features on my desktop just so I could type without stalling.

> first thoughts would be to compact your folders and if that doesn't help, vacuum the sqlite files.

Might also recommend a disk check, perhaps FS needs fixing.

At one point I cleared out all TB folders and started from scratch, problem reappeared, so I don't think that's it. Very strange indeed.

The items mentioned above should be done periodically, not just once. Perhaps there are some giant files in there too?

> The items mentioned above should be done periodically, not just once

While helpful advice - if you need to do this manual maintenance regularly, the application is broken.

Possibly, hard to know over this medium. I’ve never encountered the problem, but added the tasks to my cleanup command.

Not to denigrate the hard work of developers (particularly open source developers) who work on very complex application software like Thunderbird, but software designed for regular users needs to cope without cleanup scripts. (Yes, yes, I know that reinstalling Windows from scratch is a time honoured tradition...)

Sure. My cleanup script however does multiple things, tbird maintenance is only a tiny part.

Maybe the old folders used mbox, which stuffs entire folders of emails in a single file, while the current stardard (for a while now) is Maildir, a file per email.

I have ~1 million emails in Thunderbird. The GUI thread often gets blocked waiting while it's trying to open messages. Search is horrifically slow. This is on a fairly decent ThinkPad X1.

One day I'll finally take the plunge and jump to notmuch, which I've been told is the only email client actually capable of handling millions of emails without sucking.

Quiet often it's add-ons that are are at fault, e.g. the Lighnting calendar add-on can significantly extend launch time. Starting TB in safe mode will reveal if this is the case.

perhaps put your hardware and some benchmark numbers so people can compare? if something takes 2 seconds, you might consider that "fast" and others "slow". loads of things i think are slow i realize others don't notice (or care).

Apparently load time is 1.5-2 seconds. I swear it often feels slower than that. shrugs That's not bad.

The rendering corruption/lag issue with scrolling only happens while scrolling a message while in split-pane view, not when the message is maximized in its own tab.

Its search is excellent, but the load time of the program itself, of emails, is very slow for me, on both Windows and Linux. Then the rendering! As I scroll an email it struggles to render the content and lags behind. Same problems on release, beta, and nightly. Hmm.

It happpens under various circumstances, nothing easy to pin-point here.

Example: on two identical iMacs (4 cores, 16GB RAM, PCIe SSD) with about 100GB of local synced IMAP and the same accounts, it's slow for one user's iMac but fast on the other. Deleted TB and profile, fresh user, fresh TB install, fresh config; same result.

> As for email clients, I do keep wondering how Thunderbird can be so slow, how its rendering system can be so shite.

I don't have either of those problems, using TB for a decade. Not sure what your issue could be but first thoughts would be to compact your folders and if that doesn't help, vacuum the sqlite files.

It turns out my scrolling lag and rendering issues disappear when I disable smooth-scrolling. Simple fix. :)

>As for email clients, I do keep wondering how Thunderbird can be so slow, how its rendering system can be so shite.

Compared to Gmail or Microsoft Outlook online it's very very fast. Gmail lags severely compared to when I use Thunderbird.

The new Gmail interface is slow. Man, it's just so freaking slow. The fans get whirring, the network calls are probably blocked by the UI and if you delete a message and quickly close the page it never goes through.

Thunderbird struggles with managing and searching large mail archives in a way that Gmail doesn't. I use Thunderbird for a work email account, and its GUI thread regularly gets blocked while trying to search or open messages.

To be fair, my work email account is an exceptional case - it's got somewhere around ~1 million emails.

Gmail, particularly with the new Gmail interface launched a few months ago, also sucks performance wise, but in different ways from Thunderbird.

> Gmail lags severely compared to when I use Thunderbird.

The number of times where I am typing an email and somehow trigger hot keys is absolutely ridiculous. Mid-word and suddenly I've deleted the draft, changed labels, and muted & archived a critical conversation is absolutely infuriating. It's Outlook level bad.

Have Android’s “widgets” been removed? That’s a thing I still miss sometimes on iOS, most of my Nexus One’s homescreen was taken up by my calendar.

That's still a thing, it's just the design of widgets from app to app is disparate, usually visually clashing if multiple widgets are used. Not as clean and uniform as Windows Phone tiles.

There is an Android launcher (launcher 10, i think) that offers tiles, and some of them even live tiles. It's okay, not quite the same slick experience though.

They haven't been removed per se, but fewer apps tend to have them. Most Google apps have high quality widgets.

Widgets are alive and kicking.

I had a Windows Phone, it had a lot of good things about it, but I have no desire for anyone to resurrect it. There was just not enough developer interest/market share to sustain it.

I still miss having my next appointment on my lock screen. The Pixel gets this mostly right with calendar info on the ambient screen display, but still seems a bit shy about showing it.

Speaking of e-mail clients, come to the bright side and use blazing fast tools like `mutt`. Thanks to such tools, my day-to-day "computing experience" has become far more pleasant for the past five years.

>Windyws Phone's tile setup is still the best home screen setup for mobile. As I peep at my Android home with pretty wallpaper with acres of bhank space and uninformative icons, I pine for the informative, intelligent WP tiles.

So download widgets to show what you want?

I've invested a lot of time trying to find solutions in that manner and have always ended up with hacky results, usually hideous as well. There is no cohesive solution for that Android that I've been able to find.

> As a broader concept, a return to prevailing use of on-device processing and computation

I don't think anyone ever liked developing for the web. It solved three problems pretty well:

- Data portability: log in from anywhere, have your data at your fingertips.

- App model: no broken installations of programs into desktop environments, dependency problems, DLL hell

- No concern about the mostly broken Windows desktop: viruses, etc.

(1) is mostly a solved problem even on desktop, the web just got there first. We have everything from Dropbox to various data APIs.

(2) is largely solved in mobile app development. It's instructive to consider how: very locked-down app APIs to the operating systems (e.g. no dumping files all over the place), standard packaging formats and distribution channels (.apk, .msi), etc

(3) is also mostly solved on mobile.

I would go as far as to say that in 2018, software should be native-first. And indeed it is, in mobile. The major shift is desktop->mobile. Mobile development has much more in common with native desktop development than web development. The reason desktop isn't likely to come back is because it's a niche market. Everyone has a phone but if you aren't in tech circles, it's surprising to see how many people barely use computers at all, especially those who don't work in offices, older folks, and the less well-off. A mobile device does about 99% of what they need -- most people aren't "creators", they don't author websites, write code, edit videos, design buildings, etc.

>Everyone has a phone but if you aren't in tech circles, it's surprising to see how many people barely use computers at all, especially those who don't work in offices, older folks, and the less well-off.

I work in a warehouse, and the specific thing I do requires a laptop. They've tried to train backups for me, but they have a hard time finding people who know how to use a computer. Most people don't - they only have a smartphone.

> It's instructive to consider how: very locked-down app APIs to the operating systems (e.g. no dumping files all over the place)

But this is precisely the problem. And also why mobile apps aren't really native, nor becoming native. Developers think, "since our app is just a thin wrapper on our web service, why not just make a webview and do everything through the web?". APIs being opaque to the system means you don't own the data. Saving actual files in actual filesystem is still much better for user freedoms.

> A mobile device does about 99% of what they need -- most people aren't "creators", they don't author websites, write code, edit videos, design buildings, etc.

True to some extent, but how many of those things they would be doing if the mobile platform wasn't limiting them? Users don't imagine the hypothetical things they could be doing, they choose from what's available.

The Windows desktop is "mostly broken"? That's news to me.

