These Web elegies are starting to become a real drag to read. And I don't mean to come across as flippant in saying this. I feel a sort of concern for the spirit of the authors of these kind of pieces, which are becoming more and more commonplace.
Agreed. I don’t even know what this elegy is for. “what was wonderful about the web, was that here was a machine for finding people with similar interests and experience, anywhere in the world. That’s the web I want back.” Is that no longer the web? It seems pretty easy to find communities these days.
The web I want back is the one powered by real people making real things. But rather than complain about it I think I’d rather be one of the people who is still here, making things.
From what I've seen, a lot - I'd say most - of the communities that existed outside of the major platforms have dried up or ended (the famous/infamous bodybuilding forum apparently just went down). Web forum communities, mailing list communities, communities built around an individuals website (it sounds strange to say now, but this was pretty common a couple of decades ago). Even the offline meeting sites like Couchsurfing and Meetup are a shadow of what they were in their heyday (about 15 years or so ago).
The actual number of people actively using the internet is much higher, yet the number of people outside of a few small social media silos are much lower. So I can only surmise that these social media silos are giving the vast majority - almost everyone - what they want (those gratifying instant hits). It's just sad to know that there's apparently only a tiny sliver of humanity interested in venturing off the main path and interacting with people outside of these silos.
It kind of is though. Maybe not with discord specifically, because it’s a closed platform. but communities on open ones like matrix are just as much part of the web as any forum.
It’s not a big platform with all users in one space. There are dedicated communities with passionate people talking, sharing ideas, helping each other out.
Would you’ve considered IRC chat rooms on freenode part of the web?
> but communities on open ones like matrix are just as much part of the web as any forum. [...] Would you’ve considered IRC chat rooms on freenode part of the web?
For IRC, absolutely not. IRC (1988) even predates the web (1989). Freenode offered a web-based gateway that ran on their servers, that doesn't make IRC a web protocol.
As for the rest it's slightly fuzzier since in practice they rely on HTTP, but you can't just use a normal web browser to interface with them. You have to run a specialized client that understands the protocol to interact with them. The practicality of the firstparty clients for these platforms/protocols (or in Discord's case the only one you're officially allowed to use) being shipped in a web browser doesn't make the systems "part of the web". The server doesn't give you HTML that you can see in any browser, you make API requests for JSON in a non-W3C-standardized format and interpret it with specialized software. You can't send someone a URL to a message and have them be able to open it in any web browser without additional software.
I am well aware that it's possible to have a link to a discord/matrix message, but it only works if you're already a member of the group and you download the custom software client.
> You may be pleased to hear that those kinds of communities are actually alive and well, they just live on places like discord now.
But that is exactly what it wasn't. Discord is a proprietary walled-garden like too many others today. I tried to access a forum there once and it demanded both registration and a phone number. No thanks. That's not the open Internet.
The open communities were defined by the openness. Using standard protocols like SMTP, NNTP, HTTP (on public websites, not walled gardens).
> From what I've seen, a lot - I'd say most - of the communities that existed outside of the major platforms have dried up or ended (the famous/infamous bodybuilding forum apparently just went down). Web forum communities, mailing list communities, communities built around an individuals website (it sounds strange to say now, but this was pretty common a couple of decades ago).
I would say that is simply because bodybuilding sucks and everyone with half a brain has moved to other activities like crossfit or similar that allows you to have a nice body that you can actually do something with instead of a useless one :)
More seriously this is not my experience. Two of the websites I visit the most in 2024 are the "traditionnal participative type", ie, privately owned, not social medias but with membership and comments and stuff. One is using a custom cms based on rails and has been launched in 1998, the other one is only 15y old but is running vbulletin.
And I still encounter many websites like this. Last time I had an issue with my Piaggio motorbike, I found a vespa dedicated web forum that was really helpful and active.
I think there are lots of people deep in social medias, reddit and discord and some who are fed up by this stuff and stay out of it. The issue is that privately owned websites cost money to run usually or are very stripped of customization/features/storage and moderation is a burden so you need very motivated people to keep the light on compared to say, a discord or reddit community that can stay online for months/years while the initial admins/founders aren't even connecting to it anymore.
