[HN Gopher] AI is killing the web - can anything save it?
___________________________________________________________________
AI is killing the web - can anything save it?
Author : edward
Score : 104 points
Date : 2025-07-20 09:33 UTC (13 hours ago)
(HTM) web link (www.economist.com)
(TXT) w3m dump (www.economist.com)
| sylware wrote:
| Start to regulate the technical protocols to access the AI
| prompts.
|
| Like regulated noscript/basic (x)html interop. Or 'curl' based
| simple APIs.
|
| Basically, if the whatng cartel web engines are not anymore
| required to access and use "AIs", things will start to
| significantly move.
| input_sh wrote:
| https://archive.is/nhrYS
| pknerd wrote:
| I don't get why the articles behind paywalls are shared here.
| aw4y wrote:
| they killed the web, not the AI.
| keyringlight wrote:
| One of the things I've been wondering about with the 'digital
| detox' trends or one of the younger generations getting
| dumbphones instead of smart, is why haven't the papers found
| some way of turning back the clock to explore capitalizing on
| that when it's supposedly hard to sell news now. 24 hour news
| is decades old at this point and the constant firehose of
| events from every location on the globe is tiring especially
| if only a tiny fraction is directly relevant to you. I'd be
| interesting if they could make a more attractive
| 'news/analysis product' like a newspaper or the evening news
| broadcast which is distinct from what is readily available
| from all the other sources.
| janice1999 wrote:
| It's the other way around. Paywalls are a result of the web
| and the Ad companies which power it killing the revenue
| models of publishers. AI, which steals even more and
| repackages their content, will make it worse.
| ElisaChemy wrote:
| Totally agree, here is something is use to read without
| paywalls
| https://archive.ph/2021.09.10-164025/https://www.bostonglobe...
| kikokikokiko wrote:
| Because paywalls are optional, at least for a crowd such as the
| HN crowd. Information wants to be free.
| mdavid626 wrote:
| Yeah, just share the archive link directly.
| aspenmayer wrote:
| Canonical links are preferred, per guidelines. Paywalled
| sites that are otherwise on topic with a workaround are
| usually allowed.
|
| https://news.ycombinator.com/newsguidelines.html
|
| https://news.ycombinator.com/newsfaq.html
| sharpfuryz wrote:
| The web as we knew it -- open, chaotic, full of real voices
| already gone. Free speech isn't what it was 15 years ago; it's
| filtered, throttled, and buried under bots and algorithmic noise.
| But AI isn't the root of the problem -- it's just another layer.
| The real issue is that the current internet model no longer
| serves people; it serves platforms. Maybe it's not about saving
| the old web. Perhaps it's time to build a new one--one that puts
| users, privacy, and real expression first.
| lmpdev wrote:
| The thing that stops me pursing this idea though is how do you
| verify contributors to this new internet aren't
| platforms/businesses?
|
| Where do you draw the line?
|
| Who gets to draw the line?
| xyzzy123 wrote:
| Even if you could do it perfectly (distinguish "authentic
| people" from slop merchants) the same old actors will do the
| same old things as long as the incentives are there. They
| will just wear "real people" like skin suits. Almost worse :/
| sircastor wrote:
| This is an incomplete thought, but a friend of mine has this
| idea around reputation built through a sort-of key signing.
| You get a key, your friend gets a key, you sign each other's
| keys. The key can serve as an indicator of trust, or validity
| that an individual's contributions are meaningful (or
| something). And if your friend suddenly turns into a
| corporate shill, you could revoke that trust. And if the
| people haven't established their own trust with that person,
| their trust goes when yours does. Transitive trust.
|
| It obviously has some flaws, and could be gamed in the right
| circumstances, but I think it's an interesting idea.
| moron4hire wrote:
| Sounds like following people on a social media platform and
| only reading posts from in your network. Which is exactly
| how most people I know use Bluesky.
|
| It works better than Twitter's algorithmic feed but it's
| still not foolproof because not everyone has the same idea
| of what sort of content they are willing to trust/ track.
| DaSHacka wrote:
| Isn't this just a standard pgp web of trust?
|
| https://en.m.wikipedia.org/wiki/Web_of_trust
| salawat wrote:
| Anything that requires the end user to internalize PKI is
| dead on arrival.
|
| A) The interface won't get intuitive enough.
|
| B) The asshats will still find a way in.
|
| C) Ain't nobody ever met someone in the real world and gone
| "Yo dawg, what's your public key?"
|
| Encryption is just a machine that turns already hard
| problems into key management problems.
| asplake wrote:
| Why that line in particular? It seems not to be about the
| quality of the content. Part of the issue is that businesses
| were advised to produce useful content, but the motivation
| for doing so is disappearing. A net negative, surely?
| JKCalhoun wrote:
| A plug-in. Trusted users thumbs up/down sites and ratings are
| recorded in a database. The plug-in visually differentiates
| shite links (according to database) so others can avoid
| clicking on them (or they can hide them altogether).