It’s broken insofar as there isn’t a complete and coherent “developer story” for building a Windows desktop application that doesn’t look like it came from the 1990s.

I call it “the Photoshop test”: Pretend you’re Adobe and you just invented Photoshop in 2018. How do you build the GUI for each desktop operating system?

...considering you need to do things like OpenGL hardware acceleration, 30-bit color, support binary plugins, and access certain low-level hardware features (e.g. supporting TWAIN, Pro-level broadcast media devices and interfaces, etc).

On Windows it’s an unfortunate situation because all of the “new” application frameworks since 1998 are unsuitable: WPF is .NET only, UWP requires your application to run in a sandbox, you can’t even use MFC because it’s tightly coupled to User32 with its lack of support for 30-bit color. The only choice is the long hard road of doing almost everything by yourself through low-level interfaces with the DWM. Not pretty.

The sad thing is that UWP is very capable - if it weren’t for the sandboxing then it would be perfect (that, and you would need to build your own widget library as UWP’s stock set is almost entirely touch-first controls that are utterly inappropriate for a mouse-first desktop UI)

Why is the sandboxing such a big issue in the case of a “Photoshop” app? The UWP api seem to cover most OS services that the devs would need, or am I missing something?

UWP with its XAML framework forces you to use Direct3D, there’s no way to use OpenGL efficiently - you also are forced to surrender a lot of control to the OS, for example, all Windows Store games must use Direct3D and were forced to use VSync until very recently. As a developer it means you can’t do anything cutting-edge until after both Microsoft and NVIDIA/AMD say-so and officially support some feature. This is the unfortunate consequence of UWP’s style of API sandboxing (compared to, for example, a hypervisor approach with end-user approved red-pilling). If you were Adobe it would be difficult to maintain your reputation as the leading vendor for desktop graphics and media software if you were artificially constrained by other companies who haven’t demonstrated commitment to supporting novel platforms beyond a couple of release cycles (see: WinJS, WinRT, Windows 8 XAML, etc).

What I meant was, if you look at the state of non-expert users' PCs, they're unusable by power user standards.

There's so much malware, anti-virus that slows down all I/O, browser toolbars, etc. Just open Internet Explorer and see how long it takes. On my computer it takes <1s. On many computers I've used in peoples' homes, small offices, etc. it's more like 10.

Calling that anything but "broken" seems like a misnomer. Look at the 99% of computer users, not the 1% power users/people at companies with good IT groups.

This is an exaggeration, especially on modern Windows. I have a decent anecdotal sample size of average people whose PCs run fine and have no apparent infestations, including someone in their 70s with a laptop running Vista.

>Windows Phone

In a similar vein, Palm webOS. I still maintain that it was several years ahead of its time with the whole "mobile apps using javascript" Enyo framework.

If love to have PalmOS (pre-WebOS), still superior to Android or iOS in many ways.

Didn't it live on as webos (now in many smart TVs and printers)?

I believe LG Smart TVs are webOS based.

Under the hood it's webOS, but obviously a custom UI for a TV. I have a 2017 LG with it and it works rather well, no issues with 4K HDR on Netflix and YouTube, and other than booting, it reacts instantly as a TV should. It's a shame there are limited apps for it though - I'd really like apps for Steam Link and Kodi.

I think webOS has been open sourced and there used to be some effort [1] to port it to devices starting with the nexus 4/5. Seems the effort is still decently active with a release just last week[2]

[1]: https://en.wikipedia.org/wiki/LuneOS

[2]: https://pivotce.com/2018/11/28/luneos-november-stable-releas...

I think he means PalmOS, which has nothing to do with web or webos.

> compared to today's bloated apps that have difficulty keeping up with keystrokes on 8-core 3.5 GHz monsters

Seriously, I recently had to disable Thunderbird's spellchecking so that typing emails wouldn't stall for 5+ seconds at a time on my 8-core machine. Ironically, my older dual core machine has no such problems, and the same email accounts, OS and extensions are at play.

Something is wrong, see my other replies for potential fixes.

I really hope Purism's Librem5 pans out for this very reason. A third alternative. I don't want Windows Phone, since that would not buy me much more privacy than Android with today's Microsoft, so having an open 3rd party mobile platform would be the solution here, I think.

There's SailfishOS, which is an alternative to Android and iPhone. It's a mobile Linux OS with an optional support for Android apps.

It's not the world's most advanced mobile OS, but a good way to escape the Apple and Google ecosystems.

It is possible to purchase the OS and install it to Sony Xperia phones. https://shop.jolla.com

I guess it’s not even available for viewing in my country (the United States). Coincidentally, this must be close to what it’s like in the UK (just on a much much smaller scale) after the GDPR stuff.

I hate to be so blunt but it won’t pan out in any meaningful way. Why would it succeed where FirefoxOS and Ubuntu Mobile could not?

Not enough people value openness to make the trade off versus iOS and Android.

Does iOS not provide the privacy you need?

> Why would it succeed where FirefoxOS and Ubuntu Mobile could not?

I think Purism would be satisfied with much less initial adoption on the overall market than somebody like Mozilla/Canonical. In short, they'd probably consider it a success much sooner. Also, both FirefoxOS and Ubuntu Touch were essentially adopting the same philosophy as Android in terms of having a closed-hardware approach and a software stack that while open in principle, was largely restricted on-device to run just the apps available in the respective app-stores, made for the platform specifically, which weren't that many.

Purism is making it so that practically any Linux app can be installed out of the box and with a small amount of work, any GTK3/4 app can be made touch friendly as well. I think this could provide them the app ecosystem head start they'd need.

> Not enough people value openness to make the trade off versus iOS and Android.

I think you're, sadly, right. However if PurismOS becomes a solid choice in its own right and then you have the privacy advantage on top, it might sell.

> Does iOS not provide the privacy you need?

Kind of, that's what I use now, however I have to trust Apple on keeping its word, which may be difficult considering they're moving more and more into "services" == rent seeking. Also, I have serious problems with the war on general purpose computing Apple is involved in and the closed nature of their software, where they have the ultimate say in what I can run on a piece of hardware I paid thousand pounds for to own, not rent.

I agree it is an exciting phone, but what are the privacy gains versus an android phone with lineageOS and f-droid foss apps? AFAIK it's just the baseband code. Convincing regular users that opensourcing the baseband will solve any problems seems pretty DOA when you have the very visible google play services tracking issue and no on cares.

> what are the privacy gains versus an android phone with lineageOS and f-droid foss apps?

Google oozes from every part of Android. I very much doubt that installing LineageOS with closed-source blobs for most of the hardware makes Google/NSA/OEMs completely unaware of your activity. Moreover, Android is such a critical piece of software that probably every significant 3 letter agency in the world has some sort of a backdoor in/to it. Android suffers a security fiasco quite often and the update situation is quite awful. Lineage is also often not as stable as one'd like and the update process from one major version to another can be itself nerve-wrecking.

On the technical side, it is running ancient, out-of-tree kernels that lack a lot of the security work going upstream, which is especially concerning considering there's a new Spectre variant patch in practically every release.

Lastly, having the full GNU/Linux experience in my pocket is something I've always wanted.

Whether 'regular users' care about this is not particularly concerning to me, as I think the 'enthusiast crowd' is large enough to get the ecosystem started and once we feel comfortable recommending this to our 'regular' family members and friends, they need not to be aware of any of the additional benefits - it would be just the new smartphone that they have.

