Why did Usenet fail?

This is annecdata - not a serious academic study. Adjust your expectations accordingly.

When I first got online, the World Wide Web was still in its infancy - so CompuServe was my gateway to the Internet. I loved their well organised chat room. A couple of clicks and I could be discussing Babylon 5 with people in another country, downloading wallpapers, and uploading my poor attempts at fan-fiction. Magical.

At some point, I discovered Usenet and its decentralised collection of newsgroups. This required downloading, installing, and configuring an NNTP client. Quite the world away from CompuServe's shiny software delivered on a free CD. But I persevered and, by the time I went to university, understood all the intricacies of PLONK, ROFL, and Godwin's Law.

My university experience was dominated by Usenet. It was where nerds went to socialise. I bought and sold computer equipment, published terrible poetry, and learned about LGBT matters. I lurked in the comp.lang.* hierarchy until I was confident enough to ask my Prolog questions without making it look like I was asking for help with my university assignments.

And then, one day, I just stopped.

There are three main reasons that I remember.

First was moderation. Sites like Slashdot were - at the time - pretty good at moderating out crap comments. Usenet was a stream of consciousness and, short of blocking individuals, there was no way to separate the interesting topics from the dull. Without upvoting and downvoting, it became tedious to read. And there was very little feedback on whether people found your posts useful.

Secondly was reputation. Sites like AVForums and eBay allowed you to see who was a reliable buyer and seller. Every transaction on Usenet was a risk. You might send money or goods to a scammer. Look, eBay wasn't perfect - but it made it easy to see how many successful transactions an account had made.

Thirdly was the user interface. Usenet looked dull. In a world of animated GIFs and MySpace colour schemes, Usenet didn't even have avatar images! Sure, the spartan nature meant that you could focus on a conversation - but it didn't feel as modern and exciting as the web did. NNTP software was fragile.

And so, slowly but surely, I drifted away from newsgroups into the tender embrace of web forums, Reddit, and Slack. All of them centralised (booo!) and privately owned (boooo!) but all having a nicer UI and UX (yay!)

I should briefly mention Google buying the Deja News archive, promising to revitalise Usenet, and then promptly abandoning it. Cheers Google. Choogle.

While Reddit goes through its "are we the baddies?" moment of banning all 3rd party clients and fucking over its unpaid labour force, I thought I'd return to see if Usenet was still a viable option. There was only one modern newsreader listed by my Linux distro - Pan which hasn't been updated in a year and has a GUI which gave me instant nostalgia. That's not a good thing by the way.

I couldn't find any Open Source Android apps for reading newsgroups.

And, as it happens, my ISP killed off their Usenet server a couple of years ago.

As I've said before - Slack is just IRC with a better UI. Dropbox is FTP with a better UX. WhatsApp groups are just ListServes for people who don't know how to configure a server.

Reddit was Usenet with a sparkly front-end.

User Experience matters. That's why Usenet lost. It was hard to set up, there was a ton of terminology to learn, sticky posts with group etiquette didn't exist, trolls and grieffers couldn't be moderated away, and the whole thing looked like a 1990s shareware accountancy package.

I'm a little sad that Reddit is further enclosing the commons. And I doubt this will lead to a resurgence in Usenet. But I hope it will give open source and open standards developers a little jolt towards designing user experiences which are fun and easy to use.

Share this post on…

27 thoughts on “Why did Usenet fail?”

  1. Alex B says:

    USENET failed? I was still an active user until the late 1990s, by which time it was nearly 20 years old. Many "successful" projects can only dream of a failure of that magnitude!

    And, I'm sorry, but I find the idea that Reddit's UX is better ridiculous. I only use Reddit because it's the closest thing that still exists to USENET, but with a larger userbase. It's hard to search with precision, upvoting gives undue weight to popular-but-incorrect viewpoints, and moderators are unresponsive, capricious, and unaccountable (give me my personally-maintained regex-powered killfile, any day).

    The only thing Reddit (and other privately owned online social media forums) have going for them is an audience. People go where people will read what they have to say, and where they can find other people with interesting things to say. As new Internet users gravitated to new platforms, so even us from the old guard found that the interesting conversations were no longer happening on USENET, and so its userbase declined, perhaps terminally. Given that decline, it shouldn't be a surprise that there hasn't been much innovation in FOSS NNTP clients. I will remind you that Thunderbird can still use NNTP, though!

    1. @edent says:

      I think you've proved my point there. People flocked to things like Reddit because they found it easier to use. And power users like us built tools to consume its API in a way that made sense to us.

      You may enjoy maintaining a killfile, but most people don't find regex particularly user friendly.

      And, as it turns out, many people rely on capricious mods (of whom there were none on Usenet, I'm sure...) to block things on their behalf.

      The customer is always right in matters of taste. And, for most people, Usenet left a bad taste in their mouth.

      1. Alex B says:

        By the late 90s, the TV-ification of the Internet was well underway, and I don't think many people even tried USENET to discover if it tasted good, or not.

        The nice thing about regex-powered killfiling, is you don't have to use regexes, but the power is there if you need it. TBH, I never needed a killfile much, because the audience (of the groups I followed, at least) was better-behaved than a large proportion of the participants on Twitter, Facebook, and Reddit today.