|
| A kind of PiHole for just shitty SEO sites.
| krapp wrote:
| It already exists, it's called the Gemini protocol:
| https://geminiprotocol.net/
| jmclnx wrote:
| I moved my site to Gemini, finished middle last year.
|
| _Clients:_
|
| https://en.wikipedia.org/wiki/Gemini_(protocol)#Software
|
| _Some links to find content:_
|
| gemini://sdf.org
|
| gemini://gem.sdf.org
|
| gemini://gemi.dev/xkcd/
|
| gemini://warmedal.se/~antenna/
|
| gemini://gemini.circumlunar.space/capcom/
|
| gemini://skyjake.fi/~Cosmos/view.gmi
|
| gemini://calcuode.com/gmisub-aggregate.gmi
|
| gemini://tinylogs.gmi.bacardi55.io/
|
| gemini://sl1200.dystopic.world/juntaletras.gmi
|
| gemini://tilde.team/~khuxkm/leo/
|
| gemini://raek.se/orbits/space-elevator/
|
| gemini://fediring.net/
| thoroughburro wrote:
| Do you have readers, or is it just for you?
| jmclnx wrote:
| I do not understand. Clients link above has a list of
| clients to read gemini sites.
| VoidWhisperer wrote:
| They are asking if your site has people viewing/reading
| it after you moved it to Gemini or if it is just you
| reading it at that point.
| rambambram wrote:
| Just ignore the platforms. Use RSS on a body of self-curated
| websites/bookmarks. Click to read the articles and essays on
| their own domains (show the creators some love by doing that),
| and click around over there on that other domain.
|
| I built my own system for that, but I know for sure this is
| possible with off-the-shelf (open source) software.
|
| It takes some time to get used to this. No saturated video
| thumbnails, no infinite scrolling, no notifications. It's
| slower and feels more boring in the beginning. But it becomes a
| blessing very soon, when you go back to LinkedIn's feeds or
| Youtube's algo grid after a month and it feels like a punch in
| the stomach.
| 627467 wrote:
| I worry that AI/bot presents as a desincentive for proper RSS
| distribution. Authors may not don't want to provide easy
| access to their content by bots. Maybe paywalling? Maybe
| proof of work solves this?
| nchmy wrote:
| I used to be a heavy user of RSS, back in the Google Reader
| days. I loved it for following a wide array of different
| blogs. I'm not really sure why I stopped with rss - I
| switched to viable alternatives to Google reader when it
| died.
|
| Recently I've been keen to get back into this way of using
| the web, because I have evidently been sucked into scrolling
| on the platforms until the algorithms give me something I
| want to see.
|
| The other day, one of my favourite web dev blogs (and one of
| the only blogs I actually seek out) created this fantastic
| compendium of Web Performance resources and blog links, along
| with an associated rss opml file. Surely this is the push I
| needed to get back to the glory of the web.
|
| https://infrequently.org/links/
|
| But I definitely need to put in the effort to discover other
| eclectic blogs. I really miss reading long, authentic things
| on diverse topics
| rambambram wrote:
| Nice, thanks!
|
| I think OPML is underrated and the combination of RSS
| (Really Social Sites) and OPML (Other People's Meaningful
| Links) could give the open web a resurrection as the social
| media of choice for curious people.
|
| Right now, I'm working on integrating more and more OPML
| functionality into my RSS software. I envision a quick way
| of exploring and discovering new links/feeds from
| sites/feeds that I already follow.
| slater wrote:
| > RSS (Really Social Sites)
|
| Rich site syndication.
| j45 wrote:
| That's something the few can do, but not the many.
|
| As open source improves at user onboarding, and user
| experience, there might be a chance.
| nirui wrote:
| Ignoring is not how it works. Internet is a basically huge
| social circle, if not enough people got on broad, a site can
| die out really quickly. I've observed quite few examples of
| small community closed down because no one was there anymore,
| some websites that I loved as a child no longer exists
| because of this reason too, gone with it is all the content
| they once hosted.
|
| Here's the problem:
|
| 1. Software/Infrastructure have a cost: If you want to self-
| host, there's a consistent dread of maintaining things. It
| wears you down, slowly maybe, but eventually.
|
| 2. The problem of discovery: Back to the past, people used to
| sharing links and resource manually, often on a forum ("forum
| life", i call it). But now days people are more rely on
| platform recommendations (starts from "Just Google it"). If
| your content/link is not recommended, then you can't reach
| far. Also, people now days really hates registration (and
| memorizing/recording account/password), and they will not
| even try to use "strange" websites.
|
| 3. Government regulation: The government pushing laws upon
| laws that could restrict self-hosting content, by either
| making self-hosting difficult, or forcing websites to self-
| censor (which most personal sites just don't have enough
| admin to do).