It is not just about privacy. With Android phones, you will sooner or later end up in a position where you can't install newer LinageOS versions, simply because your kernel is too old and the required binary drivers cannot be ported to newer kernel versions. Maybe project Treble helps to solve that problem, but it will still take a few years until we will find out (unless Fuchsia kills Android).

Nevertheless, I hope that Purism will be able to deliver a product which is as easy to use as a "normal" Android device with those privacy/sustainability features as a bonus. That way I might be able to give the people who ask me which phone they should buy next some good advice.

Note that there is KaiOS -- popular in India and Brazil but under the radar on HN because it's currently mostly aimed at feature phones. It's based on FireFox OS (hence probably closer to PalmOS than Windows Phone): https://en.wikipedia.org/wiki/KaiOS

That was what I posted in the Firefox ( or was it Edge ? ) discussion, Mozilla could have had Firefox OS being the 2nd most popular OS. Instead they handed it to KaiOS.

While we're at Windows Phone I must say I miss Windows Mobile. Always found the decision to kill it and replace with Windows Phone a total disaster. At the time it still had a very dedicated user base built over many years. It got replaced with a half-baked copycat OS. I doubt many WM users had any reason to switch to WP instead of Android. It was a completely different product. Microsoft basically threw away a dedicated niche audience for a different niche audience. And I don't have any data on this but I suspect WM users were a much more affluent, valuable and "Microsoft'y" group of users.

JavaScript isn’t that slow and bloated, the browser rendering engine is a memory hog and cause of most pain. Electron is the big offender because it’s a copy of a new browser everytime.

We have TypedArrays, WebAssembly, very smart JITs. With Typescript, JavaScript is a very productive and sane language to work with.

I 100% agree with you that the current state of desktop apps are extremely bloated.

> With Typescript, JavaScript is a very productive and sane language to work with.

Mostly agree with your comment on the broader point that front-end (coding) is a lot saner than it used to be. This sentence however sounds to me like "With C, assembler is a very productive and sane language to work with." (FTR: I used to enjoy assembler.)

PFE - Programmers File Editor. Was way more advanced then any of the editors available now. Could handle GB sized files easily. Had macro that can record your operations and playback. Was very nifty and virtually consumed no memory and was simply superb. I still use it though as it is available in 32 Bit version.


Would love to see a 64 bit version and with some more advanced features.

Still loving it and it's one of my main stream editor which just works with just one executable and part of my big list of portable apps collection which mostly consists of single executable based tools for almost any task!!!

I was just talking about how it was a shame windows phone didn't catch on last night, I really think they should have doubled down on it at least as hard as they did on the original X-Box, it really felt like they gave up on it too soon.

I'm feeling the same way about Edge.

electron eats so much cpu & memory simply because it's still a web browser, it has to support all those 20 years of HTML/CSS history and it was designed to do so for 10s of tabs, so it has a lot of caches and other things.

and while javascript is certainly one of the worst choices if you care about performance, it's also one of the best choices if you need to iterate really quickly (which I believe is very important in overall, especially for new projects)

I hope there will be lighter electron one day (I'm actually working on something like this already) but it's unlikely that people will start doing UIs in systems languages.

> electron eats so much cpu & memory simply because it's > still a web browser, it has to support all those 20 years > of HTML/CSS history and it was designed to do so for 10s > of tabs, so it has a lot of caches and other things.

XULRunner does the same thing, includes, in addition, a full XML GUI language and assorted technologies and the full runtime distribution, on my Windows10 system, including debugging, crash-reporter aids, VC libraries, D3D libraries, consumes 72MB.

I wonder what do you think about this: https://github.com/cztomsik/node-webrender node+react+servo, consumes ~30M ram on my osx (if you precompile typescript files)

it's nowhere near to XUL but it's also little more powerful because of react (vdom is more dynamic than XML, it will be possible to use react-devtools and it should support HMR too)

I am actually only interested in scriptable XML technologies, so pretty much like Mozilla's XPFE (XUL) was.

I often wonder why/how xulrunner has descended into obscurity while electron has blossomed. Anyone have input on that?

I tried building something with xulrunner a while back and the experience was awful, IMO. I was up and running with Electron within an hour.

xulrunner may be a superior technology, but if it's a PITA to work with, then nobody is going to want to use it.

I haven't used electron, but in 2007 xulrunner was such a pain in the ass to write code for it wasn't worth the effort for anything where you didn't absolutely need an embedded browser, and even then there were better options.

Documentation was poor or non-existent and IIRC the API was based on the half-baked XPCOM because COM and CORBA were all the rage at the time it was created.

> Documentation was poor or non-existent

That's absolutley not true.

"Rapid Application Development with Mozilla" has been published before 2004. It is still online: http://mb.eschew.org/

XUL (and assorted technologies) is/were broadly documentet on MDN.

Not to mention xulplanet.com which had been very helpful as well.

Wasn't xulrunner based on XPCOM which Mozilla deprecated? I think this is one of the technical reasons for its decline, but I think the biggest reason was timing.

When xulrunner was being developed there wasn't the same interest in desktop software. It seems like Chromium reached a point where you could build desktop-class software, but the distribution model wasn't right. Someone then decided to adapt Chromium to solve the distribution problem and you have Electron.

If Firefox had evolved to allow the tipping point experience that Chrome did, then you might well have seen xulrunner be that thing. It was years ahead of itself, but perhaps you could question the focus of the folks on the Mozilla side who were spread across so many concepts/ areas compared to the folks at Google - but I think that's for a different thread.



Would you please review the site guidelines at https://news.ycombinator.com/newsguidelines.html and stop posting unsubstantive comments to HN?

Comments about downvoting are nearly always downvoted.

You didn't give an answer; your post was downvoted for not contributing anything.


How was the gender of Mozilla’s lawyer a factor in this?

The CEO, back then, was a female lawyer.


JS had nowhere the popularity it had today. At the time, creating a desktop app in Javascript was not regarded as serious. The reason electron became popular is because so many people from other fields (designers, integraters, students...) learn programming through copy / pasting JS and wanted JS for everything. Hence node. And now electron.

Firefox, Thunderbird (and their spin offs) were all considered serious desktop apps. They all were written with XPFE/XULRunner.

They didn't advertise that much, though. I only learned about XULRunner when I was trying out a keyboard-driven browser (Conkeror). Unlike Electron, XUL wasn't promoted.

And that is the main reason, IMO. Mozilla Inc. never wanted to do more with it.

uh, I wouldn't say electron got popular because of copy-pasting newbies.

webapp development is a lot of work and SPAs are actually even harder... and so is server development with node. I love javascript, yet I don't think it's suitable for everything.

electron got popular, because companies could reuse code and tooling, only for the cost of cpu & memory

Why is JavaScript the worst choice of you care about performance ?

Because the runtimes aren't meant for small, high-performance app, and the language itself doesn't (or wasn't, until recently) give you the primitives necessary for writing high-performance code.

automatic memory management makes certain optimizations very hard to do (you can't make cpu-cache-friendly array of structs)

javascript is very dynamic and V8 has very limited time to make its magic so the optimizations can't ever reach level of rust/c++ (but V8 can do some runtime-only optimizations)

it's very hard to predict performance of given code (and this itself is why it's the worst choice for perf-sensitive code, V8 is moving target, and what was fast yesterday might be slow tomorrow - so basically, it's not worth time and money to actually spend too much time optimizing javascript, it's much better to identify hot spots and do them in rust/c++)

It usually doesn't matter and it's more than fast enough for any kind of scripting, but it's certainly not a good fit for cpu-heavy tasks. I love javascript, to be clear.

BTW: run some node.js code with --print-opt-code so you have an idea how big amount of code is generated even for very simple things.