I'm looking forward to reading an aging GenZ-er's elegy of the web that looks fondly back on the age of TikTok and algorithmic feeds, back when people actually made things.
"I miss when the internet was an international gathering place, not a carefully guarded forum for communicating misinformation-free messaging about foreign policy and corporate life in the United States of America, England and Wales. I had a friend from China."
>It seems pretty easy to find communities these days.
Outside of HN and a few niche subreddits, I can't think of many. Most good communities I'm part of are off the web on things like Discord which isn't addressable by search engines.
There has to be a barrier, a kept gate. Back "then", simply access to the internet was enough of a filter to get a good signal-to-noise ratio.
Today it's something else, such as knowing discords through word of mouth, knowing groups on other platforms, etc. If you're not getting through this knowledge filter, you need to try harder or accept not being in there.
HN can remain public and high quality due to strong moderation and an "ugly", text-only, imageless, emojiless UI that scares away "noise"-people.
This comment was very well-written. And, it stirs up thoughts about what works and what doesn't. "What abouts" like that one battle royale game (that failed) that coupled advancement to being a good sport and respectful to others, and it kindof worked. Could that kind of "good sport" thing be engineered into public forums somehow.
But in the end, all my feeble "what if's" collapse, and I end up agreeing with you.
Discord is honestly one of the worst things to happen to the internet, and I'd blame Gen-Z for most of its popularity since they seem to have no idea why the internet was created in the first place.
I think this sounds accurate. And, on the surface, there's nothing wrong with a modernized, glorified IRC network, but it should be obvious to anyone here that this is utterly useless as a store of useful information that can be archived and indexed and searched by others later. It's like two guys sitting next to each other at a bar and striking up a conversation, and one tells the other a great tip he figured out for how to fix a very particular issue with his car. That's nice, but the other guy just got lucky he met someone with this knowledge, and all the other people with that same issue are out of luck and won't learn this valuable nugget of info because there's no record of it anywhere.
IRC wasn't run by a proprietary company that demanded your phone number to read the channels. So discord is basically the exact opposite of IRC (and everything else in the open Internet).
>AIM and forums coexisted because they serve different purposes.
That's the problem with Discord. As an AIM replacement, it's fine, because no one cares about being able to access other peoples' ephemeral private chats. But when large communities discuss things, it's very useful for that information to be publicly available and accessible from a search engine, and Discord prevents that, locking it all away from any outsiders. If your group really wants to be secret and private for some reason, I guess that's fine, but if you're a group discussing, for instance, bicycle maintenance tips, why do you want to keep others from seeing this helpful info? And why do you want to keep others out? Specialty subreddits don't have this problem.
So where do you suggest people post that is zero cost and advertisement free?
These communities were pushed onto discord because there isn't a better place for them.
The regional live music discord I'm in would be much better served by being on a web forum but running a forum costs money. Or they could use facebook but not everyone in the community has/wants a facebook account. Not everyone uses reddit/twitter/insert your favorite social media.
> why do you want to keep others from seeing this helpful info?
How paranoid would one have to be to think people are moving to discord to be malicious? The garden doesn't have walls to keep you out, the walls are to keep the advertisers/bots out.
>So where do you suggest people post that is zero cost and advertisement free?
These days, Reddit seems to be the best option available (unfortunately).
Back during better days, we had lots of independent forums running software like PhpBB, which served this purpose well. For some reason that I don't really understand, back then (when computers were less powerful and efficient than now) that worked well enough, but now everyone's screaming about hosting costs and how expensive it is to run even a single bulletin board on the internet so they turn to Discord when they don't like Reddit. Why hosting costs apparently weren't such a big deal 15-20 years ago with crappier hardware and much less storage space, I honestly don't know.
I said 0 cost and ad free but your answer does not satisfy those requirements, because your interest are at odds with the interests of these various communities.
Communities generally don't need their content to be indexable. People in the community can use Discords search feature. Indexable content is only a benefit to people outside the community.
So why should these communities move away from discord to help you? Especially given that you do not respect their interests.
Do you get what I'm saying?
> Why hosting costs apparently weren't such a big deal 15-20 years ago?
