You know, the people who talk about USENET and IRC are people who don't want to be "updated" to what modern users demand. I don't know if you've had a look around you lately, but the modern internet is a dumpster fire.
Anyways, ideally you would be able to connect to services with whichever client you want. You could use a pretty eye candy client, or then not. With the web, however, it's an impossible feat to implement a standards compliant browser.
> the people who talk about USENET and IRC are people who don't want to be "updated" to what modern users demand.
IRC is objectively terrible. If you ask a question, the only way to get the answer is if you stay logged in until someone answers you. That's like buying a phone that doesn't ring unless you're already holding it up to your ear, or an email client that deletes all your email unless you happen to be looking at your inbox at the moment a new message arrives.
I disagree with your false equivalency. IRC is akin to a cellphone without voicemail, as when your disconnected from the network (cellular or IRC) you just don't get communications.
Scrollback and voicemail can be useful, but a good chunk of the population does not want either, hence unread backscroll and unconfigured/full voicemail boxes being very common.
>but a good chunk of the population does not want either
The largest Freenode channels I'm on like #node.js and #javascript have about 10 regulars max, myself included. It's pretty much dead for any purpose other than hanging out with a handful of curmudgeons feeding off a corpse out of habit.
Doesn't seem like "a good chunk" to me. IRC lacks the only feature I really care about in a chat network: new people to talk to, users.
I wonder how many HNers who praise IRC only like the idea of it but don't even use it these days, either.
And I haven't even heard HNers bring up usenet outside of downloading free shit so I know that's especially dead wrt this thread as a decentralized Reddit.
So it always makes me chuckle to see "yeah, well, some people like it like that!" which is apparently virtually nobody. Or it's like how my two friends and I loved the failed, abandoned supermall in my hometown because the three of us could longboard down its parking garage.
What makes IRC good is that it's a text-only chat with a wide range of highly configurable, blazing fast clients. Not supporting images, quoting or reactions could be considered a feature, since these distract from the actual conversation at hand (consider newbies clogging the conversation with screenshots, or 100 people adding reaction emojis to a message).
Young people now think it’s ok to send out a message and respond whenever they feel like it. Communication is no longer a conversation in the tradition sense, it’s a series of monologues or statements where the sender usually only cares about whether you are impressed by the statement.
IRC was all about conversation. You went there to have a conversation, like a party line voice bridge. You even waited around for someone to have a conversation with... oh I miss it...
This is like the famous comment in the "Show HN: Dropbox" thread that said you could just use rsync/ftp/git instead. Meanwhile Dropbox is now worth over $8B.
Not sure why the value of anything makes my point invalid? I answered there was a simple way for anyone using IRC to keep sessions alive and therefore not miss any message. This is an off the shelf solution, not something you have to build by yourself.
Synchronous and asynchronous communication both have their fundamental strengths and weaknesses. In a chat room you have the possibility of a real-time back and forth. That's invaluable at times.
You might have a preference for one over another, or each might be appropriate for different situations. But to claim one is objectively terrible is ridiculous and only makes you look like you haven't taken the time to understand it at its most basic level, or you're too solipsistic to be able to recognize the existence of value in something outside of your own preference.
Slack is a superset of IRC, except for the protocol not being open.
The UX in slack is not easily replicated in IRC (for example, how does/can global search work in IRC?)
Therefore, it is actually correct to say IRC is objectively worse in those measures. The only measure that IRC beats slack is the open protocol (which, if I'm being honest, not many users care,ala most existing chat services moved to private protocols and haven't lost all their users to IRC or xmpp).
> IRC is objectively terrible. If you ask a question, the only way to get the answer is if you stay logged in until someone answers you. That's like buying a phone that doesn't ring unless you're already holding it up to your ear, or an email client that deletes all your email unless you happen to be looking at your inbox at the moment a new message arrives.
Well it's a good thing Slack and Discord don't work exactly the same way but with a nicer UI and no need for a bouncer to emulate a persistent session.
These two words next to each other really bother me. That something is "terrible" is not objective, whatever that means. All we have are our subjective valuations.
What I meant was more like, when viewed impartially rather than through the lens of nostalgia. Two people can do that and still have different opinions.