"High-performance lightweight desktop email clients"


Don't dismiss it because of the lack of bells and whistles on its web page, it runs circles around everything else. Outlook mail folders also can be imported through an external utility. Runs on Linux, FreeBSD, Windows and reportedly can be built on MacOSX.

Evolution is also OK. Not sure how it compared to claws-mail feature wise but they both have a long history.

> High-performance lightweight desktop email clients


"Continuously improving since 1998"

> We need a third viable option in the mobile space

Agree. It feels like apple respect users privacy which is great, but are you not interested in ios you are stuck with android and how google collects everything about its users, which feels really bad. I also think that "google play store" is rather bad. So it really needs competition here and a 3rd option.

What I really hope is that Microsoft (which have the muscles) should embrace android fully and build an "MS android app store" and an ecosystem around it for android so there could be an real option and competition to google and "google play store" on android.

WP died a lot because it did have to few apps. But because this should be android, developers shouldn't need to convert any apps, just simple upload the same app to "MS android app store" too, more or less. This way they could get a lot of apps in their app store and fast. To get developers to upload they could give better deals (ex lower %) and for users they could give an better user experience, focus on privacy like apple to differentiate them self from google and play store etc to get traction and popularity.

If MS did and got some traction, smartphone manufacturers would have an real option not to be forced to install play store and all googles apps just to be able to sell their phones, because there whould be an viable option for it users to download all android apps they want anyways. And competition is always good for consumers. At least for me, I would love to buy an android phone without google, with someone that respected and focused on privacy, and with an good app store out of the box.

I'm also a big fan of Windows Phone and am sorry to see its demise. While Android and iOS now share many similar UX interactions, Windows Phone, in contrast, is a refreshing change.

For those of you haven't tried a Windows Phone, this video gives a good overview of some features:

I Used a Windows Phone for a Week in 2018! I Will Miss It:


For email client, what about Claws? https://www.claws-mail.org

It seems to be missing an OSX client, but Windows and Linux is supported.

It is pretty telling of just how successful Yahoo, Google and Microsoft (and others) have been with webmail that the best client for multiple email accounts available is Thunderbird. It isn't awful, it works, but it is also kind of sluggish and pauses in the interfaces all the time.

"desktop email clients" - I run thunderbird and "eM Client" - both are handling a large amount of email fine most of the time.

When I have 100 firefox tabs open and it starts to go wonky my thunderbird also has issues doing just about anything. Sometimes even get the "stop this script" on occasion.

I think this has more to do with the java heavy pages in firefox getting bogged down while other software is trying to run backup tasks and AV and such.

I still enjoyed the older outlook, but with the current two I'm not longing for shiny new tools lately. I also realize that my lack of trust in cloud servicing means I am not using some likely popular features like calendar notifications and other things that may be important to others. So bias, ymmv, etc.

For email, on Mac I’m using MailMate. It’s everything I wanted Thunderbird to be.

On the other hand I think the web is awesome and I’m 36, I remember very well what the landscape looked before web apps. So I don’t understand the fetish for native to be honest.

Yes, +1 for Windows phone! I found it much easier to use than the iPhone in general, apart from its lack of apps (which of course is what killed it).

"As a broader concept, a return to prevailing use of on-device processing and computation."

Will probably come back with edge-computing.

> High-performance lightweight desktop email clients.

Interesting, haven't had performance problems with email in about well, never. Been using Thurderbird for a decade. Not perfect, but slowness isn't a problem.

Every time I try searching my Inbox in Thunderbird, it feels painful. Meanwhile, using my mail provider's (Fastmail) web interface, its snappy.

Don’t have a problem with that either. Try the tips in my other replies.

If you resurrect those technologies they would die again right after.

Nobody cared about Windows phone and nobody noticed when it died.

It enjoys the same prestige as Microsoft Bob.

> "High-performance lightweight desktop email clients"

I hope the SublimeHQ gods are listening to this

> Windows phone

The gigants buy up competitors and bury it, shot down the project.

We would have more freedom and viable options as consumers if not the big companies such as Facebook, Google buy up the competition. It should be illegal to do this. The world is a more poorer marketplace because of this.

They bought... Windows? Microsoft is one of the giants here. People didn't like the phone.

That and Microsoft didn't support developers.

Picasa (desktop app). It hit a sweet spot for me - linux compatible, very easy to do the majority of simple edits (crop, rotate, colour tweaks), and a few other nice features.

The only significant failing was that it couldn't handle removable media at all well.

Shotwell is the best I've found so far, but it's not quite the right feature balance for me.

https://en.wikipedia.org/wiki/Picasa https://en.wikipedia.org/wiki/Shotwell_(software)

I used picass for one thing and one thing only: image organization.

I could point it to a directory and it would read all the images. It wouldn’t copy them to it’s database, it wouldn’t try to change their format. It would simply read them as fast as I’ve ever seen and show me what’s in that directory. Then it would keep the directory hierarchy in place. I haven’t been able to find anything similar. I search for a similar piece of software every now and then but nothing shows up. Lightroom is close but it’s god awfully slow. I just loved how fast Picasa.

I actually did some cleaning on my computer last night and found the last Picasa dmg. Maybe I run it and see if it still works.

Picasa was definitely a state of the art app. Image controls, overall speed and smoothness of the user interface! I still cannot find anything close to it. I ended up using ImageRanger for browsing my pictures and it seems to be quite fast and non intrusive. It looks like these days almost every device and app is trying to force you to use their subscription based cloud storage or service.

Have you tried XnViewMP[0]? I use it on Windows, but they also have OSX and Linux builds.

[0] - https://www.xnview.com/en/xnviewmp/

What about ACDSee?

I missed it too!

After the Nth photo sharing website that I'd early-adopted decided to close up shop, I determined I wanted to own the next solution I invested time into, and I founded PhotoStructure.

I've got 20-odd hard drives from laptops and servers and backups. No software that I tried, either open or closed source, would do what I wanted: organize everything into a nice, timestamped, deduped folder structure.

Many years ago, I'd shot myself in the foot by using tools to do JPEG editing and rotation, but those tools quietly deleted EXIF metadata, so PhotoStructure applies a suite of metadata inference heuristics to heal those holes, too.

The MVP is focused on high-quality metadata extraction and inference, and has a simple web-based UI. Simple editing support, along with GPS POI and face detection, is planned.

After spending more than a decade in the ads business, and (helping build) ML-powered behavior targeting based on metadata, it blows my mind that so many of us give the most rich metadata stream, our photos and videos, for free, to the FAANG. PhotoStructure isn't just an effort of love, it's also, at least in some way, penance.

I've got a limited number of beta users trying it out right now. If you're willing to share your feedback, please consider signing up. Use of PhotoStructure during the beta period is free.


As a semi-pro photog who is very keen to quit Adobe completely, I'm very keen to try it out.

Curious about privacy aspects though. When you say it's a private cloud, does it communicate back to your servers at all?

Any goals for Linux compatibility?

I'd love something that can store the photos on a NAS or network share drive, runs its processing in a Docker container on my LAN, and serves a web app locally with absolutely no external internet access. That's basically my dream photo manager. I can connect to my own LAN remotely to access it then, without any need for the privacy/security risk of hosting my (and my clients) photos on some 3rd party server.

I have installers for Mac, 64 bit Windows 10, and 64 bit Ubuntu. My CI suite runs all ~2,600 tests on all three platforms after every commit.

Your images, videos, and metadata stay yours, and are not uploaded anywhere.