It's not that hosting costs are so much higher now. In the 2020s I ran a forum for ~10$ a year but then I realized exposing my home IP was a bad idea. Then I ran the forum for $50 a year and... no one but me used it.
Recently one of my friends paid $500 to set up a forum using a dedicated provider /w a support plan. Fingers crossed it sees more activity than mine did.
So you can still host a forum pretty cheap but it requires trading in your personal time or paying $500 and given most Americans don't have $500 liquid you would need donations/organization (which requires more of your personal time).
It's not impossible but it's not super easy either.
You're just repeatedly asking that these communities deal with advertising to make things easier for you personally because you think they need your participation to survive.
But what if they don't need you?
P.s. not every community recruits primarily through the internet.
> Why hosting costs apparently weren't such a big deal 15-20 years ago with crappier hardware and much less storage space, I honestly don't know.
Back in the so called "golden days", I actually hosted a forum with a friend. His uncle offered his old office PC(bought at a bargain) and internet with fixed IP for us to tinker with. Our tiny forum has local kids and some "supervising adults" doing moderation. Many of our international members were mostly on dialup, so even if we didn't have lovely CDN, the slowness was accepted as normal thing. We did not face much spamming, grifting and any trouble makers were policed out immediately by moderators.
In present days, I am very afraid to put something online, because if the moderators are not vigilant, I could very well expect a police letter by a week or two asking about certain type fo contents and my infra provider banning or null routing my forum. It was easy back in days, now the risks are too high and trouble makers have more resources than well intent folks.
It is starting to change, for me, Discord has being going downhill for a couple of years.
Client updates are becoming more common and more buggy, as if we were on a beta channel. Marketing for Nitro is becoming more agressive, and now, there are "Quests", which are essentially ad banners for games.
100%. I've been writing HTML since 94, and there truly was no golden age. There have been (get ready for this) good and bad things about each era.
It has always been possible to create communities, or break away and do something different. I really like ActivityPub, but it does not really change the dynamic.
With all of that said, sometimes getting a fancy new notebook makes you more likely to draw/write, which seems to be essentially what Dave is experiencing. Nothing wrong with that!
I still remember the screen names of folks I interacted with in and on IRC, AIM, ezboard, IGN boards, forums, etc.
I only remember a handful of HN usernames and absolutely zero handles from elsewhere.
Forums used to have large avatars and signatures you could customize and communities were much more tightly-knit. I imagine that's what Discord and VRChat do now.
> I still remember the screen names of folks I interacted with in and on IRC, AIM, ezboard, IGN boards, forums, etc.
> I only remember a handful of HN usernames and absolutely zero handles from elsewhere.
Same. It's strange that in terms of interpersonal communication, we appear to be worse off than where were 25 years ago with AIM and e-mail. Back then it was common to carry on pretty deep conversations and relationships with people from all over the world. Part of my nightly routine was going over the day with my friends (both near and far). I would send long e-mail to my distant friends on their birthday, talking about all the things that had happened in my life, and they would do the same. I'd even keep in contact for years with pen pals I had met all over the world.
There's no reason this couldn't be done today, but I suppose most people prefer the much shallower interactions we have now.
That's a feature. HN is expressly built to encourage high-quality content in the comments - by removing or discouraging things that tend to bring poor content with them. Not showing up/down-votes counts, not allowing avatars and discouraging lengthy signatures, not showing previews of images linked, not allowing to reply immediately to comments (and the length of "immediately" getting longer the deeper the thread) - the HN is coded this way to encourage sticking to guidelines and making thoughtful comments focused on informative, interesting content.
It does make HN a worse place for building interpersonal relations, no question about it - but it also eliminates or blocks many of the enshitiffication vectors. Not all of them, and we do see changes over time in what's acceptable, going in the direction that some will dislike, but I think the fact that HN still existing in relatively constant form almost 20 years later shows that the strategy works, to an extent.
