[ Home | Articles | Account | People | Projects | FAQ ] 
Back when I first discovered the internet archie was a great search tool and tin was the best internet app I'd ever seen. Back in those days (about ten years ago) no-one ever spoke about the `free software community', but there clearly was one. Nowadays everyone seems compeled to muse on the nature of the community, but it seems to me that it is now less of a community and more of a clique.
In the early nineties the free software community was just a lot of people who shared thier software patches over USENET, and USENET was the keystone of that community. Nowadays however sites like slashdot seem to be the keystone, and the differences between USENET and WWW bulletin boards is, it seems to me, both disturbing and destructive.
The main difference between USENET and bulletin boards is that every bulletin board (advogato excluded) has an entry point that is strictly controlled by the site managers and therefore gives enourmous controll over opinions expressed by those managers. The only threads started are those given the OK by the managers, and typically every thread starts with a comment by the managers, giving their opinions much greater significance than anyone elses. Bulletin boards are really little different from the standard media and are therefore anathema to the distributed, anarchic environment that the internet allows and USENET impliments.
Another problem with bulletin boards is the short lifespan of threads. Because once a thread starter dissapears off the main page the thread dies, the average lifespan of a slashdot thread is only a few hours. This is hardly a good environment for rational, thought out discussions. The only talk that goes on in such sites is merely a statement of each side's opinions without any real dialogue. Advogato manages to avoid this problem by having a relatively low turnover of new articles, but I think there is a problem with the design itself - USENET has no such problems with high volume news groups.
Another problem that I perceive is the way `anonymous coward' accounts are managed on slashdot. When pinet was still active each anon account was given a unique identifier so that you could kill file the idiots while still getting posts from people who, for whatever reason, wanted to remain anonymous. The slashdot system excludes the implimentation of a kill file system that could reject specific anons.
Finally, the internet allows for one to have a completely distributed design environment, but bulletin boards have strict controlls over what software is used to controll the threads.
When the WWW first appeared we had a great oportunity to create something much better than USENET but we failed and ended up with something much worse. On USENET no-one has controll over the starting of threads, no-one's opinions are given special status and threads can go on untill it dies a natural death.
With the dominance of slashdot we now have a clique of Malda/Raymond followers whose opinions are the only ones picked up by the general media. The rest of us have our opinions sidelined into outsider status - Hardly what could be described as a community.
No development work and little communication of any value to the free software community happens on slashdot. When was the last time you saw a thread on slashdot where someone posted some code, someone else refined it, and it was polished up and released as free software? When was the first time you saw this? I've never seen it happen on slashdot. That's what mailing lists are for.I think you miss when you say free software development moved from usenet to web-bulletin-boards. The real move was from usenet + mailing lists to mostly just mailing lists, with a smattering of web sites to provide news. You may be right that this encourages cliqueish behavior. But if you want to find people posting code, commenting on it, and making something happen -- free software development in its essence -- look no further than the linux-kernel mailing list, the debian-boot (and debian-devel on its good days) mailing lists, bugtraq, mutt-dev, the zillion mailing lists hosted at sourceforge. The list goes on and on.
Your mention of the general media has some valididy, but Rob and co are hardly the only source of opinions I've seen in the media. Stallman has gotten an equal amount of press coverage for instance, as far as I can tell. And the press always distorts things; they're guilty of every sin you ascribed to slashdot.
When the WWW first appeared we had a great oportunity to create something much better than USENET but we failed and ended up with something much worse.No argument there. I'm holding out some hope because of the WikiWikiWeb and emerging sites like advogato, but the web is indeed a failure so far.
The best free-software discussion scheme I've seen yet is mozilla.org's bi-directionally gated mailing-list/newsgroups. Mailing lists seem to be very popular for free software development but I find them obstructive - the process of subscribing is time-consuming, and then for sanity's sake you need to create a folder for each list you subscribe to and set up a filter rule, and then set your mailer to sort that folder threaded, and even then you can't browse past discussions to make sure your question hasn't already been asked. Sometimes you can find an archive, but it's a real lottery. Since a lot of the time you'll only want to get involved in development of something in reaction to some particular problem or potential improvement, this is a high barrier to entry.
Newsgroups are much more convenient as you can subscribe and browse past discussions without any commitment or cost. Your newsreader is probably already automatically threaded and there is nothing else that needs to be done to join. There's no artificial barrier to getting into a particular project's community.
