byebye AOL, long live usenet

[piranha]

AOL is discontinuing access to usenet. i'm seeing people celebrate the end of "eternal september". as if.

i, too, would like to go back to those days. but this is one of those "can't turn back the clock" things. AOL's influx at the time was bad for usenet, but it was certainly not the only bad thing that happened -- i think spam became much more destructive to it than AOL. its disappearance won't bring back the good old days. but neither will it become what the the author of the article linked above claims: But the Usenet will nonetheless become a smaller, less interesting place once AOL turns off its newsgroup servers. -- oh, bullshit. IMO when usenet was smaller, it was more interesting than it is now.

after more than 2 decades experience with online communication media, i think there is a certain range of contributors with which a community functions well, and a certain range of influx of newcomers that it can handle, and when those ranges are exceeded, the community suffers and becomes less useful. there's bound to be research by sociologists and maybe anthropologists about this, *heh*; i should look for that to see whether my anecdotal evidence is born out by it, or whether i am all wet.

Comments

[desh]

Besides, it's not like those AOLers will lose access to Usenet entirely. Far from it. I mean, I have a real NNTP account, and even I still use google groups sometimes...

[huaman.livejournal.com]

The sad thing is, around the time I stopped paying much attention to USENET due to all the other demands on my time, the AOL users were getting on towards being the clueful old hands.

I, too, would wish for a return to the USENET of days gone by, but the truth is, it's gone. I concur that spam hurt worse than eternal September, and that what made it really ridiculous to try to manage was the sheer volume. I think it's almost 5 years since I stopped having a newsfeed, and I haven't even bothered to preserve my home server's leafnode access. It's been over a year since I've even typed "trn."

USENET is dead. Long live USENET. If it comes back, and you hear, will you let me know?

"It's not dead yet!"

[brooksmoses]

"Nobody goes there any more; it's too crowded". --Yogi Berra.

I can't see how "it's hard to manage because of the sheer volume" means that Usenet is dead; it seems more like evidence that it's thriving. But, then, the Usenet that I know and love is still going fine -- and, yeah, there's far more stuff on groups that I'm interested in than I have time to pay attention to, but real life is kinda like that too. Maybe I just read different groups.

It would appear that alt.fan.wednesday, though, is quite distinctly dead, sadly.

Usenet vs. NNTP

[eagle]

Stepping back from subjective impressions and looking solely at where I spend my reading and posting time, my interest in Usenet in general is clearly waning. This is particularly true if you define Usenet as the public groups, Big Eight plus alt.*, since most of the attention I give to Usenet I give to private hierarchies that are much more interesting to me than what I can find in the public Usenet.

My interest in NNTP as a protocol and its associated toolset is, if anything, growing. It turns out to be the right solution to a lot of interesting problems that have nothing to do with Usenet as it historically has existed. It's far simpler than IMAP to write software for, the automatically expiring storage model is amazingly useful for some applications, NNTP support actually works better and more reliably in a lot of desktop software than IMAP does still (and probably always will, given the complexity of IMAP), and Usenet software is stable as hell and just works. I don't have the problems with my news server that we have with IMAP, where mailboxes occasionally just get corrupted and have to be reconstructed, or there are weird load spikes.

A news server is a great way to handle sharing access to mailing lists, particularly mailing lists that you want to peek at periodically but don't care about reading all the time. It's the best system for handling role addresses that I've ever seen, since everyone can review everyone else's responses and see the whole thread. I'll take a good scoring/killfiling newsreader any day over procmail to read a noise-heavy role address like postmaster; procmail may be theoretically more powerful, but I can write Gnus scoring rules far, far faster than I can write procmail recipies and I don't want to have to devote the local storage space to all that mail or pull it all down from an IMAP or POP server to filter out. Usenet is a wonderful way to build a viewable mail archive for mailing lists; I defy anyone to find a web-based mailing list archive that's as easily searchable and navigatable as a two-year accumulation of traffic in a newsgroup combined with a good newsreader. Etc.

So I expect I'll still be working on news software long after I've mostly stopped reading public newsgroups.

[novazembla.livejournal.com]

i think there is a certain range of contributors with which a community functions well, and a certain range of influx of newcomers that it can handle, and when those ranges are exceeded, the community suffers and becomes less useful.

Let's test this theory out by diverting all those AOL users to this blog instead.

Anyway, usenet pretty much outlived its usefulness, and good alternatives abound.

@%

Re: byebye AOL, long live usenet

[pir-anha.livejournal.com]

good alternatives abound? like, where? i have yet to find something that works as well as usenet. LJ certainly ain't it.

Re: byebye AOL, long live usenet

[brooksmoses]

You beat me to it.

A brief comparison of comp.text.tex and tex_latex would show that, at least in one case, Usenet is still distinctly where the community is.

On the other hand, I'd say that the Model Car Message Board proves that the community is a lot more important than the interface, for all that nntp is still the best interface I've seen for this kind of thing, and webboards generally are far from it.

Re: byebye AOL, long live usenet

[novazembla.livejournal.com]

That's funny; most of the time, when I've used Google to search for a clue to a TeX (or LaTeX or XeLaTeX or fontinst or whatever) problem, it's pointed me to a TUG listserve archive somewhere, not to any archives of comp.text.tex.