|
| 4. _Some_ people who has the capability and know-hows on
| solving the problem are "solving" it the wrong way. Instead
| of creating systems that modern users would love to use, they
| tries "being back the old way" so do speak, but not giving
| any consideration on why people abandoned "the old way" in
| the first place. The software they created maybe even quite
| hostile to regular non-tech-savoy people, but hey at least
| they themselves thinking it's cool.
|
| There are few projects gets it right, like Mastodon, and
| maybe Blue Sky etc. But, then these project still don't earn
| a lot of money and political capital, meaning it still can't
| escape the point 1 above and maybe point 3 as well.
|
| Over all, I think it's less that the platforms exploiting the
| Internet, it's mainly that most people just "moved on" to
| what could make their life easier. Internet is a tool after
| all.
|
| P.S. If someone wants to solve the social media over-
| monopolization problem, I'd recommend that you make sure
| you're "user forced", user, user, user, regular old man/woman
| John/Marry Doe user. That's how you create social
| circle/network effect and that's how you grow and sustain.
| asdff wrote:
| I am an RSS user but it is pretty frusterating these days
| being one. All of the I guess "first tier" sort of sites
| you'd really want an RSS feed for don't have one any more or
| offer a truncated one that forces you onto the platform (yes
| I roll my own morss, doesn't always get the content). You are
| left with sort of second tier news websites that pollute
| their feed with reposted AP content you might even see on
| several same feeds you follow.
|
| And the biggest issue is that no one is starting a new site
| and implementing RSS. Seems like for a lot of RSS feeds I
| follow, the only reason they still exist is because the
| webmaster has not yet culled the service for whatever reason;
| like some of these links are found on vestigial web pages
| that look like 2007 internet whereas the rest of the site is
| modern.
|
| And it makes sense why RSS is dying. It is a huge free bone
| tossed to the community. You don't see free bones tossed out
| anymore without a string attached to pull you back into some
| profit making angle. Everyone wants you on their site so they
| can serve you ads. They don't want you using a feed reader
| and getting that content without having to see an ad.
| rambambram wrote:
| I have to agree with you. Completely.
|
| On one hand I think it's a shame and I do miss feeds on
| certain (big) websites, but on the other it makes me
| appreciate the small web or indie web or just open web
| more.
|
| Feels like rehab after two decades of 'social media'. But
| the open web is the ultimate form of social media itself,
| if you'd ask me. I plea for a name change of RSS to Really
| Social Sites. I already started calling it like that in my
| own software.
| patrickk wrote:
| Freetube is a way to achieve this with YouTube. You just get
| RSS feeds, you avoid endless scrolling.
|
| https://freetubeapp.io/
| marginalia_nu wrote:
| You're using the wrong tools to browse the web if it seems that
| is the case.
|
| The weird, creative, bordering on unhinged part of the web is
| still very much around and alive. It's just that you need to
| depart from the major social media sites and search engines if
| you want to find it again.
| barbs wrote:
| Well said. There's a good search engine for that, maybe
| you've heard of it?
|
| https://marginalia-search.com/
|
| ;)
| marginalia_nu wrote:
| Well as it happens...
| pjerem wrote:
| Hahaha
| thoroughburro wrote:
| I've tried Marginalia about... probably 10 times, at this
| point? Every time I want niche search results. I haven't
| found an interesting site through it, yet.
|
| I love the concept and want it to work! I pay for Kagi; I
| value search.
| marginalia_nu wrote:
| The explore mode[1] is probably the tool you're looking
| for if you're just looking for something interesting /
| demonstration that the weird web still exists.
|
| [1] https://marginalia-search.com/explore
| spacemadness wrote:
| This is a step in the right direction. Thanks for this.
| JKCalhoun wrote:
| Bring
|
| Back
|
| Web
|
| Rings
|
| (But seriously, I think I would love to rat-hole down
| interesting web rings.)
| larodi wrote:
| Delete all social media immediately. It's the equivalent of
| Neo unplugging himself, taking these tubes out of his throat.
| belter wrote:
| > The web as we knew it -- open, chaotic, full of real voices
| already gone. Free
|
| Commented on a site whose top pages are curated manually....
| tropicalfruit wrote:
| > one that puts users, privacy, and real expression first
|
| users aint that special.
| pupppet wrote:
| It will never happen as long as Google is able to gatekeep the
| Internet with its search and browser. Even if you could find
| enough power users to break out and create something that hits
| critical mass, user-powered indexes don't scale. Whomever
| swoops in to fix the problem immediately becomes the new
| Google.
| nine_k wrote:
| The problem of the pre-platform Web is the difficulty of
| discovery; your interesting content will have but a few
| readers.
|
| The lure of platforms, like Twitter, or, well, HN, is that your
| content can potentially be seen by "everyone". Going viral is
| fun but not that important; being seen by the right people you
| never knew, or never had a hope to grab attention of, is much
| move valuable. This leads to much stronger cross-pollination.
|
| (Spam is a problem here, but spam is also a problem in similar
| biological systems; blooming plants release tons of pollen, and
| then tons of seeds, most of them fruitless.)