Currently I've got error reporting that phones home if there are critical problems detected, but the log events only include the stack trace and possibly the path to the problem file.

PhotoStructure spins up a webserver bound to localhost by default. In other words, other machines in your LAN can't open the PhotoStructure web UI (unless you set an environment variable or use ssh port forwarding).

Sign up via the website or send me an email, I'd love to hear your thoughts and feedback.

I've signed up via the site. Sounds very promising.

I still use Picasa, because it is so fast and easy to just copy an SD card somewhere to my hard drive and have it show up at the top of Picasa seconds later.

Then I browse the pictures, star the ones I like and process them.

An no other software I have tried so far comes close in terms of speed and UI efficiency.

So I would also love to try PhotoStructure and have signed up via your landing page.

signed up. ready to provide feedback. :)

My favorite feature of Picasa was duplicate image detection. IIRC, it would actually find duplicate images by the visual of the image and not just the bits. So it could match smaller versions of the same image. This helped me clean up countless duplicate copies of photos all over my different disk stores.

I have found Pix to be the good replacement of Picasa.


Never heard about it before. How did you find out about that one?

Default installed on LM 19. It’s great

any info on a port of Pix to Mac OS? Looks great?

That might be interesting, you may try to compile it on Mac like any other glib/gtk app.

For my research I have scripts that scrape various sites, downloading images and encoding the web page info into the JPG IPTC metadata fields. Upon opening Picasa, it automatically detects the newly scraped photos and imports them along with the IPTC data, which can then be searched. Picasa is amazing for its speed, indexing, and use of metadata. I wish Google had open-sourced it. I have no interest in loading these huge libraries of photos to the cloud. I'm wouldn't say I'm a data hoarder, but I have 4TB of images and don't want to pay Google or anyone else $x/mo. to store them. As more software moves to the cloud and requires us to also move our data to the cloud, I'm afraid options like Picasa will disappear.

Are your scripts available? I do something similar but hadn't thought about encoding the web page info into the JPG.

I am still mad at Google for killing this project. Picasa was the best photo app EVER! It was also very commonly used among genealogists and in genealogy centers where visitors often have low computer literacy. The facial recognition was fantastic! As far as I know, it all ran locally. It was great to be able to just select a person and go through all of the photos with that person in them. Great for genalogy, too. I wish they would open the source code on this!

Elegant easy to use interface. Something my wife, mother, and others without any computer expertise used religiously and were saddened by demise. Now Google Photos on mobile is pretty good, but the web UX feels like it was pasted together without any thought as to user experience and both seem to make assumptions about what users know.

I've thought about it (more than) a bit.

Part of the problem is that only a small fraction of the population needs a high-performance photo manager. Most of the market has always been for "color snaps" and that has just gotten worth with the proliferation of cell phones.

If you are handling lots of RAWS and you care about speed then you have to start with hardware.

For instance, if you care about performance you just can't use a mac. Forget about it.

An application that cares about performance might tell you to ditch the hard drive on your computer for an SSD if you want to run it and that we'd rather give you your money than just give you the same bad performance you expect from Lightroom, Photos, etc.

Having your images on a RAIDed network server could be good but if you are using WiFi performance will be bad. If you feel entitled to keep using your 100 Mbps Ethernet hub on your DSL modem than performance will be bad.

Wedding photographers spend $5000 for a camera and would probably get much more than $2500 of value from a $2500 photo management suite, particularly if you factor the lower blood pressure from not staring at a spinning beach ball all day and the lower health care costs and extra years of life they could get.

Knuth's "premature optimization is the root of all evil" might have made since back in the 360 mainframe day when you couldn't get N very high so an N^2 algorithm wasn't as bad as it is today.

People today have the wishful thinking that they can find the "one" bottleneck and open it and that was sometime true in the past for prototype projects (eg. less so in the age of networking, deep cache hierarchy, ...) but in real-life systems there are usually 5-10 bottlenecks that all need to be cleared if you want to make a difference in performance that people will feel.

Most people think they can clear two or three of them and will spend a lot of effort for it and will argue until they are blue in the face and bankrupt that they don't need to fix the other ones.

Yep, another one for Picasa. I still use it, was able to grab the installer from one of Google's internal home pages,, when they took it down on .com

Love it for quick and easy edit controls. Like the speed and looks. Wish some remaining bugs were fixed.

"internal" = international ( i think it was uk homepage)

I miss Picasa so much. I switched to Lightroom but I hate the subscription model and how heavy it is. I just want a light fast way of managing 500k photos and doing basic edits. Rotate, crop, export at specific resolutions for sharing, tagging, etc. I appreciate the power in Lightroom but I just don't need it.

Also, I'd love to switch to Linux full-time but that's the main program I'm lacking. I've tried most the suggestions but none handle high numbers of photos without becoming intolerably slow.

irfanview should be adequate, .. wine on ubuntu

Google photos is much better. There's not really any reason to continue using Picasa anymore. Most of the features it had were extremely primitive, like the facial recognition.

I don’t think Google Photo is a desktop software. And it cannot be used offline.

The Opera 12 browser, for umpteen reasons:

- UX-wise, it solved many issues browsers today are still struggling with. Vivaldi is working in a similar direction, but they've basically started from scratch again so aren't close to the point of development Opera had achieved by version 12.

- These HN posts from today's frontpage [0] [1] give some idea of why we really need Presto today. Not only would it just be good to have more competitors in the space, but also, for most of its existence Presto was ahead of its competitors in every detail (except for the one important, and very political, issue of "site compatibility", a.k.a. making badly written websites work). It also, similarly, got many UX points working at an engine level that modern browsers still get wrong, like text selection of html content and link text, responsiveness of scroll and of page links during navigation, etc.

- M2 will never be surpassed by Thunderbird, and quite possibly will never be surpassed full stop (though I really hope I'm wrong about that).

... so many other things

[0] https://news.ycombinator.com/item?id=18626316

[1] https://news.ycombinator.com/item?id=18622516

My favourite part of Opera was Opera Unite. People made fun of it because it was advertised as reinventing the web and stuff like that, but it was really nice. Basically what it did was providing an interface for hosting a webserver in the browser and having a central repository to install extensions for it. There were extensions with simple games, pastebin functionality, webhosting. It was very easy to setup and share stuff with your friend. Sure, you can host a webserver with nginx and install something on it, but it will take you hours or days to setup, while this was instant.

Just FYI, you can select link text in Firefox and, I think, Chrome if you hold Alt.

Learned that rather recently even though I'm practically living on the web.

And if that doesn't work for you, for Firefox there's https://addons.mozilla.org/en-US/firefox/addon/drag-select-l... which is awesome. It is one of the first add-ons I always add.

Thank you so much for this. I always have trouble selecting parts of links on reddit.

The Presto source code was leaked not too long ago, so legal issues notwithstanding, you might actually see this one happen. Some Russians and Chinese have already managed to build and patch it with new features.

The commit SHA-1 at the tip of the leaked presto.git repository is 8cd9aeda16e82d5babc1070a0791f0ef2de6fad5, which helps to find it in various places.

Microsoft MapPoint. This was a phenomenal piece of mapping software and had features I haven’t been able to find in any of the online map replacements.

When getting directions, you could designate a number of areas on the map as “avoid area”.

You could map all sorts of census data. Want to colorize the map based on crime rate divided by median home price? No problem.

You could import lots of data from a spreadsheet and do useful things with it. I used this when looking for my first house. I scraped the data from realtor.com and made maps where the icon changed on the number of bedrooms and the color changed based on the price.

You could make drivetime zones. Want to see how far you can drive in 30 minutes from your office? No problem.