Your use case is what mailing lists or usenet are for, of course IRC is an objectively terrible mailing list, it's also an objectively terrible toaster.
However, the spam mostly died off now. It's possible to have great convos with boomers complaining about each other finally. A small number of people came back because of the pseudoanonymous qualities. Honestly the only thing that really sucks is Google Groups users bumping 15 year old threads.
Oh and binaries didn't go away, that's a feature and you just don't have to sync alt.bin.* if you want to avoid. I think it's great though for snatching rare/out of print stuff.
The spam died off because Usenet died off. If users ever went back to Usenet, the spam would come back again behind them. There's nothing in the design of Usenet that can stop spammers from spamming.
(There wasn't anything back in the '90s in the design of email to stop spammers from spamming, either; but enough people cared enough about email that billions were spent to add that spam-proofing. Nobody cared enough about Usenet to invest a similar amount, so it's still as unprotected as it was when Monica Lewinsky was in the news.)
Modern internet is both good and bad. Depends on the site. Generalizations don't work. However if those technologies are to ever be popular than they'll need to be redesigned to keep pace with what the vast majority expect, which is in response to the OPs comment. Otherwise you can already go post on usenet right now.
> "Anyways, ideally you would be able to connect to services with whichever client you want."
Make it happen. This is my other point. People complain that all these fantastic legacy technologies aren't used but nobody puts in the work to modernize and use them. If it was that simple, where are all the companies doing it?
Is it companies, that would or should do it? Colleges and other publicly-funded entities are who did it the first time. Companies can only do what's profitable. (Or what they can convince investors will someday be profitable, but we can ignore that aberration in the long term.)
I didn't mean it that deeply, just that if people who believe these older technologies can drive modern incarnations then I would expect them to have produced products by now.
I usually adjust my brakes so that my front brake isn't tight enough to throw me over the handlebars in panic. I used to bike a lot (and quite recklessly) in my teens, so I got kinda paranoid about overly sensitive front brakes.
Anyways, and without knowing the details of your accident, if you were biking 30km/h in an area where you can't see if someone comes from around a corner, then you were biking too fast. Still the girl's fault for jumping in front of you without looking.
So is this just a VPN that's baked into the browser?
One of the problems with VPNs is that you're putting a lot of faith in your VPN provider. I trust Mozilla, and would gladly pay them for that service (I would use Freedome, but afaik they don't support Linux). However, it would have to be usable outside of my browser.
Tangentually, I wish Mozilla also offered paid email (or email with a premium plan), which is another service that requires a lot of trust. It would help provide alternative sources of income to keep Firefox alive, and Thunderbird could be a stunningly good email client if they had more resources to pour into it.
Mozilla's product is trust and control. Although they are non-profit, I see no issue with them offering paid services.
As much as I like the idea of baking better privacy tools into the browser, it's hard for me to get enthusiastic about the idea of making Cloudflare even more of an official man-in-the-middle for all network traffic than they already are.
A better question that we should be asking is, how the hell did we get to the point where we need a third party proprietary platform to serve a static file efficiently? I remember a time when mainframes would automatically place orders for their own parts when they broke down and when personal computers empowered people to easily create and remix. Somewhere between then and now we forgot how to make things simple and easy to use. Somehow despite the advances of HTTP/2, WebRTC, the upcoming WebTransport, web hosting is now harder than ever even though things are supposed to be more efficient. Apache and NGINX are far from accessible to your average user. Countless sites depend on proprietary "as a service" oligopolies like Cloudflare and Netlify. Hosting an email server these days are almost an exercise in frustration; what happened to the mythical unikernel? Where is my secure, turnkey email server image? Unikernels were supposed to make ops easier and things more secure. Somehow they never showed up despite all the hype on HN. Zero config self hosting projects like Sandstorm are half dead. It's easy to complain about tech giants, but we are not exactly providing end user alternatives. The world does not need yet another Lisp interpreter, the world needs high quality zero maintenance software that is easy and accessible.
We are so much better at adding complexity than removing it. There are tons of incentives that drive that: the difficulty of upgrading old stuff, the need for companies to invent reasons to exist, bureaucracy, job safety and creating reasons for employment (the personal version of corporate self promotion), featurism and comparison on features, etc. There are almost no incentives pointing the other way.