| ntstr wrote:
| No one noticed the parent post is LLM slop?
|
| Spams of groups of threes (open, chaotic, full of real voices -
| filtered, throttled, and buried - users, privacy, real
| expression)
|
| It's not just X - it's Y type of sentence structure Vapid
| marketing style writing that has no real substance (Maybe it's
| not about saving the old web. Perhaps it's time to build a new
| one)
|
| Of course, there are emdashes too, they may not betray LLM
| alone as they exist in literature and a minority like to use
| them in internet comments but when they are present along with
| other signs of slop they are still a strong tell, particularly
| when they are numerous.
|
| Is this satire? or trolling? it is concerning everyone replies
| to it as if there had been human thought behind this drivel.
| r33b33 wrote:
| Don't forget semicolon. Normal people don't use that.
| gaws wrote:
| Semicolons are fine so long as you know how to use them.
| Marsymars wrote:
| I don't know that I'd call myself normal, but I use
| semicolons regularly, though infrequently.
| Marsymars wrote:
| I didn't notice because I unconsciously skim over slop-
| looking comments without evaluating whether it's human-
| written or not, and only read the more interesting comments.
| r33b33 wrote:
| LLM reply. At least get rid of the dashes, come on.
| nonvibecoding wrote:
| This didn't just start now. It's been fading for over a decade. I
| remember when every forum had its own look, strange layouts,
| unique colors, and a vibe you couldn't really describe but you
| felt it.
|
| Now everything feels the same. Same layout, same font, same clean
| boxy design. Sites copy each other. AI just made it more obvious,
| but the soul started slipping away long before that
| thom wrote:
| I remember usenet where every forum was exactly the same and it
| was still better than today, so I'm not convinced this is a
| fundamental symptom of our current problems. To me it's more
| that the internet has lost any sort of physical, spatial,
| kinetic quality. There's no time or place, no nooks and
| crannies to disappear into with friends. Just an unyielding
| cacophony. I agree it's all undifferentiated but it's not the
| aesthetics that are the problem for me.
| nonvibecoding wrote:
| Yeah, maybe you're right. Could be nostalgia playing tricks
| on me. I just remember how exciting it felt to join a new
| forum, or discover something like eMule, Sababa DC, or random
| p2p tools.
|
| Everything felt raw and full of possibility. Even if a lot of
| it looked the same, it didn't feel the same. There was this
| sense of exploring something alive.
| thom wrote:
| It's possible that various Discord servers, or obscure
| streamer chatrooms still feel like this, and we're just
| old. But it definitely feels like the default has become
| very top-down and public instead of bottom-up and intimate.
| spacemadness wrote:
| I think the difference there is streamers are just there
| to get money from their audiences. Doing something they
| like sure, but a vast majority are trying to make a
| living. That has a different context entirely.
| thom wrote:
| I think there's an extremely long tail of streamers and
| associated chat communities that are untroubled by any
| form of financial rewards. When I speak to people in
| those communities it sounds to me like the closest thing
| to IRC in the 90s - tight-knit groups with regular
| comings together at specific times and places, being
| their whole selves with each other.
| pjc50 wrote:
| "Context collapse"? The phenomenon that, no matter where you
| go and what the nominal topic of discussion is, it always
| comes back to US politics.
| BriggyDwiggs42 wrote:
| I think the issue is optimization. As these sites have grown
| more efficient at gaining and exploiting (like a natural
| resource) users for money, they've optimized away mechanisms
| people used to form community and such. Moving to a feed of
| recommendations instead of a feed of people you follow is an
| easy example, but there must be a thousand little examples
| like that.
|
| Fundamentally, if the goal is to make money, then that's what
| will be optimized for, and in this case that goal appears to
| be in conflict with the formation and maintenance of
| community. It was just a matter of time.
| lmpdev wrote:
| My memory of this was Facebook overtaking MySpace
|
| I remember being 13-years-old and completely baffled people
| preferred the platform where I had no say over the HTML on my
| page.
|
| I didn't understand how people could prefer a boilerplate with
| profile picture and name over an actual artefact made by the
| person.
| nonvibecoding wrote:
| I loved Myspace. You could talk directly to bands members (At
| least the unknown punk bands I was following back then)
|
| Once they lost all the pre-2016 content, I think that was it.
| Hard to make a comeback after something like this
|
| https://www.theguardian.com/technology/2019/mar/18/myspace-l.
| ..
| aspenmayer wrote:
| It's a long shot, but you might find some resources here:
|
| https://wiki.archiveteam.org/index.php/Myspace
| alganet wrote:
| I know it sounds counter-intuitive, but I think we need less
| collaboration, less competition, and less team dynamics in
| general. Anything that does cross-pollination should be opaque.
|
| More individuals cultivating personal points of view drastically
| different from homogenized masses.
|
| That extends way beyond the web though.