        And whilst there were certainly capricious USENET mods, they couldn't prevent someone from setting up alt.pizzas.unmoderated or whatever. Reddit's flat hierarchy means people are unlikely to discover better-moderated versions of subreddits that are badly-moderated, but have the "obvious" name.

  2. says:

    Usenet failed, for me, due to a combination of (ISPs include any 3rd party feed providers):
    1) volume of the feed (meaning ISPs were finding it harder and harder to maintain an adequate copy due to the size) leading to fewer feeds being available for users. Initially just alt.binaries.* got omitted due to size, but that grew due to point 2.
    2) moderation/spam/copyright issues: most of Usenet was unmoderated and spammers found it more efficient than email to send out their messages leading to many groups being overrun. The lack of moderation also lead to groups being "repurposed" by people wanting to spread copyrighted material leading to lots of groups being blocked and then ISPs going "We can't keep manually removing messages blocking individual groups, so we are blocking all alt.* for starters..."
    3) ISPs took very little action against USENET spammers themselves leading to posts from their subscribers being dropped by various other ISPs without knowledge of the drop being visible either to the poster or the ISPs customers
    4) The factors above leading to USENET getting more splintered: you might not be able to see the groups you want, conversations may happen without you being aware due to the filtering. This lead to fewer consistent users especially with the rise of the web with specialist "low-spam" (at the time) moderatable bulletin/message boards acting as a "pull factor".
    5) The cost/effort vs customer demand for ISPs meant more incentive to just stop offering the service pushing the few remaining consumers towards 3rd party providers (many of which were run on a low cost/fee basis which attracted spammers leading to higher costs for that provider and then being blocked: so they discontinued offerings).
    6) The user experience of USENET, even at its peak, wasn't beginner friendly (those people complaining about picking a Mastodon server being anti-newbie friendly would have big issues with getting started on USENET) meaning there wasn't the takeup of new users to replace those leaving.

    Just my 6 cents! (and me regretting reading this post on mobile and replying via mobile!)

  3. John says:

    There are free places to go, as you may know, one example:


    And I still find good posts on USENET even now, but yes I have a rather large killfile. But regex in killfiles depends upon your client. In tin(1), when you kill something you are prompted to how to kill. No knowledge of regex required.

  4. said on m.blank.org:

    @Edent @davidgerard a few thoughts from someone who was all too there at the time:

    Spam expands to fill available space, and this interacted very poorly w/ usenet's lack of anything other than client-side moderation. getting value out of usenet took ever-increasing work for everyone, and everyone had their breaking point

    (Okay, slight lie: Usenet did have one server-side moderation mechanism: CANCEL messages. Boy oh boy did that scale extremely poorly. Hilariously poorly.)

    Reply | Reply to original comment on m.blank.org
  5. said on emacs.ch:


    #Usenet​/NNTP, apart from some non-destructive updates, still is Usenet/​#NNTP.

    The web was meant to be the union of documents in the internet, no matter which protocol or text format. The real browsers were able to access more than only HTTP(S).

    The web has turned into a single protocol party. The diversity has been restricted to document types. Current browsers are divisive (even the smolnet ones, but that's a different rant...), old browsers were inclusive. And lots of security disasters were added under the hood of bling-bling.

    So Usenet/NNTP did not improve much, but did not lose either.

    The web is not any more what it was meant to be, gained bling-bling and dangers and loses on the original goals of connecting everything.

    So who did fail?

    Reply | Reply to original comment on emacs.ch
  6. said on m.blank.org:

    @Edent @davidgerard 2. The core technical contributors continued to treat the shibboleths and ad-hoc preferences of early adopters as holy writ even after the userbase expanded 100X. There was incredible resistance to client-side MIME support until netscape forced the issue (just for example).

    Everyone joked about “eternal september" but in retrospect that was the dying reflex of a bunch of poorly socialized nerds (I very include myself) who saw control of their little fiefdoms passing away.

    Reply | Reply to original comment on m.blank.org
  7. said on m.blank.org:

    @Edent @davidgerard 3. Turning NNTP into the world's largest distributed repository of pirated TV and movies may, in retrospect, have had some downsides. To this day I'm genuinely shocked that more providers didn't drop and block the alt.binaries hierarchy way earlier than they eventually did. Not that I'm morally opposed to media piracy in a lot of cases, but as early as 1998 every Usenet admin was spending the majority of their time trying to manage things that were not conversations.

    Reply | Reply to original comment on m.blank.org
  8. foul says:

    I loathe the constant need for stimulation that "winning" internet shops put on sale. I really loathe the user. You're right nonetheless and this post of yours is insightful in case someone would like to make a(nother) Qt or Electron app for Usenet or IRC.

  9. said on chaos.social:

    thanks for the interesting post @Edent

    Reddits current plans are another reminder that centralized proprietary service providers can't be trusted.

    Unfortunately that is where the majority goes, because it is simpler.... and everybody else is already there, too...

    looking at facebook (in the past), Instagram, WhatsApp, Discord... etc.