Don't worry in about 5 years they'll come to the conclusion I did about 5 years ago (because I was like that in 2014) - that is to say, they're realize it's futile and liberate themselves from the angst of it all and just accept the world is shit and the internet is shit and there's no going back and you will be free from such horrific lust for the dead remains of an archaic past. You will be glad when new potential "solutions" come up, but be sad to see the loss of expansion of said places and ultimately learn to accept that it's "creative destruction" on a smaller level, and the only people who care are the old heads who are a dying generation. We're like hippies at a dead show, with a few young hangers on who love to hear the glory of the old days, But they'll soon be dead too. Freedom in acceptance. Freedom in serenity. Free at last, thank god almighty, they will be free of such mental strife at last.
Winer in particular often writes in a style that approaches technology with the attitude that even the most obvious ideas weren't truly finished until he's had a hand in them.
This is controversial and perhaps rude but I've started to copy-paste these articles into ChatGPT to summarize into a single paragraph to help me decide if it is worth reading or not, for me.
Again, I mean no offense but humans only have 2,207,520,000 seconds on average. You can be goddamn right I'll take all the help I can allocating this precious resource called time.
Example output from this article:
> The author reflects on their early experiences in Silicon Valley, feeling at home among peers in the tech world but later realizing that many influential figures, including gatekeepers and executives, lacked a deep understanding of software. This disconnect was disheartening, as key decision-makers often misunderstood the potential and intricacies of the technology they were overseeing. Despite this, the author celebrates the web for its ability to connect people globally with shared interests and expertise, yearning for a return to a time when the internet was a space for meaningful connections and innovation. They also express enthusiasm for their current blogging on platforms like WordPress and Mastodon, envisioning future improvements and business potential.
Learn skimming? I say this as a old person who actually had "skimming" and "scanning" (different things!) as part of my elementary reading curriculum. We even had to practice skimming the whole (printed) NYT in 5 minutes. This piece is seven short paragraphs. I can get the gist and tone in a few seconds. Quicker, in fact, than reading the chatgpt snoozefest.
I'll happily dump an 80 page patent or court ruling into chatgpt for a summary. But if an actual human wrote a few hundred word personal blog post, I'd rather read it.
If you think that human mind works like this, then you'll find out that those precious ideas and things you want to think about come once in a blue moon if ever. ChatGPT is going to make you even dumber than you already sound like.
Not so much rude as.. silly. What's your goal here? Optimal resource consumption? Can't relate. It's not like absorbing more information faster makes me smarter or more capable. Or happier. If anything it's the opposite.
Hate to admit this but reading the comments will give a person way more knowledge than any given submission and if the comments are good the article is required reading before adding your own comment. Not sure ChatGPT is worth the effort here. By the time you cut and paste then read the gibberish one could have already read the top few comments (you can use the minus to minimize them and read the next thread). Like the time spent leaving this comment would be better used any where else, but yet here it is (all 300 seconds or if you like 300,000 ms)
The original reply was looking for a summary of the article which ChatGPT can supply, not more info, which is typically more where one might go to the comments and join the discussion. Two different streets on my map.
Just be aware that comments on HN are subject to all sorts of opinions skewed by bias and sycophantism, the same way Reddit or Mastodon silos tend to be. I think a lot of users think HN has some sort of immunity to that, so the tribalism tends to be strong and stubborn.
Agree 100% with everything you said and the comment I left was very low quality. It would be interesting to see a ChatGPT summary for each submission to HN and the comments to give you an overview of the content that was auto generated. The main point was the time it takes to cut, and paste is not zero. Personally, spending that time reading the top comments seems more useful (to me) than reading a ChatGPT summary of the article, but like you say two different streets.
> I later learned that the execs at most of the tech companies were similarly clueless on what made software possible, and basic stuff like trading off time for space and vice versa.
Why would an exec ever need to know about time/space tradeoffs? That's an engineering detail that is almost never relevant at the executive level. If you're pitching a business idea at this level, you're going to have a rough time.
>> was thinking about making a product, they’d invite this person to hear the pitch and if they didn’t like or understand it, they wouldn’t make it.
He says this like it's a bad thing. When really the company is just trying to find out if the thing being pitched can be sold to a not-engineer.
Too often engineers love an idea because it "can be done". They can list features all day long, but lack the skills to understand (or articulate) the benefits to the prospective customer.
Prioritizing time makes a product fast. Prioritizing space makes it cheap. Customers often do care about the outcome of a tech decision even if they don't care about the decision itself. Therefore the leaders at a company should care too, or should at least empower people below them to make good decisions.