| nine_k wrote:
| Pafnuty Chebyshev, a Russian mathematician who discovered a
| bunch of important things, deliberately limited his intake of
| other mathematicians' works, in order to force himself think in
| original ways, not ways suggested by others' works.
|
| This medicine needs to be taken in moderation though, else one
| can end up reinventing some key wheels instead of speeding
| forward on these wheels, like
| https://fliptomato.wordpress.com/2007/03/19/medical-research...
| austin-cheney wrote:
| I thought social media killed the Web 20 years ago. RIP
| anilgulecha wrote:
| Making it federated (so it's a true network of people's sites) is
| what can theoretically save things. But given under 0.001% can
| self-host, I don't see how that can work .. the centralized
| services are slated to win.
|
| Perhaps some global law could help - significantly
| disincentivizing for centralization and network effects.
| IanCal wrote:
| I feel like the barrier for self hosting could be so much
| lower. The resources required to host a static site are tiny
| and even a dynamic one with comments accessed by all the people
| I actually know could easily run on a cheap router.
| jen729w wrote:
| I think self-hosting is a distraction. You can make your own
| site using Astro and deploy it for free to Netlify and still
| get 99% of what we're talking about here.
|
| If that was less scary maybe more people would do it!
| JKCalhoun wrote:
| Welcome to MySpace!
| zer00eyz wrote:
| > But given under 0.001% can self-host, I don't see how that
| can work
|
| The place where the web is still great is where you have to be
| invested to be a real participant. Everyone can yell about
| politics in a text box on twiter/FB/reddit/HN or post photos to
| IG/Dataing site Or videos to twitch/YouTube.
|
| If you can host something, even for a small number of people
| your one of the rare few. If your "into" something where there
| is a focused community then your back into one of those 1%
| pools where people vibe and participate.
|
| To make an analogy of it: The web is now a tourist town.
| Everyone is interested in making money off the visitors with
| the flashy lights and signs luring them into the over priced
| tourist traps. The locals, the natives, the REAL .01% know
| where the cheap places with great food and local flavor are.
| tobyhinloopen wrote:
| The web was already dead.
|
| > We care about your privacy. Can we please put a camera in your
| toilet seat for a personalized experience? > > [ ACCEPT ]
|
| Browsing the web is a nightmare these days, I rarely visit "new"
| websites
|
| > Subscribe to our spam for a 10% off coupon > > [ ] [SEND]
|
| It is just a pain to visit any website these days... anyone
| involved creating these modern monstrosities should just fire
| themselves and go on a hike or something.
|
| > We rely on invasive, tracking ads! Please enable your adblocker
| so we can get 0.00001 USD, please. > > [IVE DISABLED MY FIREWALL
| AND ANTI-VIRUS] [PAY 999 USD A MONTH FOR AN AD-FREE EXPERIENCE]
| casey2 wrote:
| It's still not as annoying as the assorted influencers who repeat
| The Economist headlines and articles back at me
|
| Anyway this article is about AI replacing web search, not
| "killing the web" which I would take as it somehow deleting or
| overwriting content on existing webpages. Or generating so much
| spam as to make the web unusable for the average person.
|
| Large sites that can't exist without "traffic" already killed the
| web a long time ago. A paywall is the proper solution, not ads in
| content and content in ads. That means you will have lower
| traffic, it doesn't mean you are being killed. It just means you
| stopped assaulting passersby who are linked to your site.
| rambambram wrote:
| > Anyway this article is about AI replacing web search, not
| "killing the web"
|
| Indeed, exaggerating title. But we all have to get the idea the
| web is really dying, so we give up working on it. We have to
| get that idea because the genie of the web is already out of
| the bottle for 30+ years. That stuff is going nowhere. The open
| web is a hindrance for big businesses. Big business wants to
| keep internet infrastructure to push apps, AI and what not, but
| does not want to keep the open web.
| mmcconnell1618 wrote:
| I just read Technofeudalism by Yanis Varoufakis which has an
| interesting perspective that "cloud capitalism" is replacing
| traditional capitalism and competition. A few players are
| assembling their own fiefdoms inside dominant web/mobile
| platforms.
| https://www.penguinrandomhouse.com/books/751443/technofeudal...
|
| The internet doesn't have a clear, simple, micro-payment system
| that would allow people to reward value, so instead we have an
| attention based system where the number of likes and followers
| grants social status and financial opportunity.
| jrexilius wrote:
| When cryptocurrency first started getting attention
| (2010,2011-ish?) I was so excited that a potential
| micropayments system would come out of it and solve this
| problem. Sadly it did not go that way..
| egypturnash wrote:
| god fr real, everyone making shit and selling it online is
| _still_ paying their tithe to Visa /MC and working under
| their rules as to what you can and can't sell, and possibly
| another tithe to Paypal or Square or Stripe or whoever on top
| of that. Crypto's just a giant sucker trap and the amount of
| stuff you can buy with it without paying your tithe to
| Visa/MC to turn it into real money is infinitesimal.