    What i love about reddit is, that they have a multitude of topics i can discuss with one account. Lemmy won't be that I'm afraid.

    Reply | Reply to original comment on chaos.social
  10. dgold says:

    Usenet "failed" because it was a country club. It relied on its strangeness, its difficulty cliff of engagement, and its self-selection of users to keep it "safe". I loved the place (I still mourn for a.f.p.) but even then I could see that the place glistened shiny white and wealthier.

    When it was opened to the "masses" (more primarily white americans privileged enough to own both a computer and a landline) it failed almost immediately.

  11. DinoNerd says:

    I must be older than you; when I got online, CompuServe(*) wasn't even a gleam in an MBA's eye. My gateway to the net - still not the Internet - was a modem and the UUCP protocol. Somewhere upstream was the backbone, which was still pretty close to being the ARPANET. My workplace - a small Canadian branch of a big German company - got online first. We all loved being able to communicate with researchers in e.g. Israel with only a day or so for the message to arrive.

    This was email, but there was also Usenet. I loved Usenet, and became increasingly sophisticated in my choice of client software. The best client was trn, which presented messages in threads, and allowed you to mark uninteresting threads to be ignored, as well as blocking specific annoying posters. IIRC, you could also use regular expressions to select what to ignore - or what to read.

    Like you, I also haunted parts of the comp.lang.* hierarchy, picking up expertise on sometimes extremely arcane details, though my language of choice was C, not Prolog.

    Later I acquired my own home system, with its own modem, and a place in the UUCP network. I downloaded and built various software packages. (It took days to download the complete source tarball for X11.)

    Time passed, and the Internet appeared, with nodes mostly connected at all times. the modems I used to connect to internet nodes got a lot faster. Eventually I got DSL, and joined the always-connected set.

    Meanwhile, Usenet got larger and larger, with an ever poorer signal to noise ratio. Most of the groups I frequented by that time were moderated - too many people were sending out too much spam to all the others. But what killed it for me was a policy decision by my then employer, that charged for headers downloaded - which trn had to do in order to reject an article. (No protocol was available for it to ask its server to send only articles matching a specific pattern.) My use of trn was causing a really large spike in the internal funny money charged to my department.

    I eventually got access to Usenet again via an ISP. But by then the bad had mostly driven out the good.

    However, the interface for trn was never an issue for me. What's not to like about an interface that does the job really well, which has stayed the same long enough for me to become an expert user? Even another equally effective interface wouldn't be as good while I was learning it.

    FWIW, I still judge interfaces to forums of all kinds by the trn standard. So far, none I've encountered have the same flexibility and power.

    (*) Actually, Wikipedia claims that CompuServe "dominated the industry during the 1980s" and was founded in 1969. I got online in 1985 or 1986. So this claim is totally wrong, even though I doubt I'd ever heard of it in the '80s.

    1. fredcy says:

      I too was an avid trn user back in the day. I started using "netnews" in Bell Labs back around 1982 and wasted far too much time over the years clicking through the groups that I followed. I seem to recall an earlier 'rn' client, later improved by trn. Trn's ability to kill entire sub-threads was brilliant -- I wish Hacker News had such a thing now. I don't recall just why I stopped participating in usenet. I think my ISP's support for usenet got increasingly weaker and the content was ever noisier.

  12. said on infosec.exchange:

    @Edent A big part of it for my friends and I was accessibility.

    As it slowed down it quickly got monopolised by a handful of shady providers pushing the unspoken piracy angle, and the ‘public’ ones soon dried up as the ISPs got fed up of constant DMCA takedown requests and other pressure. It also took not insignificant resources to host and as you mention, what little moderation there was also faded away.

    For myself and my close circle, we were Amiga users (for our personal computers) quite far into the early 2000s, and still routinely used Usenet over BBS/IRC and other web based social platforms, simply because the client support was (mostly) better.

    Even as late as 2015, the likes of BT/Plusnet etc. still hosted their own Usenet servers as a legacy hangover but eventually pulled the plug. Their documentation is still live afaik.

    I’m not on the ‘bring back Usenet’ train as it’s largely run its course, but I do remember my time with it fondly.

    Reply | Reply to original comment on infosec.exchange
  13. Tony Stark says:

    Usenet never died, everyone forgets basic electronics

    Two or more computers connected in a network become and behave as a single device. Everyone in the tech industry has wanted to kill piracy and move us towards mainframe computing since the 60's. Under copyright software is licensed, so you don't own the technology.

    Media and tech companies got together and formed the trusted computing initiative, it's a bid to remove plaintext binary access and root access to digital devices from all users, since they know the public is stupid and illiterate.

    There 25+ year quest to kill the PC is coming to fruition in windows 10/11/12. You no longer own your PC.

    AKA secure boot was an attack on binary plaintext exe access of your device.

    They perfected trusted computing using consoles and mobile phones, once the iphone hit in 2007 they've had more then a decade to perfect TC tech.

    It's the fritz chip on steroids.


What are your reckons?

All comments are moderated and may not be published immediately. Your email address will not be published.Allowed HTML: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong> <p> <pre> <br> <img src="" alt="" title="" srcset="">