What I find distressing is that there seems to be almost no desire by anybody to gate any project's mailing lists to newsgroups in this way. As far as I know even Sourceforge, which has the ability to do so if anyone does, isn't doing this. I really can't figure out why...
The final step would be to write a web-nntp gateway that would make it possible to read discussions usefully on the web, so that whatever medium someone prefers, they can use. If you really wanted, you could make that gateway look like slashdot ;)
I like the idea of gated mailing lists <-> newsgroups; I'm on a few. But the reason that usenet newsgroups are no longer being used as nicely as they once were is twofold:a) news readers are more difficult to use
b) usenet has been overrun by spam
Item a can and would be be dealt with if it weren't for item b. Item b can't really be helped. Usenet is a completely open system full of tons of email addresses willing to receive spam as well as having a large audience for posted spam on all non-moderated newsgroups.
Setting up your mail program to send out foolproof no-spam style email addresses is about as tough as setting up your mail client to filter mailing lists into seperate mailboxes (if even possible in many cases where mail gateways or firewalls prevent such things). On top of that, not all people filter their mail strictly by mailing list but either let it all go to an inbox or to some semi-topic related mailbox (such as all freebsd lists in one box, and all linux lists in another).
Practically no serious work gets done on Web discussion boards. About the only contribution is that a discussion may spark a idea that gives birth to a project; the Free CDDB projects started from a Slashdot thread, for example. I'm doubtful that things like Wikis can ever approach the ease of use or flexibility of an archived mailing list.I only got on Usenet around 1991-92, and don't know if there was ever serious code discussion on it, with deep technical discussion and patches flying around in comp.* newsgroups; can anyone who was around confirm or deny? (Oh, maybe some of the gnu.* newsgroups that are gated to mailing lists, such as the ncurses list, would have detailed discussions.)
I suspect the new development is the fragmentation of the community. I wasn't around then, but it seems reasonable that in the 1980s/early 1990s the community was small enough that everybody knew everybody else, or at least knew who they were. (Again, anyone who was around then care to comment?) Today that isn't the case; for example, I know very well who's who in the Python community, but couldn't name any of the current primary developers of the GIMP or bash or the various KDE apps that I use. I suppose it doesn't matter much; each group will just trust the other groups to do their best.
(Oh, and Mailman, for one, has a mailing-list/NNTP gateway that's used for comp.lang.python; it works reasonably well, despite the occasional hiccup or misconfigured news server.)
I wasn't just concerned with code discussion - That does only occur on mailing lists. But the discussion of the wider issues only took place on USENET and it seems to me that the opinions being expressed has narrowed since the advent of bulletin boards.One thing about code discussion on USENET that has been lost is that people used to post patches for emacs etc that made it work to thier own particular quirks. It was a great way to find patches that made programmes work the way you wanted them to. This seems to have dissapeared completely and we now only get the official versions. This could be easilly fixed though by having a freshmeat style site just for patches that don't make it into the official releases.
I'd disagree about the patches to some extent. Just take a look at the tons of patches for the linux kernel out there, lots of them in common use by lots of people. Sourceforge also supports people uploading their own unofficial patches.We may also be seeing a maturing of the development model, where more patches get sent upstream and integrated, with further 3rd party patch integration being done by the various distributions. For example, while I maintain 1 personal patch of a program, I have sent hundreds upstream, and maintain several dozen different patches to the debian packages I am resonsible for. After all, large numbers of third party patches are not very sensible in the long haul, so there is pressure to get your patch in upstream if you can.
You mention that opinions seem to have narrowed -- I think that the opinions you read on slashdot are indeed narrowed, to the point of being caricatures. Issues arn't as black-and-white as a slashdot discussion will paint them. A certian amount of sensationalism and simplification seem to, sadly, be the way to make yourself heard on slashdot. The opinions expressed on mailing lists contain more shades of grey and more thought.
You're probably right about discussion of wider issues. When all valuable discussion happens on focused mailing lists, discussion of wider topics is either off topic and can't be found by a wider audience, or doesn't happen at all. What can we do about this? Advogato may be very helpful in that regard, if it scales.