| medion wrote:
| No. All great things come to an end - artistic movements,
| cultural, nations, etc etc - the end of the internet is now.
| _nalply wrote:
| AI is one sharp tool cutting slices from the old internet. But
| perpetrators have used different tools from the start: SEO spam,
| algorithmic feeds, embrace/extend/extinguish, building moats, the
| attention economy, and many others. AI is just the next
| newfangled sharp tool.
|
| In other words, I don't think that AI is killing the web.
|
| It's being profit-oriented and running amok in an unleashed way.
| It's prisoner's dilemma. You know, if you don't do it then
| someone else will do it and you lose. Enshittification is one
| consequence. The internet experienced it from the beginning. But
| only about fifteen years ago companies learnt how to squeeze the
| last drop out and, like in the tragedy of the commons, everybody
| is worse off.
|
| And what's the most catastrophic? People are confused. They look
| at the tools but not at some famous people behind these rampages.
| Of course as leaders they just optimize the hell out of the
| internet with the target that their companies thrive. But in
| doing so they cause heavy damage.
| dankobgd wrote:
| paywall can surely save it
| noiv wrote:
| I think, the web was killed before by human slob search engines
| can't or won't filter. Now we find out, a little longer prompt in
| an AI chat returns better results. So what?
| JKCalhoun wrote:
| But for the existence of YouTube, I think the ad business
| killed the internet.
| SalariedSlave wrote:
| The "web" is already just business infrastructure. It already
| was, much prior to AI. I would challenge the assumption that
| there is anything worth saving.
| JKCalhoun wrote:
| Yeah, the sentiment I was looking for in this thread. I could
| get over the death of the Web pretty quickly, I think.
| deadbabe wrote:
| Web is obsolete. Going forward AI is the first and maybe last
| step to getting information about a topic. No need to sift
| through ads, forum drama, clickbait blog posts, comments etc...
| just straight compiled information into your brain as quickly as
| possible. Yea sometimes it's wrong, but sometimes things you find
| on the wild web are wrong anyway, just deal with it.
|
| I find that when people pine for the old web, what they're really
| asking for is some way to connect to other people and see things
| that people have written or made just for fun in a genuine way,
| without it being performative, derivative or for other
| motivations.
|
| In theory social media should have been this, but people's
| constant need to accumulate validation or tendency to produce
| meme-like content adversely affects the quality of their output,
| giving it a machined style feel that rarely feels genuine or true
| to their human nature. Instead of seeing people's true
| personalities, you see their "masks".
|
| Thus the issue is not rooted in a technical problem but rather a
| cultural one: people no longer naively share things that don't
| fuel their ego in the most perfect way.
| _hao wrote:
| Until that same AI starts shilling ads and certain viewpoints
| peddled by their owners in the output... This will happen 100%
| (ads, the other bit has already happened). The economics of all
| of these models doesn't work as is. There will be a major
| squeeze down the line.
| JKCalhoun wrote:
| Some of us have dipped our toes in local LLMs. To be sure,
| the ones I can run on my hardware always pale when compared
| to the online ones. But perhaps in time the ones you can run
| locally will be good enough.
|
| Or perhaps an Apple or Kagi will host an LLM with no built-in
| monetization skewing its answers.
| jillesvangurp wrote:
| At the same time, apps are also a bit in decline. People still
| make them but the whole race for making it to the top 10 in the
| app stores seems to have faded away. And a lot of them are simple
| web page wrappers. People still install some apps but more on a
| need to have basis than that they are constantly adding/removing
| apps. So, I don't buy this "the web is in decline" framing.
|
| Change is a constant on the web. Things were very different in
| 1995 (plain html, no good search engines), 2005 (no widespread
| web capable smart phones usage yet, Google, AJAX), 2015 (peak
| social media and app hype), and 2025 (social media has shifted to
| new apps and lots of people are disengaging entirely, AI is
| starting to threaten Google, content aggregators serve most web
| content).
|
| For 2035, I would predict that AI will drive a need for
| authenticity. Existing platforms don't provide this because they
| lack content signatures. We've had the tools to reliably sign
| content for decades. But we don't use those a lot except for DRM
| content behind paywalls (for commercial reasons). So, you can't
| really tell apart the AI generated propaganda, marketing,
| misinformation, etc. from authentic human created content by
| individuals you care about. And that might be contributing to
| people disengaging a bit. But you can see the beginnings of this
| on platforms like bluesky and signal which push end to end
| encryption and user verification. People might share AI nonsense
| via these platforms. But they seem to be less about that as say
| X, Tik Tok or Instagram are. We sometimes watermark our images.
| We don't digitally sign them. Why is that?