You can probably do a lot of this with ArcGIS, but I don’t know since its way out of my price range.

I keep hoping that Microsoft will bring some of this to Bing Maps. I would pay to use these features again and adding them to Azure would definitely differentiate them from Google Maps.

I miss MapPoint too. I used it as a component of a silly web scraper I built to plan routes for visiting yard sales many summers ago. (My wife and I were really into yard sales at the time.) It would scrape down the local newspapers' classified ads, do some processing on the text, and output an optimal route ending back at home taking into account the varied starting times for the sales. It was easy to get MapPoint to do cool stuff.

QGIS is free/open source and can do a lot of this type of stuff too. More like ArcGIS than MapPoint.

I actually like QGIS more than ArcGIS. It's not without some rough edges, but I found it way easier to use than ArcGIS (which is a bit of a UI disaster, IMO).

Thanks for the intro to QGIS.

Won't happen. Apps with those features get accused of being racist.


The OP appears to be talking about just mapping demographic datasets, plenty of software does that today and isn't accused of being racist.

The issue you linked to seems different. I agree that the headline is hysterical, but the patent mentions, among other sensible ideas, that it would plot paths factoring in demographic data. Which is just plain weird when you already have crime data available.

Some modern luxury cars have this feature built in. And what’s racist about avoiding dangerous areas?

Please keep the whole "we are being surpressed" bs out of HN.

Well we're definitely being suppressed by you it would seem!

Esri offers an ArcGIS for personal use program. $100/year gets you ArcGIS Pro, ArcGIS Online, and a bunch of the apps.

I've never used MapPoint so I don't know how it compares - I expect the ArcGIS stuff is more powerful but less usable (for your particular needs) than a more focused solution.

Note: Esri calls isochrones (30 minute drive time) 'service areas' and you have to search for it under 'network analysis'.

I would resurrect Microsoft Streets and Trips. I have never used it, but my father still does, and swears it is better than any online mapping software. Part of the reason I would like to resurrect it is that then I would no longer have to figure out how to install his ancient CD copy of it on whatever netbook he buys every 2-3 years.

Some of the fondest memories I have from childhood involved Microsoft Streets and Tips. Father was a military man, so we were always traveling either for moves, or for some other reason related to his career.

He delegated to me the task of putting together his route using Streets and Tips, a great introduction to navigation, mapping and spatial reasoning. Been a cartography geek (not to mention massive road geek) ever since.

Take the time and create an iso of that and throw it on GDrive/OneDrive/Dropbox.

Will be a lot easier in the future.

> When getting directions, you could designate a number of areas on the map as “avoid area”.

It's noteworthy, because I've only seen this feature in Rolls-Royce navigation in recent times.

OSMAnd and Navigator on Android allow you to "disable" road sections. Not an area feature, but for things like the I-74 Bridge project where construction blockages change sometimes almost weekly, it's a big help to toggle individual blocks/intersections individually.

Not very discoverable, this is the first I read about it.

Fairly sure my parents old TomTom navigation thing had that option too.

It does seem somewhat paranoid. Might be useful to exclude certain routes with an easy interface, but tremendously harmful by seemingly legitimizing the myth of “no-go” areas that has recently become a favorite talking point of the alt-right in my (perfectly safe) country.

There's a neighborhood along my drive home that Waze will often try to navigate through to save a few minutes of driving. This rarely actually saves the time it claims, and is disruptive enough to people living in the neighborhood that entrances to the neighborhood now have signs excluding non-local traffic from entering.

Waze still tries to navigate me through this neighborhood all the time. It makes the app nearly worthless for finding routes home. There are also some other streets I've learned not to trust when Waze tells me to take them home, as they often have much more traffic than predicted. I'd love to be able to block those as well.

Some roads were shut down for construction near where I work a few months ago. Uber and Lyft don't seem to take data like that into account, so I had a hell of a time getting home from work - I'd have two or three canceled rides before I got one that happened not to be routed through any closed roads.

I believe the entire use case you present was already acknowledged in the comment you are replying to, I. e. the “easy interface” to “exclude certain routes”.

The comment I was replying said that the functionality might be useful. This does not acknowledge any particular use case for the functionality, and leaves open the possibility that it might not be useful. Someone who isn't sure whether this functionality has legitimate uses or not probably isn't aware of what the legitimate uses are.

The parent is providing an alternative reason for avoiding particular neighborhoods.

Also as votes are hidden here me-toos provide useful info.

I work with many foreign women who have no idea the safe and dangerous areas of our city. You might consider from their point of view the value of such a service, rather than worrying about more meta-tribal political battles. Some areas of my city have higher harassment and crime rates, as a man this is less concerning for me, but it's something many women need to take much more seriously.

If you find people staying safe to be tremendously harmful to your politics it might be time to reconsider your viewpoints.

Well if you live in sleepy Luxembourg you'll probably not need that feature, but other countries might need it.


The ability to avoid an area is a useful and apolitical capability. I often want my satnav to exclude driving through cities as it's more relaxing and often faster to drive around, and when there are roadworks or large scale events it is really helpful to be able to exclude them.

Not only this is not recent, it has nothing to do with politics or ideology, at all.

No-go zones have been used all over the world for decades.

Not all over the world. I'm in Seattle and I'd probably have to drive to Chicago or Tijuana to find a place I wouldn't feel safe driving through.

I don't disagree about "no-go" areas, but since the comment to which you replied didn't mention those, and your own comment set off a sprawl of off-topic comments, I wish you hadn't written this here.

I believe the mention of Rolls-Royce, in association with “avoid area” may (legitimately) create the impression that this is a feature for pearl-clutching snobs afraid of “dark people” coming for their women and other belongings.

It’s perfectly okay to consider this and still say the feature’s benefits are overwhelming. But the rabiat outburst of denying even the possibility is an indictment of the posters more than the suggestion that triggered them.

Isochrone is the word you're looking for, for the distance time feature. Openrouteservice can do that, based on osm data (of course; what other source could they use really).



I would happily pay for an update that fixed outstanding bugs and ensured compatibility with future versions of macOS. It would be nice to have some new features, too, but it is basically "good enough" as it is and I dread the day I can no longer run it. I'm really not sure what I will do.

The claim that Photos was going to be a meaningful replacement for Aperture was obviously a lie at the time it was released and even nearly five years later it is still not true.

I'd have moved to Lightroom had Adobe not moved to a subscription model.

Oh man, so much this.

When Photos just came out and Apple was making the claims about how they've taken everything good from Aperture and stuck it into Photos I was pretty optimistic considering how solid of an app Aperture became over the years.

Now, few days down the road, pretty clear that Photos is just a marginally better iPhoto with the vast majority of features that were useful for professionals such as Stacks, never coming back.

The dumbification of professional apps to the lowest common denominator continues ...

Stacks, 2-up and 4-up comparison, the Loupe, star ratings, fast browsing of raw files... so much missing functionality. Lightroom IMO was never as good as Aperture, I really wish Apple would’ve spun it off instead of just letting it wither.

I'm not a pro and loved Aperture. It made my workflow so easy and it was such a fast program. I haven't found anything that lets me cruise through photos so fast and pare them down. Left/Right arrows, Z to look at details, X to flag as a reject made running through hundreds of photos a breeze.

I'm trying to find an alternative and, quite frankly, everything I've tried is not great. I ended up buying AfterShot Pro, but I'm not terribly satisfied with it. It runs on the Mac, but it's really hurt by the fact that it's not actually a Mac application.