XML may be giving us a window of opportunity to fix things up. With a server like jabber you could use any client you want for reading and have the client use XSL to impliment kill files and so on. This would make a web bridge through apache fairly trivial and you could use your own personal XSL doc to get rid of people and threads that you aren't interested in (don't flame me if this can't be done - I only work with plain XML).I think one of the most important changes that could be made to advogato would be to add the ability to make threads sticky so that they hang around for those who are still interested in the discussion. This would allow a thread to pop again when someone new discovers it as often happens on USENET.
Are there any free (speech) nntp<->web gateways? Last time I looked I didn't find any (and was somewhat surprised at this). It seems this would be a valuable thing to have; if there aren't any, would members of the "advogato community" like to collaborate on one?
Thanks for a thought-provoking article. I agree with many of your central arguments. Here's my own take on the issues.I think splork has it right when he says the problem is spam. There was a time when Usenet wasn't overrun by spam, but that was back in the days when it was nearly impossible to get Internet access unless you were a student at a major university, and not easy unless you were in computer science. That was simply the first of many walls used to protect a community. There's absolutely no way we can go back to those days, and I'm not sure we'd even want to - even if you believe that was the golden era of Usenet, you have to admit a lot of people were excluded.
Now that everybody and their scam-artist brother is on the Internet, the walls have to be built differently. In particular, they have to explicitly exclude the unwanted, rather than implicitly through difficulty of access. I guess the main exception is a group that is only known by a small group of "insiders." However, such a group always is always in danger of being overrun, especially as it grows.
Now that the decentralized Usenet is being supplanted by the increasingly centralized Web, the main model is that of ownership of the Web site. As mettw points out, the site owner generally gets to choose which articles get posted, how the discussion is structured, and (to a much lesser extent) who gets to participate. Because of this centralized nature, everybody wants to own a chunk of the good stuff. This, I think, is what bothers me about services like SourceForge. Yes, it's a wonderful service to free software, and appears to be quite open to all legitimate users, but at the end of the day, it's owned by a publicly traded corporation, who after all wouldn't be running the site if it weren't to their business advantage.
Usenet, by contrast, is not owned by anyone. I remember a number of anecdotes about clueless people unhappy with things posted to Usenet, asking "the owner" of the newsgroup to do something about it. Those of us in the know sat back and chuckled, smug in our knowledge that Usenet belonged to all of us. Of course, that's not true these days. All Web sites (including Advogato) do have an owner.
I've been thinking about these kinds of issues a lot, and have been mulling over a modest proposal. I'm suggesting that we build a new infrastructure, totally decentralized and free of centralized ownership (like Usenet, Gnutella, and FreeNet), yet confronting the issue of spam head-on. I think the group trust metric developed for Advogato is a very interesting step in this direction. While the original motivation for the idea was to build a better PKI, I think it may find its most interesting application in authenticating messages in spam-resistant communication systems.
While the Advogato group trust metric is currently instantiated inside a centralized Web server (thus allowing me to control the trust root, and thus ultimately who gets accepted and who doesn't), I have great hopes that it can make the transition to a distributed, decentralized system, in which everybody gets to decide their own individual trust root.
I really don't know whether this modest proposal will work out in practice, but I do think it's worth trying. Maybe in my copious free time, I'll try building a prototype.
I can't see how spam is the problem. Pulling up my newsreader right now and making an unscientific survey, spam is an insignificant part of the newsgroup traffic: comp.lang.dylan: 50 articles, 2 spams; comp.lang.python: 165 articles, 3 different spamsChecking some other groups: comp.lang.perl.misc: 361 messages, 0 spams (does someone cancel them?); comp.lang.c: 2216 messages, around 50 spams (2% of messages)
The problem isn't spam, since spam makes up a small part of the volume in those groups. The problem is the nature of the discussion in those groups, and the scale of it:
- The comp.lang.dylan volume is mostly a low-grade flamewar crossposted with comp.lang.lisp.
- The other 3 groups are full of mostly newbie, or at least low-level questions. comp.lang.c and comp.lang.perl.misc suffer very badly from people who can't be bothered to read manuals.
- comp.lang.python is a bit better, and there are a very few interesting technical threads such as a discussion of aspect-oriented programming, but little of it is concerned with improving the interpreter or the library.