I doubt consumers at the checkout line buying software are thinking about how fast or slow the software will run on their hardware compared to the competitor software
An exec that does not understand core tenets (and challenges) of the industry his company operates on is an exec that will drive said company out of relevance. Whatever the size of the company.
Boeing is one but a glaring recent example of that.
That description of time/space trade-off is completely orthogonal to the concept of software. If that was the author's intent it would be applicable to all consumer-facing industries.
Is it a space-time trade-off if you end up with little space and little time? Twitter chose to minimize both by storing less, that’s not really a “trade”.
I usually understand the term to mean that using more space lets you use less time and vice versa.
It didn't. It was the best place to find the best news stories, blog posts, journal papers and all the rest on any subject that interested.
Someone who you think is smart,informed and honest says "Mary, who I know, is smart and works in this field wrote about this issue here: link" Could take you to fantastic info you don't see otherwise. Obviously critical thinking and possibly verification required.
Now you don't need that because links are all de-amplified and many of the most thoughtful and interesting have left. This is the real shame of it all.
It's also pretty disheartening that certain famous 'tech' podcasts are hosted by people who've never written a line of code in their lives. Like who are you to comment on where AI will be in 10 years? You don't even know what back propagation is.
Building a product or business is a multi-skilled task. Broadly speaking you need to cover product development, marketing, finance.
Good leadership covers all 3 bases, but leadership in the latter 2 is more important than leadership in product.
Which sounds backwards, but you're the tech guy providing the tech. Leadership is figuring out what / how / to who of marketing coupled with the balancing of finances to make the business work.
VC funding of course up-ends this model. It removes financing and marketing (and even product fit) from the equation. Here's a pile of money, "go build something and we'll figure out the rest later..."
Sure 90% fail to figure it out, and the business part fails, but that's OK because the VC (although not always the techie) comes out a winner. It's a great model to handle the edge-case of good ideas that are hard to pull off.
If you end up in a non-vc business though, don't be surprised if most of the layers above you don't care about the tech.
There’s a difference between technical at the leadership positions vs every position.
Nvidia’s CEO is technical. Google’s was technical for a very long time. AMD’s is technical. Amazon’s was for a very long time. Microsoft’s was for a very long time and is again.
I think it’s a bold claim to make that technical people are only capable of designing for technical people. There’s plenty of people who understand how to build products that connect to the consumer and non-technical people don’t have a unique ability to do so.
You imply you believe that technical people are one-dimensional and lack some fundamental skill. You imply that some other people have this empathetic skill that technical people lack...
The fundamental skill I think technical people lack is understanding how non-technical people use software and technology. We just can’t imagine how non-tech people see things, and since most users of most products are non-tech users, many decisions should be made by those types of people.
I agree that many technical people don't give a shit about users or usability. However it is also true that non-technical people often don't give a shit.
Making sweeping assumptions about a group will only harm yourself.
Good techos must also be good at talking with users and understanding user needs: it is a critical skill for designing user interfaces (software or hardware). Anyone lacking user facing skills severely limits their salary and career choices.
There's a reason so many development methods and startup advice focuses on the user.
I worked in a very small corner of the modern Silicon Valley area recently. I was surprised to see so many companies, VCs, and founders giving such a similar mystique to a certain youtuber/streamer that i won't name here.
Its very confusing to see people who have been in the industry for potentially decades more or less blindly following the whims of a person with very little real world experience in the industry to back up such strong, opinionated claims. Interesting to see based on the OP that its not necessarily a new phenomenon.
This is like how my dad says SNL was awesome when he was 19 and sucked when I was 19. But I think it was awesome when I was 19 and sucks now that my kid is 19.
I may be missing the point of this blog but I don't think I am. I too miss the era of blogging but this isn't it. The only thing this piece does is name drop--without having the courage to actually name names--and point out how those "famous" (we can't judge, as they're unnamed) people were so wrong back then and the writer was so right. The line that actually answers the title is a cliche and isn't elaborated on: the web brings like-minded people together from all over the world.
Is Silicon Valley still like how the author describes it? I ask this as a Canadian considering moving down there. I want to go to a place where I feel understood.