Adobe is out because of the subscription model. I'm willing to buy the software, and I'm even willing to periodically buy new versions when it makes sense, but $10/month forever is hard to stomach.

Fast Raw Viewer. It has very few frills, but it lets me do selects way faster than even Aperture. Whether processing in Lightroom, photoshop, or Photos, I always run larger batches through this app so I can quickly toss rejects and pick a few selects.

Many years ago, when I was first getting into digital photography, I was faced with the Aperture-vs-Lightroom decision. Both seemed promising, with Aperture even having some advantages. What ultimately made me choose Lightroom, however, was the multi-platform support (I've always liked MacOSX laptops, but prefer to use Windows and/or Linux on the desktop).

Several months later, something happened that basically reaffirmed my decision. Nikon released a new camera, which I bought at launch. Lightroom added support for its RAW format in a minor patch almost immediately. Meanwhile, Aperture couldn't be bothered to add support until 6 months later and a major version update.

Flash-forward, I really wish Adobe didn't become so obsessed with their subscription models. While cloud storage/backup definitely has its advantages, I'd much rather simply own my software and use my own servers. (But since the "run it yourself" options are always grossly inferior, I just put up with their new approach and make sure I have a local copy.)

On the flip side, for some camera models Apple has added support more quickly. And while Photos is no Aperture, it does have pretty good RAw profiles and lens correction for almost all popular DSLRs, mirrorless cameras, and all the 1st-party lenses I’ve ever owned.

You could try Luminar 2018. Library management is coming, and it's developing tools are (IMO) better than lightroom. It's also very reasonably priced. That plus an Ubuntu vm for importing via shotwell is what I use.

> Library management is coming

Specifically, the date is December 18. There's also a (small) discount if you buy in the next 8 days: https://skylum.com/luminar

Not involved at all, just discovered this while looking around.

Iirc, you can buy Lightroom 6 perpetually, no creative cloud needed. Might be worth looking into.

That's what I still use but it's getting long in tooth and you're starting to notice lack of support for new formats and the like. I wouldn't really recommend someone who isn't already using it to switch. I don't really mind the subscription because it's software I use frequently but I need to get around to cleaning out my catalog before I switch over.

You can use Adobe DNG converter (free) or other 3rd party tools to convert new raw files to DNGs understood by Lightroom 6.

Is no one going to mention Pebble? It's the only smartwatch I've ever wanted, and I was so sad that it got bought and died right when I had enough money to spend some on a smart watch. It just had the perfect combo of good design, long battery life, and usability.

Pebble definitely deserves a mention. Lost mine this October on a high dive. Had it for three years never had an issue with it. I was feeling okay with it for awhile, but then I started missing meetings at work and time sensitive texts - was too used to having easy access to it my calendar and notifications.

Started looking for a comparable smartwatch: always on, good battery life, and Android notifications, there's just nothing there.

Ended up ordering a Pebble Time Steel from eBay. I feel like a part of my life and sanity was restored.

I found the same thing in an amazfit bip

Over two years after Pebble shut down I am still wearing my Pebble Time Round every day and it works just as well as the day I got it. There's really good community support with Rebble.

You can still find Pebbles for sale in various places and I'd recommend getting one over any of the Android Wear watches. They're just that much better.

Same here with my Pebble Time Steel. It still gets 7 days battery life, works with the latest Android, and the weather & speech services all still work after subscribing to Rebble.

I keep looking at the Fitbit Versa, which is clearly Pebble's spiritual successor, but it doesn't have an always-on screen or the Time Steel's 7 - 10 day battery life. I just hope Fitbit keeps the Versa around by the time I need to replace my Pebble.

My Pebble Time Steel has been in my drawer for 2 years I think. I bought it on release and didn't really use it for that long because I honestly didn't found it useful. What do you use it for? Most of the time I just find it easier to take out my phone.

Bluetooth Smart Unlock makes taking out my phone easier, since I don't have to put in a code or fiddle with the fingerprint reader. I've really gotten used to it.

But the killer feature to me is notifications. I can keep my phone completely silent with no vibration and never miss an important call or text or calendar reminder. I never have to pat my pocket to see if my phone is vibrating. I can decline spam calls and decide whether to respond to messages without taking my phone out. I can even leave my phone in silent mode on the counter across the room and not miss calls or texts.

I also love the Google Maps integration (third party app). Being able to see my next turn on my wrist is very useful and the navigation screen is more informative and easier to read than the native Google Maps Android Wear integration. When I get in my car that doesn't have Android Auto, I can still say "OK Google, navigate to X" and without taking my phone out of my pocket I get voice directions through car Bluetooth audio and visual directions on my wrist.

It's also just a good watch. There are lots of faces to choose from, some with extra information like weather or stock prices or step count or whatever. It's far thinner than any other smartwatch and you don't have to flick your wrist just to turn on the screen and see the time.

I use it for as few things as possible and it does them very well. I only want notifications for things that may be immediately relevant (phone call, text msg, calendar event reminders) so when I walk away from my phone I'm not worried about missing something important. It works with Smart Lock on Android so my phone stays unlocked when I'm close and locked when I'm not. Vibrating alarm clock that won't wake the whole house, sleep & exercise tracker without needing to carry my phone. I don't actually "use" it very often, but it is a very passive yet essential part of my routine and keeping myself on track. I am very forgetful and used to have entire weekends pass by where I forget to check my phone unless something reminds me to and the Pebble helps make sure I get only the essential information when I need it without unnecessary interruption. It also shows time and weather at a glance ;)

I have one sitting in my drawer; happy to send to you for free if you're interested in owning one. Just drop me a line.

Hey dude I'd be really happy to snag that from you, but I think it'd only be fair if I paid for shipping.

I couldn't find an email address for you, mine is HANEEF at PDX dot EDU. I've never actually had a smart watch before, but hearing all these people talking about the pebble is making me really interested to try this one.

I too have a Pebble that I'm happy to provide to anyone as well. OG Kickstarter edition. Free to a good home.

if the offer is still available, I'll take it. I'll pay shipping (I'm in europe, assume you are in the US).

Try the Amazfit BIP, it’s similar (e ink screen), cheaper ($69) but does lack a bit on software / integrations side

Does not look comparable. Appears to be basically a heart rate and step counter. One of the great things about smart watches is notifications.

As far as I can find on the internet, it does support notifications, even says so on their official site. So not sure what your comment is referring to.

Ah yes. The pictures I found made it look like it didn't have a proper display but it actually does. Definitely not e-ink though.

I believe it's transflective LCD display

It is e ink... the screen is always on, works in bright light and lasts 30 days+

E-ink is a specific technology. Not just any daylight-visible low power display.

It has notifications, shows texts, let’s you reject calls etc.

although its no pebble, the amazfit bip is the closest contender available today

What does data security/privacy look like?

I used to be big into those dorky digital watches when I was growing up, and I've been tempted by fitness trackers. My concerns are:

A) Is the Bluetooth connection secure? Are they taking steps to make it so the Bluetooth connection isn't easily trackable? How easy is it, for instance, for me to set rules to turn off the Bluetooth when the watch isn't actively syncing?

B) Does the watch send any data to someone else's remote servers, at all? Do I need to go through a 3rd party to get access to my own data (ie. what Fitbit does)?

I already have a Smartphone, I don't want my watch to act like one. I don't want to have to do a bunch of syncing through a 3rd-party server that will get shut down and sold off with all of my data someday. I'd pay a lot of money for a device that uses my Smartphone as an access-point/hub, but that largely just works offline and independently.

Holy crap, I've been looking for something to replace my pebble... This product can't be real... 30 day battery life?!