|
| Just speculating here but the web could use a big upgrade here
| and do more than just certify domain name ownership. Which is
| fairly meaningless if the domain is some big network with many
| millions of users. What about certifying content itself? Reliably
| tie content to their creators in a way that can't be forged. IMHO
| this is long overdue and the related UX challenges are there but
| solvable in principle. DRM is a prime example of a fairly usable
| implementation. Just works if you paid for the content. Signed
| content would make it very challenging to pass off AI gibberish
| as authentic if it's not signed by a reputable private key. And
| if it happened anyway, that would damage the reputation of that
| key. I don't exclude the possibility of reputable AIs emerging.
| How would you tell those apart from the disreputable ones?
| amelius wrote:
| Maybe start a new movement, similar to the Amish. And have a
| completely separated version of the internet.
| bell-cot wrote:
| Problem #1 - to "save it", you first have to define the idealized
| and/or snapshot-in-time web that you want to save. Don't expect
| much agreement here, especially on the details.
|
| Problem #2 - if you aren't the Emperor of Earth or some such, how
| could you make your ideal web stable over time, in today's world?
| kkfx wrote:
| We are many, search engines are the mean to discover things
| because even with usenet it's impossible for a human to discover
| via URLs and links enough information on the web, that's the real
| revolution: links are useful but not enough. Search engines are
| the best tool we have had so far to find knowledge around the
| web, now LLMs try to surpass traditional search engines milking
| knowledge from web contents, like we have many articles about
| wildfires in a region, but let's say not one about wildfire
| trends in that region, an LLM could try to spot a trend milking
| all articles in a significant timeframe. The Conrad Gessner's
| Biblioteca Universalis dream.
|
| So well, LLMs do not kill the web, eat it. We are still almost
| the sole valid source of data for LLMs.
|
| What really killed the web are social networks as proprietary
| walled gardens instead of an open Usenet with a web companion for
| stuff to be preserved for posterity or too long/complex for a
| mere post. What killed the web is the fact that ISPs do not offer
| an open homeserver instead of a closed box called "router" even
| if it's a limited homeserver. With an open version, with IPv6,
| anyone could buy a domain name and publish from his/shes own iron
| a blog with a ready-to-write software, with automatic RSS feeds,
| newsletters etc. If we give such tool to the masses the original
| web will be back but it would mean free speech and
| giants/politicians etc have free speech preferring ways to master
| public topics through their platforms to hide from most stuff
| they dislike and push ideas they like...
| salawat wrote:
| Search engine indexes being turned into copyright enforcement
| levers also significantly killed the net as it created scarcity
| in info dissemination for the sake of maintaining info
| asymmetry.
|
| Go ahead and try to find JLG equipment/service manuals on the
| open net anymore. I'll wait.
| kkfx wrote:
| they are anyway needed and we also have YaCy as an example,
| and other distributed search solution. The point is that most
| do not participate so only commercial one get enough
| resources to be useful.
| senectus1 wrote:
| I _think_ the economics will save it.
|
| AI isnt cost effective. The investors are going to want their
| money back very soon due to outside economic influences... they
| wont get it back and many of these AI pop ups are going to fold.
| the rest are going to scale back and jack up prices.
| vannucci wrote:
| Personally this is what I'm hoping for. Stories I read about
| services sold as AI turning out to be minimum wage workers
| tells me that as much as everyone thinks this is the dawn of a
| new age of hyperintelligent machines we haven't gotten as far
| as we wanted to as fast as we wanted, or hoped.
| theshackleford wrote:
| Given that incredibly capable models can be run on fairly low
| cost hardware, how will this really change anything?
| kristianc wrote:
| A huge chunk of online content (especially what ranked on Google
| )was already SEO churned sludge, and I'm not I buy the argument
| that elite publishers and creators like the New York Times, The
| Economist, and The Atlantic have ever really depended on Google.
| When the Economist sells itself to advertisers it doesn't talk
| about its web traffic numbers, it talks about the fact that it's
| read by CEOs.
|
| You're likely to see content creators pull their work behind
| access-controlled spaces (which might actually work out better
| than the current bargain of it being free but unreadable, recipes
| buried by long winding stories, etc). You might see the weird web
| emerge again as search engines are able to discover it under a
| pile of SEO sludge.
| sofixa wrote:
| The Economist and FT no, but a lot of the other more mainstream
| (read by a wider audience) media like Guardian, NY Times,
| Washington Post, Le Monde, Le Figaro, etc. depend a lot on
| Google traffic. There were numerous legal disputes over this
| dependence, how Google circumvented it for users (the quick
| answers that made it so a lot of queries were resolved without
| even needing to visit the source website), and profit sharing.
| kristianc wrote:
| You see I even disagree with that. People don't accidentally
| discover the Guardian, NY Times, Washington Post, Le Monde,
| Le Figaro via Google, their muscle memory is trained to these
| publications because that's where they go to get their
| opinions and worldview validated.
|
| Of course they can get that from ChatGPT too, but it hits
| different when you realise ChatGPT validates everything you
| say anyway.