I entirely agree with amk. With spam ebbing recently I restarted reading of comp.os.linux.* and noticed that several second tier Linux developers post there (notably Andi Klein and Jeff Garzik). But boy, are they difficult to find among "my cable does not work" posts.I strongly disagree with the guy who said that newsreaders are hard to use. I participate in a standalone newsgroup for Air Warrior players, posters to which come from all ways of life: oil drilling engineers, pizza men, cargo pilots, teachers, taxi drivers, photographers, rock musicians, carpenters. Most of them use Forte Free Agent or Openlook. One gentleman, a 80 year old WWII veteran, uses Quarterdeck News Center (I never heard of such software. Did you?).
What raph says is pretty cool. But I do not know anyone who can sit and implement that.
I used to have an opinion similar to yours about the free software community. That it had become mostly a bunch of poseurs. That 'free software sites' were being jammed with nothing but hits from Windows boxes by people who were interested in having the social 'cool value' of being associated with the community without any of the effort.And there are indeed a shitload of people out there that are like that. But the people that you are lamenting the lack of are still out there, still submitting software patches, still working and learning. It's a shame that many forums have become all noise and no signal, but it hasn't eliminated anything, just diluted it quite a bit.
The community is what you want it to be. There are parts of it that are horribly cliquish, and parts that are the same as they were in the early 90s. Maybe you just need to start looking for different 'hangouts' so to speak. I find that technical conferences are good places to go, because generally the people who are only in it for social value don't take the time to go to the conferences where there's nothing but a pack of nerds sitting around discussing extremely arcane details of various programs.
As far as the circles I travel in, slashdot hasn't been the 'center' of the community for the past 1 1/2-2 years. It's still an interesting site, but it's WAY to high-level to be interesting as a 'community' hangout. They talk about perl, they talk about linux, but it's always at a level bording on marketing information rather than technical nitty-gritty. There are plenty of other places out there, you just have to look hard. :)
USENET is nice, but it although it may scale to large numbers of people it doesn't scale too well to a large diversity of people. I used to read soc.support.depression and at one point the regulars on that group started a minor flame war with some alt.depression people who came over and started giving a big (((((((Hug))))))) to every poster who asked a question. Now that might be welcomed by people who are really emotional and tender-minded, but the regulars at soc.support.depression weren't like that and didn't like the idea of their little newsgroup being turned into that sort of a group.With everyone getting on the internet we need some way to controll membership of these groups so that you can hang out with people like yourself. Every realy world club has some way of doing this, the Hell's Angels having the ultimate of making people serve as pledges for atleast six months to weed out the posers from those who really are ultra tough-minded.
Now that I think about it, the trust metrics used on Advogato might be a great way to do this.
My main reason for posting the article though was that groups like slashdot are easy to find whereas a good newsgroup is not. So slashdot gets all of the media attention despite the fact that its poor design means that no rational discussion goes on there. With XML we should have a chance to design a better USENET that also has the ease of locating a website. I think the WikkiWikkiWeb idea could make a good contribution to the new system as well - although maybe only in those cases where a group is actually trying to reach some end goal.
I have no particular opinion on it, but some people including Peter da Silva have tried to come up with a better Usenet than Usenet, under the name of Usenet II, trying to reclaim the lost glory of Usenet from spam and such. People seem to have very strong opinions about Usenet II, one way or another.I love the feeling of a small BBS (such as I used to call when I was 12 or 15), but that's not perfect for much of what people want to do with computer communication.
It's funny that people use computers to talk to people about the best way to use computers to talk to people -- but it makes sense to me.
Another thing that has changed is that advocates of free software have become much more vocal in seeking mainstream attention, including management, industry, and media attention. Richard Stallman and Eric Raymond are most famous for their efforts to spread a message actively to those who were completely unfamiliar with it (even though their messages are slightly different). But previously free software had very little effective public outreach or recruiting.Nowadays there is much more of a free software movement, and many more people who identify themselves as part of a movement. There is a real growth of interest in activist and ideological stuff, and in public relations. This has really irritated some people, even including some activists. The results of this stuff are very mixed, but really quite amazing -- I remember when I was amazed if any newspaper or magazine ever mentioned Linux at all. Now the free software movement, in its various forms, is influential and interesting in parts of the "outside world".
Is this a distraction from the important business of writing and sharing code?
I had a look through USENET II, but it just seemed to be the same old USENET with a few new rules. Thinking about the trust metric idea a bit more, an idea that occured to me is that we could have semi-moderated groups where if your level is, say, apprentice or greater then you can automatically post a message, but for those on a lower level you would need to have your post approved by a master before being allowed.