Me and my girlfriend been using the Bip for a month now.

We run every day.

The battery life is pretty phenomenal, the step counter and weather bits are handy, the GPS just works, you need the AmazTools app if you want to share your runs to Strava (but it works great), I have an alarm set to shake my wrist every morning at 7am, and if I destroy the watch.. it was all of $70.

The continuous heart rate monitoring is also handy for our gym and zone workouts. I wish it had multiple levels of heart zone alarms, like the TICKR FIT app -- which is amazing for more refined zone training, but it will tell you when you pop out of zone 2 on a run, so it's still useful.

Sleep tracking works great for my girlfriend, doesn't work great for me. So I turned it off and that doesn't bother me.

I think it's a great starter watch, I can see upgrading to the Garmen vivoactive 3 Music, since I'm not very interested in an Apple Watch.

I just preordered the Amazfit Verge. I've tried smart watches in the past, and I never could find any use for them, as I'm the type of person who silences all notifications, but now I play tennis and maybe it'll be useful there. It was mostly just an impulse purchase, but I hope I don't regret it.

It’s not quite 30 days for me, more like a couple of weeks, but I have everything enabled.

It genuinely is really great, if it had an app ecosystem, slightly higher resolution and was a little smaller / more stylish it would be perfect. A FOSS OS and support for emojis would also be nice

I have two Pebble/Rebble watches that I still wear. I love them!

Microsoft Money. I didn't have to give access to all my accounts to some website, or pay an annual subscription for upgrades. I got keyboard shortcuts and reports and I could search for things to find out how old my computer is or when I went to Vegas.

It seems like in the physical economy, if you create something of value and you aren't making money off it, you will sell it and someone else can make money off it. With software like Picasa, it's just gone. Some programs release a sunset version, but you never know how long it will work and you can't promote it because they might take it down.

The interface may not be the same, but I highly recommend http://plaintextaccounting.org. I've used ledger and currently am using beancount. Both are great!

There's also gnucash, never used but its interface may be similar to Microsoft Money (which I also never used).

I've found Beancount [1] to be a nicer version of Ledger. It is similar syntax but it enforces rigid rules to ensure you don't make as many mistakes. You can also use Fava [2] as a nice local web-based account browser.

[1] http://furius.ca/beancount/

[2] https://beancount.github.io/fava/

I’m a tortured Gnucash user. The best I can say of it is that it’s free, and it works.

I would not recommend anyone start using it if they are not already.

I use MS Money (2004 edition) in a Windows 2K VM. Unless I write my own finance software, I don't believe I'll ever stop.

You should be able to use "Money Plus Sunset Deluxe" (https://www.microsoft.com/en-us/download/details.aspx?id=207...) which works for me on Windows 10.

My Dad used Money 99 for years until he wasn't able to run it any longer. He switched to gnucash and seems to like it well enough.

In the case of Money, they did release a sunset version (https://www.microsoft.com/en-us/download/details.aspx?id=207...) that continues to work...

I know it's not going to be popular, but mint.com has far more features and is a much better package all together. Yes, you need to give your bank login info.

I use AceMoney for this, and it's still great after more than a decade of use in our businesses. Great support. Highly recommended.

Money dance is the closest I found that will download from financial institutions and auto import. I switched when my banks stopped supporting my Msmoney vm.

Since 2007 I have all my expenses (every single item, drugstore, stocks, funds, etc.) in KMyMoney. It works very well.

yes!!!!!!! hate hate hate quicken with a burning passion.

The parent comment immediately reminded how I loved the ux and flow of "quicken home and business 98" - I have hated every quickbooks version since.

Didn't inuit buy and kill / sell quicken or something. If they would of just kept the ux wrapper from Qhnb 98 and layered that over the DB of quickbooks it'd be okey with me.

I too hate hate hate quickbooks, and price structure, layout and all that.

I have a few software packages in mind:

1. The Genera operating system for Symbolics Lisp machines. Ever since I've gotten bitten by the Smalltalk and Lisp machine bugs, I've been wanting to use Genera, but I was born at the beginning of the last AI winter, and unfortunately the operating system is still proprietary, with no word about whether or not it will become open sourced. It's regrettable that the proverbial baby was thrown out with the bathwater when Lisp machines lost out in the marketplace to RISC workstations and the x86 running Lisp on other OSes; there are a lot of interesting lessons we can learn from Genera in today's operating systems and programming environments, and an entire generation of computer scientists and software engineers are unfortunately completely unfamiliar with Lisp machines.

2. I've heard wonderful things about the productivity software developed by Lighthouse Design for NeXTSTEP and OPENSTEP. I haven't used them myself, but I've read that many NeXTSTEP users were devoted fans of Lighthouse Design software, and the presentation tool that the company designed influenced the design of Apple's Keynote.app. Unfortunately when Sun acquired Lighthouse Design, Sun hasn't done much with the Lighthouse Design codebase. It would have been nice had these programs been updated and further refined for Mac OS X.

3. I would love for someone to resurrect the ideas of Apple's OpenDoc platform, which allows for component-based development of GUI applications, much like how pipes and redirection in Unix allows for different command-line utilities to be used together.

My dream is an operating system where all objects can be inspected (like Smalltalk or a Lisp machine OS), where there is a component framework similar to OpenDoc that allows for component-based GUI application development, and where there's a REPL that allows power users to have complete control over their applications (both command-line and GUI). All applications would adhere to some type of well-designed usability guideline (I'm thinking about the classic Apple Human Interface Guidelines from the pre-OS X days), and the interface would be reminiscent of Mac OS 8.

WRT OpenDoc stuff, you are actually describing COM and ActiveX. COM basically defines an ABI and memory layout for different languages to use to access objects and ActiveX (aka OCX/OLE control) uses it to provide reusable components (mostly visual) that are self-describing so you can access their properties and call their methods without having prior knowledge to them. It was mainly used with classic Visual Basic, but also other environments supported it, like Delphi and today Lazarus also has support for it.

Its use on the open Internet was a nobrainer from an implementation perspective, although it killed the reputation of the entire thing (make a Google search for ActiveX and 99% of the content you'll find is about its use on Internet Explorer) and .NET cemented its demise.

But yeah, the idea is solid and actually other platforms had similar tech, like BeOS's replicants, AmigaOS's... something (i don't remember the name :-P) and i think KDE's KParts was originally supposed to be something like ActiveX.

I share your sentiments. With regards to Genera, have you seen this?


This runs off of an unofficial emulator that is a monstrous hack. There are lots of gotchas, bugs and so on.

On the other hand, an official emulator is in the works and should be released at some point.

Wait, what do you mean? The current owner of Symbolics authorized the development of a new I-Machine emulator? Who is working on it, Kalman Reti and Brad Parker? This is very interesting news, please elaborate.

I don’t think Brad Parker is involved. You can email Reti for more information.

What is his current email address? The one at symbolics-dks.com?

Number three sounds really cool. Command line interfaces is what I eat and breathe all day because there's no replacement for it, and I've often thought how that is stuck in time. I'm not sure what the proper solution would be, otherwise I'd have made it, but it sounds like this might be worth a shot.

I really like your dream. There are probably people on this forum with the right skills to make it happen if you want to pursue it in earnest.

OpenDoc is unnecessary on Lisp machines, in Genera you can already build applications by combining any CLOS classes that exist in the system. There is no rigid "application" black box that needs to be broken down into OpenDoc components, everything is an object that calls methods of other objects.

I always see interest in these sort of operating systems. Are there any serious attempts at creating one?

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