Depends on you and a bunch of luck. There is a higher then average amount of technical folks, but a lot of them just are in it for a paycheck and focus on normal activities. If you put yourself out there, you can find a community, but it might involve some luck, depending on what you want out of it.
I have a writing tool I call wordLand, it connects directly to WordPress, and from there, one of my sites is hooked up to Mastodon via ActivityPub.. The reason I’m having so much fun with this mofo is that I have most of the features of textcasting now, and it’s all flowing out through Masto, and I’m not typing into a freaking tiny little text box!
Applying the philosophy of podcasting to text.
Interop between social media apps based on the features writers need.
A prototype of self-hosted editor that "publishes" format-appropriate versions to different social media sites, while the author retains copyright, data and compute control of the original source. By https://davewiner.com/, pioneer of weblogging, RSS syndication and more.
Kind of funny to see Dave Winer of all people saying he doesn't sweat the technical details here.
So I think this is what's going on. Dave had an idea he called textcasting which he describes on his textcasting.org site, I didn't see anything really new there - syndicating text from your blog to any other app or social media site, seems to be the basic idea.
A protocol called ActivityPub has been around for a while which enables this exact scenario, it's the core of Mastodon etc. and I believe is even supported by Threads.
Dave was pleasantly surprised to discover that this all just works on his Wordpress.com blog. I think the underlying implementation is probably the ActivityPub WordPress plugin which is developed primarily by Matthias Pfefferle - https://wordpress.org/plugins/activitypub/ - you can install and run this plugin on any WordPress site.
I agree with Dave where he says Automattic is sleeping on the potential here, I feel ActivityPub support should have been rolled into WP Core a few years ago. The tech for this federated media is getting pretty good. Mastodon is a thing, it works, adoption is slowly creeping up but still not really huge. WP could roll out ActivityPub support to 800 million websites and then if you so choose your own website then becomes a node in the social network and you don't have to sign up with anybody else to make that happen. This is once-in-a-generation sort of potential perhaps. Pfefferle did the original version of the plugin himself and Automattic hired him a few years ago.
In theory anyone who deals with a lot of WP installs could be putting their own money/time/code into this vision (your personal website is your social media profile, basically), and piggybacking on the established fediverse as modest as it may be, and growing it.
So it's a POSSE[0] system, like everything should be.
You keep your content on systems and platforms you control. Then republish to walled gardens and silos, either via a link to your own platform (which is downranked by algorithms...) or repost the content.
I have been trying to understand what is being suggested here and all I can think of is some Buffer.co-like posting UI for the desktop. The parts of podcasting that make it discoverable, consumable, shareable is already in the web for text…
> Mastodon, while they’re doing excellent work, is trying to wrangle an already large community into a set of consistent interfaces.. Automattic has a small team whose only job is to make WordPress work with Mastodon. So I can build software that works with Mastodon without venturing into the rough seas of Mastodon-land, I can stay on the cruise-liner, which is the WordPress API.
Wordpress doesn’t have native built post to Facebook/Twitter/etc. functionality so I’m not sure what the author is still advocating for…an API layer to post to social networks still is just a Buffer.co-like. I don’t disagree but what the author is suggesting will post to social networks, then I am supposed to use whatever UI to consume the posts?
To be honest, if the author of this communicated pitches as ineffectively as this piece communicated the topic in its title, I am not surprised that whatever nameless gatekeeper he's talking about passed on his pitch.
That's not even to get into the fact that executives don't necessarily need to understand software implementation details to size up an idea.
The author expected the people who were gatekeeping which tech got built to understand the tech that they were deciding on [0]. Instead they did not understand tech but understood business. Which means that a techie explaining the idea to them had to couch it in business terms. Y'know: a pitch.
If a techie is explaining a tech idea to another techie, we use completely different language than if we're communicating to a non-techie. This blog post feels like that, like they're talking to a group of peers rather than an audience with a different viewpoint on the world.
[0] I think it's a little naive, but perfectly understandable, to expect this kind of person to understand the tech. It is interesting that these decisions are made by business folk with no understanding of the tech. There are, after all, business people who do understand tech available to make this kind of decision.