Ever since I found out about CritLink, I've thought about how it (or something similar) could be used to track threads that span potentially many Web sites. Using an annotation system, you could post your responses on your site, avoiding "content serfdom" as Raph calls it. Eventually you'll run into the problem of too many annotations cluttering things, so you can use a Web-of-trust like Advogato.
To schoen: the main premise of your post seems to be that the change of behaviour of free software advocates distract people from the important business of writing and sharing code. You might think that you can redefine the meaning of words at your will but geek-style writing and sharing of code was never important in the free software movement. Just not to confuse matters: when you refer to "free software" you refer to FSF and the philosophy created by RMS. When we agree on that there is nothing that RMS says which places emphasis on the beauty or the amount of code. It's all about the freedom. In that respect nothing has changed. RMS has always promoted free software as an important social idea not the way for making hackers happy. The only thing that has changed is that now much more people are paying attention than back in 1984.In general: don't you guys go overboard here? The title of the article should be "USENET was much better in the old days". On what grounds it pronounces the death of free software community I fail to notice. I'll really spend my time tearing this post apart. So let's begin our journey
What is written: "in good old days there was no public notion and awarness of free software community but the community clearly was there". What it says: "I don't really have an argument about existance of free software community at that time so I'll add gratiuitious word "clearly" so that everyone sees how obvious what I say is and accept my arbitrary opinion as a fact". "Now community turned into a clique"==arbitrary opinion without any proof, how do you define clique, how does it manifest, who are those people that suddenly turned into a clique? "back then free software community was people throwing patches and themselfes using USENET". So now I'll actually define the community as I please so that I can complain later that it turned bad. "USENET ruled now slashdot rules and it's bad, disturbing, destructive and we're all gonna die". Does it appear to the author that slashdot community is just different community from the USENET community, maybe it's the community to which metw doesn't belong, but hey, find (or start) your own community. Slashdot community is not by any means free software community ("news for nerds" not "news for free software zealots") so no claims about quality or behaviour of slashdot community bear any relation to free software community. Then we have technical analysis of real and imagined weaknesses of www boards vs. usenet (which also tries to paint a false image of total censorship of www boards without giving any proofs). Those points would have some merits in the "Social and technical comparison of www boards and USENET" article but not "Free software movement is dying" one.
But the real problem and raison d'etre of this article is in the last paragraph where the whole truth is being told and bold statements are being made. One needs not be Freud to see that the only problem mettw has is his ego and the fact that no one cares about his opinions. General media and some undefined clique follows this... this Malda (creator of one of the most popular web sites out there) and this other guy Raymond (is he the one who coined term Open Source, published a few highly influencial articles and subsequently promoted Open Source?) instead of mighty and opinionated mettw, a-guy-who-hacks-on-3-programs. Of course it doesn't mean that Malda and this other guy has much more (both in quantity and quality) to say than mettw. That would be a too simple explanation, one that some egos couldn't take. So they restlessly look for answers and they finally reach the enlightenment: it means, that the free software community is dying.
I hope that all of you hard-core USENET fans enjoyed this little flame. Just one for the sake of the old good days, whaddaya say?
Jesus... I wasn't writing a PhD thesis, it's just a web bulletin board article - I would have thought that most people are capable of understanding that the title was just a bit of hyperbole. The idea behind my article is that web bulletin boards like slashdot are so poorly designed that only a certain set of opinions ever get air time. They can't even post an article from rms of all people without making a snide comment about it being too radical for the sensitive.The internet used to be a place where you could at least get people to listen to you, your opinions may have had little effect on the world but atleast you were heard. Does wanting to be heard mean that I am an egotist? Does not wanting to be cast into the role of a `geek' who is mostly interested in toy guns, leggo and star wars make me an egotist as well? God forbid!
The internet used to impliment the best ideals of democracy - no one could shut you up. But nowadays it impliments the worst ideals of business - You only present one side and then try to get everyone to repeat your mantra. The arrival of business on the internet was a great step forward for society, but we lost something that should instead have been made much better.
I joined Avagato just so I could comment on this thread. To me, this is one of the most interesting aspects of the Internet. Here we've got this group mind going, complete with constructive processes and public dsyfunction ... how do we make it run smoothly?
The combination of USENET plus the old DejaNews worked really well for me. The only real problem I have with it are the missing articles at the new Deja.