This is all a sign of growth, most of it for the better.
There are now billions of people accessing the web, so sometimes a web site needs the resources of a company like Cloudflare to handle traffic spikes.
Decentralized email has been a victim of its own success: because there is no central email authority, spammers and bots can easily flood email boxes. If you don't mind the spam, it's actually not hard at all to set up an email server, but most people hate spam, so most people don't want to set up an email server. There is no pure technological solution to spam, so we fall back on companies to help manage it.
Thanks for the reminder about Sandstorm. I intend to try it out sometime. I hope it's not dying.
Sandstorm kind of still there. They discontinued the free tier for their hosted platform because they ran out of money. The founders went to work elsewhere but maintained the project on the side last time I checked.
I think it’s a shame, it’s a lovely concept. The Capability-based security alone is game-changing.
It's not a sign of growth. Among other things, it's a sign that we have grown complacent about complexity and are not doing our job of keeping it under control.
I am still not sure how much one can trust Cloudflare as an entitiy. At some point people started putting loads of stuff behind Cloudflare, enabling them to be the perfect MITM, which is concerning. Probably only a question of time until some profit seeking people come around and see opportunity in it and then we are screwed even more, than we are already with Google captchas. Then we will not be able to use many more websites any longer, because someone in their incredible wisdom decided to put everything behind Cloudflare. Scary.
You could say the same thing about any CDN - Cloudfront, Fastly, Akamai, etc.
Is it right to say that the only reason Cloudflare is the forefront of this concern is because of their business model of offering the CDN for free, while the others have a much more limited free tier or service or none at all?
I am almost equally worried about other very popular CDNs. However, being the most used CDN makes some people short-circuit and not think about the dangers any longer Just like with captchas from Google. Many people simply put them onto their websites without ever thinking about that, "because everyone does it". If so many people put stuff behind Cloudflare, then at some point the same kind of people, who put captchas without thinking, will put stuff behind Cloudflare without thinking.
Scripts and other stuff from first party usually seem to me at least more trustworthy than something from a third party. It also saves me the mental step of thinking: "Hmmm, why are there scripts loaded from a third party? Is this some kind of ads stuff?"
If a website does not work without unblocking third party scripts, there is some chance, that I will simply abandon it. When a website's purpose is to inform me about something and I do not see the need for any interactivity, I might also abandon it, if it does not show content without unblocking scripts in general, including first party. Web frameworks, which do not take care of at least presenting something when scripts are not unblocked, thus make a website less trustworthy for me.
If you care about privacy then you SHOULD say the same thing about any CDN. Sucks that there's all this awesome infrastructure that we can't use anymore, but that's the trade off.
If you don't know who you can trust to provide a trustworthy proxy service, then there's a lot to be said for choosing a provider who can already MITM a good chunk of your traffic even before you turn on their proxy.
As I see it, "[i]f you don't know who you can trust to provide a trustworthy proxy service", you distribute trust among multiple providers, such that they must collaborate to pwn you. That's the basis of Tor. And you can do something similar, albeit far weaker, by using nested chains of VPN services.
Sadly, this is exactly what's being marketed as 'VPN' for about 5-10 years. Not sure who started this and whether we can ever correct this misuse of the term.
Not really. It's what's marketed as "VPN service".
And a key point of VPN services is that they don't share the ISP-assigned IP addresses of their users with anyone else. And they don't retain any logs, which an adversary could obtain in one way or another. Indeed, the best ones run totally in RAM, and don't have writable storage.
Edit: Also, using "VPN" in that context is not at all deceptive. Enterprise VPNs not only provide secure access to private resources. They also provide secure (and yes, often controlled) access to the Internet generally. And that's just what VPN services do. Except, mostly, for the "controlled" aspect. Although some VPN services do feature blocking of malicious sites, malware downloads, etc. Some even block age-inappropriate content.
Not going to happen overnight and even then technically they cannot access your emails because only you hold the password to private keys (if you trust they encrypt your emails with your public key before storing). I prefer keeping my emails local so pop does the job.
Google can access your emails but something like protonmail can't (if you trust them to encrypt your emails).