| sofixa wrote:
| > You see I even disagree with that. People don't
| accidentally discover the Guardian, NY Times, Washington
| Post, Le Monde, Le Figaro via Google, their muscle memory
| is trained to these publications because that's where they
| go to get their opinions and worldview validated
|
| That's for daily news reading. If you search for news (like
| what happened with the Spanish/Iberian grid), you'd use
| Google. And you shouldn't use ChatGPT because it wastes a
| ton of resources to just hallucinate anyways, whereas a
| Google search gets you the direct links to the sources.
| ChiMan wrote:
| Seems possible that one possible unintended consequence of AI
| could be a rebirth of the Web as something closer to what we
| knew. Because why use search at all for general inquiry when AI
| can satisfy much of that?
|
| More critically, it's not hard to imagine that, with AI-boosted
| boosted coding, a thousand bespoke search engines and other
| platforms being just around the corner, radically changing the
| economics of platform lock-in. When you can build your own
| version of Google Search with the help of AI and do the same with
| social media or any other centralizing Internet force, then
| platforms cease to be platforms at all. With AI, the challenges
| of self-hosting could become quite manageable as well. And while
| we're at it, some version of the same, individual-centered
| computing economics on your own devices seems possible.
|
| In these senses, it's quite possible that Jobs's vision of
| computing as extensions of individuals rather than individuals
| being extensions of computing is again at hand, with the magic of
| self-curated order from a chaotic Net not far behind.
| dehrmann wrote:
| > With fewer visitors, Stack Overflow is seeing fewer questions
| posted on its message boards
|
| When you operate a community that's hostile to questions that
| have already been answered, are poorly researched, or are
| homework, don't be surprised when people start taking those
| questions elsewhere, and don't be surprised when they start
| asking their _good_ questions elsewhere, too.
| kmfrk wrote:
| Yeah, asking a programming question without some bitter old
| coder tut-tutting you is very much a selling point with AI
| chatbots, regardless of my reservations with the overall trend.
| jay_kyburz wrote:
| The sarcastic enthusiasm and fake humanity from the LLMs is
| wearing thin as well if you ask me.
| theshackleford wrote:
| The difference is I can instruct an LLM (and have) to knock
| that shit off.
| bilbo0s wrote:
| True. Humans don't take direction very well.
| bazoom42 wrote:
| Sure, but this is only possible because the LLM is trained on
| those answers.
| karaterobot wrote:
| Proving that AI is not just parroting back what it reads on
| the web, ChatGPT manages to correct my programming mistakes
| without making me feel bad. If it learned from Substack,
| I'm glad it learned selectively!
| iknowstuff wrote:
| It's an easy critique of stack overflow, sure, but the same
| applies to reddit tbh. It's quickly becoming far more
| worthwhile to chat with AI than get angry at stupid,
| predictably reactionary reddit comments - and you're not
| reaching many people, you're just used for training a model, or
| for advertising opportunities for sleazy subreddit owners
| mittensc wrote:
| Try and follow /new on raspberry_pi or similar...
|
| You'll see reason for the hate, mainly with people not
| bothering to spend any time searching before posting.
|
| And it is getting worse, new people asking help: 'but chatgpt
| told me X', 'I followed chatgpt and it doesnt work, please
| help fix bug', or some idiots that might burn the house down
| and deserve yelling (li-ion batteries aren't a joke, ac
| current likewise)
|
| Or... LLM generated stuff... which is equal to spam...
|
| If some people like doing unappreciated tech support all
| power to them, others might fight through spam to find nice
| items, I mostly stopped bothering and looking for something
| else. (also yelling at idiots that might kill themselves)
| LtWorf wrote:
| But without stackoverflow how do you think the AI will be able
| to reply about next year's new programming language?
| riku_iki wrote:
| LLM can give many answers using absorbed docs and codebase.
|
| Rest still could be asked/answered on SO or github.
| JKCalhoun wrote:
| > Openai is soon expected to launch a browser of its own.
|
| Is that right? I'm not sure how I feel about that. Actually, I
| think I know how I feel about it.
| PaulHoule wrote:
| AI _inference_ can save it because it can be used to make tools
| that reverse "enshittification" Cloudflare slams the door in the
| the face of this "exit".
| ppqqrr wrote:
| the web died years ago, for a different reason: labor monopoly.
| The web, and software in general, stopped reflecting or serving
| users, when the ruling class started pouring massive capital to
| dismantle any paradigm for major web/software development other
| than ones vetted by pedigree VCs and planned for obsolescence by
| acquisition. Gen AI is actually the only thing that could've
| punctured their hold on world software - the vague air of
| confidence they maintain when they talk about "their future of
| AI" is a facade. There's already a new Web (or Webs) coming that
| will dwarf the current one, and it won't be coming from them - so
| they're clinging to their only means of control: the AI token
| infra.
| righthand wrote:
| > There's already a new Web (or Webs) coming that will dwarf
| the current one
|
| Care to elaborate? What is this new web if no one is
| incentivized to publish, only consume?
___________________________________________________________________
(page generated 2025-07-20 23:00 UTC)