I join mailing lists periodically, but they don't seem as efficent to me as something I can dip into (like USENET) or dredge (like the old DejaNews). I would very much like to see a project that tries to combine the best of these.
Again none of your statements strike me as true (except the one of my incapability to get the hidden meaning in your article - but that's the way I am: intelectually challenged).Slashdot badly designed? According to what metric? I would make that bold statement that Slashdot is designed exactly the way Malda wanted it to be. If we judge by its popularity its design is brilliant. Slashdot is what people want to have. It's not what you want to have but this is not an evidence of bad design. It's an evidence that you have different taste and needs. I'm not even sure what is exactly the design flaw in Slashdot according to you. After all you can post comments just as everyone else so your voice can be heard. Is this the fact that Malda and co. filters the articles? This is their site and this is their right and their editorial policy. They run the show. Maybe it's Malda who's badly designed? If you want site that allows anyone to post articles why don't you start www.freedot.org? The problem (aka. bad design) with that is that not every opinion is equal in quality and not every news piece is as important as thousand other news pieces.
Internet, land of the freedom no more? In what way? Unless someone shuts your site down physically there is nothing to stop you from shelling $70 for the domain, $50 per month for hosting and voila, you have as much freedom and democracy as you want. How did that change?
And what is your point again? Who are those evil people who "present one side and try to get everyone"? Besides, what do you expect from people? Present all possible sides? You don't do it yourself. Your article is one-sided. You arbitrarily smash Slashdot in the face. From the good, the bad and the ugly you only mention the ugly (in your own judgement). I don't expect you to present my opinion on the subject because you don't know it. How realistic is your ideal of people presenting multiple sides? Will five be enough?
Here on Advogato we can carry on a long and thought out argument over several days, but on slashdot what we are doing here would be impossible. Now both Advogato and Slashdot use the same design principles and the only real difference between them is volume. This is highly suggestive that the current bulletin board design does not scale well.To me that seems to encourage cliqueish behaviour. You may disagree with that conclusion but I stand by the assertion that no development of ideas ever occurs on slashdot. The threads have too short a lifespan for that to happen.
On setting up freedot.com: That's another major difference between the web and USENET that was pointed out by raph. The web structures things more in a business model of one group owning a site and each site being in competition with each other. USENET on the other hand is more like a town meeting where there are a few rules of behaviour, but everyone gets time to express themselves and argue and there is no-one setting the adgenda.
I beleive that if the internet is to have some value in sharing opinions then we need to use some of the aspects of USENET in discussion sites. Now USENET had its own problems, but I think we can make something better by taking the basic principles behind it and wedding them to the new ideas.
So what would a better system be?
Could we borrow from USENET a "free range" of unrestricted ideas and postings?
Could multiple methods of viewing, promoting, and certifying that information coexist?
Could sites like Avagato experiment with different value-metrics on the same information base?
Could users set their own criteria for email notification?
Could I do my own queries, but ignoring value-metrics, when I wish?
Here's some of my thoughts:
- One problem with USENET is the huge volume of traffic that has to propogate through the internet. With the death of UUCP there is absolutely no reason for us to do it this way anymore. A better solution would be to use a freenet style system, but with a pre-emptive caching so that a followup ends up near to those who are reading the thread.
- Each newsgroup could have a controll file describing such things as:
- Where the FAQ can be found, or the location of a FAQ-O-Matic for the group.
- Where a search engine for the group can be found or where a web gateway for it is.
- Content-Type accepted for the group, so that a group could automatically have binary files etc deleted by servers.
- Cross-posting limits so that we can get rid of one of the major problems with USENET.
- A list of trust metric roots for the group.
- Trust metrics would allow groups to have degrees of moderation, so that an apprentice or higher might be able to automatically post, but others would have to have their posts appoved by a Master.
- ACLs on posts within the newsgroups. This would apply not only to the controll file for the group, but the newsgroup may set up a project and give a set of people WikiWikiWeb style access to the post.
just because a group is small, doesn't make it a clique. you're assuming exclusionary practise, but I think it's a simple matter of small groups of people being able to work and socialize well together, to become a bit of a family. that such small groups gently enforce (either by occasional content editing, or simply by bloc arguing) their own standards within their discussion forums is really to be expected. web boards with "owners" are not significantly more exclusionary than netnews with opinionated flamers, netiquitte and FAQs. in either case if you perceive some serious totalitarianism going down it's easy enough to set up your own little fief, surround yourself with agreeable people, and pretend you're right too.the only reason, imho, that you perceive usenet of yesteryear to be so much superior is because it was small enough for the small-group dynamic to grow in most newsgroups, instead of the few nowadays which are obscure enough to duck mainstream intrest. they're not actively shunning others, just enjoying some peace and professionalism out of the way of the endless hordes of gabbering, etiquitte-challenged newbies demanding tech support.