Sounds good... in theory. Truth is we don't know this, their mail client is proprietary and even if it was open source we still wouldn't know what their servers are logging. I am in a similar situation and I trust Google more than ProtonMail with all that Tesonet data weirdness pointing back to one guy. These shell companies can go and disappear overnight, something to consider.
Trust is multiple things. I "trust" that Google will manage my private Gmail account consistent with my expectations—which are effectiveness and long-term durability. I have no expectation of algorithmic privacy and thus there's no trust to break there.
3rd option. Running your own for incoming, using a service for outgoing. You get all the trust benefits of running your own server without having to worry about IP blacklists or any other crap. The nicer ones even generate your SPF, DMARC entries you add to your DNS.
I like also that it is no problem to use in countries which block VPN, like China. I had no problems accessing Posteo, but no way of accessing Gmail.
It also does not nag me every single time I change the VPN server I use, because I seem to be in a different location. I know this is supposed to be a security feature of Gmail, but man is it annoying not to be able to access your mail, because of that.
In Posteo you can also activate 2FA afaik, if you like such thing.
It just works, and I am glad to be able to give support to free software, while at the same time I also gain from it, by having an e-mail service, which is ethically way more acceptable than Gmail and is working very well.
They can just forward the encrypted bytes (I don't know if that's how they actually do it). Cloudflare does mitm decryption in order to inject its own code and captcha in the pages.
> Although they are non-profit, I see no issue with them offering paid services.
I believe that's the goal. They're introducing a lot of services in the relatively short time span: Firefox Notes, Lockwise (password manager), Send (file-sharing service), Screenshots (image sharing service), and now a VPN.
After all these projects mature a bit, they're probably gonna slap a subscription on top and offer Firefox for free, Firefox + services for a price.
I'm personally totally fine with that way of diversifying their income. I already donate to Mozilla somewhat regularly, so the only thing that might prevent me from taking that offer is a steep price.
EDIT: Technically speaking, Pocket is kind of a bookmarking service with a built-in premium option already, but they're really, really not giving it love. "login with Google" button still has the lowercase "g" icon. That's like a two-minute fix that hasn't happened since 2015.
i hope they get the pricing right, which by my estimation is probably 2-tier pricing (on top of the free browser). an inexpensive base tier, maybe $3/month, and a premium tier at ~$10/month.
diversity of donors/customers has long-term benefits (better market insights, for example), so that lower tier is important to attract the less affluent. the higher tier is for techies like us who can better afford to support a more free and open internet.
> Tangentually, I wish Mozilla also offered paid email hot (or email with a premium plan), which is another service that requires a lot of trust. It would help provide alternative sources of income to keep Firefox alive.
1. If Mozilla email is used to support Firefox in part, then it will be overpriced. I am not sure if many would like to pay more than a token amount for charity.
2. Generally, software companies have increased tendency and incentive to horizontally expand because of the low barriers of entry. These incentives and tendencies are exactly what led to Microsoft and Google become these monopolies that exploit other players in the market to their advantage. I would rather Mozilla remain small, neither vertically integrated (like that attempt at FirfoxOS), or horizontally integrate (like offering email on top of a browser). I say this as someone who uses Thunderbird as their primary email client. The bazaar model of software development protects user freedoms more than the Cathedral model.
Firefox is successfully funded by Google through selling default search options. Unless Firefox's revenue were to suddenly and dramatically drop I don't see why they'd need to increase new product prices to fund an already funded project.
And Google, of course, with a 50 state anti-trust probe opening against them, would probably fund Firefox for free just to desperately promote the idea of competition. I'm reminded about how Microsoft funded Apple during the MS monopoly era as a "see, we have competition" response to investigators.
Totally agree on email. I've long been saying that they should just acquire/partner with FastMail, seems like a great fit and would allow Mozilla to better align their revenue stream with their users.
Also, this constant killing and reviving of the Test Pilot brand looks incredibly short-sighted and stupid to me.
I love both Firefox and FastMail, but I'd never want one to acquire the other. Vertical integration introduces lock-in. Even if it's not intentional, it's far too easy for developers to introduce "conveniences" that only work if you're fully adopting their ecosystem.