@%

Re: byebye AOL, long live usenet

[brooksmoses]

Hmm -- I suspect that that's more a factor of searching the web versus searching newsgroups; I don't think there are web-based archives of comp.text.tex (other than Google Groups), and so they wouldn't show up on Google unless you specifically do a newsgroup search.

That said, I do admit that my comparison is a bit limited by only comparing the Usenet and Livejournal communities, and leaving out the various other ones (e.g., from TUG mailing lists) that I'm not familiar with.

[king-tirian.livejournal.com]

I agree that Usenet has been more interesting than it is now, but I don't think that's a function of its population. Rather, IMHO, it was about its ubiquity. I mean, in 1990, if I was interested in General Hospital, there was only one place on the internet to go and talk to people about it, and I didn't need to be a rocket scientist to figure out that the address in the namespace would be something close to rec.arts.tv.soap-operas.abc.general-hospital. And not only would you know the way there, but you'd know that everyone else on the internet who wanted to talk about GH would be there too. Nowadays, if you wanted to find a place to talk about Neopets, you could go to alt.games.neopets, but an overwhelming majority of people are on an uncountable number of phpBB sites out there and there is no directory that would point you to the "good" ones.

Still, Usenet's death is not AOL's fault, either in 1994 or in 2005. Usenet's problems are the same ones as always. (A) It held to its pre-HTML standards to the point that people got sick of being flamed for their broken newsreaders and went somewhere else. (B) The forgability of headers similarly called for a change in standards to something more tamper-proof, instead of just writing cancelbots that issued another forged message for each spam message. (C) There weren't enough volunteers recruited and trained to keep up with the growth, and then there wasn't enough to handle covering the burn-outs. (D) I think it probably is a legal liability that an ISP stores a copy of a stolen binary for the advertised benefit of its subscribers, especially when Usenet propogation is the work of the ISP staff and not the private webspace of its members.

I'm not even saying that I'd fix these problems if I had a time machine and more knowledge and persuasiveness than I will ever possess. Products have a life-cycle, and it's pretty amazing that Usenet survived over fifteen generations of Moore's Law before I got pissy and declared it dead.

Usenet's problems

[eagle]

Well, I'll continue to maintain that Usenet's failure to quickly adopt a rich text format, such as MIME, had no significant impact on its success or failure. It may have had a social impact due to people being annoyed at the endless flamewars about it (and flamewars in meta-threads are one of the most obnoxious aspects of almost any forum and can drain the life out of it regardless of the subject of the meta-thread). But it wasn't something Usenet needed technically to survive.

I think you omitted the primary reason why so many people use web boards instead of Usenet -- control. You can't control a Usenet newsgroup effectively, which has implications all over the map from protecting the group from well-meaning but wrong-headed overprotectiveness to making it impossible to deal with certain types of attacks. Furthermore, that lack of control completely kills the interest of any commercial organization, since they rely on control to force-feed people advertising, to draw attention to the rest of their site, to closely associate the site with them and their brand, and all of the other things that commercial organizations love to do.

Some of these properties of Usenet are good. Some of them are bad. All of them combine to mean that the commercialized Internet was inevitably going to search for a different solution than Usenet because Usenet couldn't support the Internet business model. There was no effective way to add advertising (and even with HTML posts, there wouldn't be -- that wasn't the barrier), there was no way to brand and control a newsgroup, and there was no way to control the way that the newsgroup looked to all of your customers.

The result is that commercial interests, and the many people who are not commercial but who have similar interests in control and in egoboo from having a forum associated with them, went elsewhere, and commercial companies can generate a lot of interesting content when they want to, content that people really want to see. It is somewhat more logical to go to a support forum run by the company whose product you're using, or to a discussion board tied to the vendor from which you're buying the product you're discussing. Add in the ubiquity of the web browser as effectively the sole user interface for a large percentage of current Internet users and the lack of good support for newsgroups in most web browsers for a very long time, and you get a significant attention drain away from Usenet.

I don't think any of the problems you named above had any significant impact compared to that one. The volunteer situation has never really been as bad as it looked; the problem was more getting the volunteers into a position to do what they wanted to do. The legal liability has, in my opinion, been massively overstated and pretty much cut off at the knees at this point by the safe harbor provisions of the DMCA (note that even Ellison's case is proceeding solely on a technicality). And while Usenet does indeed have a forgery problem, so does e-mail, and I've not seen any drop-off in the use of mailing lists; if anything, they've significantly increased, and have surpassed Usenet in some areas as the standard location for in-depth technical discussion. E-mail also has spam problems that dwarf those of Usenet; we actually beat spam on Usenet, and on a well-run server you hardly ever see any. But e-mail is thriving.

I think people's desire to send pretty messages is highly overestimated. Most of the HTML e-mail is actually less readable than the original text message. There are specific cases where it's useful and where people actively use it in a way that makes messages better, namely specific announcement lists or periodic updates sent by usually web-based sites, but that was never Usenet's primary niche anyway. In general discussions, which is where Usenet always thrived, it is in practice nearly worthless.

Anyway, just my two cents. I don't think AOL dropping Usenet will make a significant difference; the presence of AOL on Usenet has been massively declining for years now. What would really make an impact would be Google Groups going away (and I would argue that in a lot of ways, it would be a substantial improvement).