I think it's actually a good thing the internet has this ability to create new and slightly more civilized discussion forums on demand because it encourages old-timers to stick around much more indefinitely (by simply hiding) without running out of patience, throwing up their arms and retreating to a mossy cave. imagine there was a fixed upper limit on the number of web boards / newsgroups, like on TV, and only the most popular succeeded. bleah.
also, though I wouldn't normally pick out a passing typo, you appear to consistently think control is spelled "controll", which it's not (at least not commonly, english as a living language, yada yada).
Since no one else seems to care, I'd just like to state my appreciation for kjk's flame. It definitely did bring back memories.One of the most subtle problems with slashdot's moderation system is that flames get _negative_ points. It seems like an obvious thing to do, but when you filter out strongly differing opinions, only basically agreeable posts survive.
Similarly, Advogato's person-filtering system does basically the same thing, but with a larger granularity and (unless we get "negative certification") less effectively.
Someday I'll write an essay about this, but not until I have a solution :)
I don't think that the Free Software Community is going to disappear anytime soon, though I've never been able to figure out how to really feel part of a community that communicates soley through the the internet.I feel that article is really about the problems the community is facing simply communicating. Once you learn not to post to usenet with an email address that you want to keep clean from spam , you won't get much of a problem with spam. Many newsgroups are moderated, so they don't get clogged up with spam. Off topic posts are usually rejected by the moderators.
Weblogs have the same problem with trolls, and off topic posts. They can resort to the same fix moderation. But moderation has its own problems. The moderators have the power to censor and that power can be tempting to abuse.
There are at least two reasons for communication : 1. relating information about a subject of interest, or 2. simply socialising, getting to know other people and their opinions and ideas, or perhaps simply letting off steam.
Perhaps a list , whether its mail, web or news based, could be improved by using some means to separate the content into either informative or social. If the list is already moderated this should be relatively easy to achieve. It may also help to cut down the amount of content that needs to be searched when looking for information. The task of archiving the content is always going to be a problem.
I have been using USENET since about 1991, and have been posting an FAQ since about 1997, so I have seen some of the changes that have been aluded to in this thread.
Why back when, the experienced users always dreded the coming of September, when a new batch of clueless students wpuld flood into the university labs. That was almost the only way that one could participate in USENET at the time. When the National Capital Freenet opened at Carleton U it was only one of a handful of sites that allowed ordinary folks to use USENET, and it rapidly became one of the top ten posting sites in the world.
At that time, Commercial posts were not allowed, and so any spam was minimal, and generaly flamed. It was uncommon for a user to be able to post anonymously, (except through anon.penet.fi) and so spams tended to be shortlived.
I can still remeber being told that half of the posts from the UK came to North america Via xxx.bnr.ca because BNR (Now folded back into Nortel) had a dedicated high speed line between the Carling Ave Labs and their british site.
Then of course the commercial net arrived, and with it a major shift in perception. Before then the skating group I was following had posts from the likes of sd.hp.com and commodore.com that actually were high quality posts from actual skaters. after the .com posts were all over the map. Some folks strated adding a well known large site to their KillFile.
the curent status of usenet is that the volume is so high that news sites will spill content in 6 hours. Many sites will not pass on any articles that are cross posted to more than 4 or 5 Groups.. (Mine gets caught often because it has to go to all five relivent groups plus rec.answers and News.answers ). I was eventually forced to use the News Posting server at MIT, because posting at home, only a few issues ever got to the RTFM server.
The USENET flood method is very useful in that a copy of a posting is probaly located on a server near you. But by duplicating a post thousands of times it is likely that the ratio of copies to readers approches unity. This is in sharp contrast to the single point of failure inherent in the web forum model, where the loss of a single server can totaly stop all communication.
Mailing Lists, with the large bandwidth requirements of one copy per user, combined with the single point of failure with one server per list are perhap the worst of both worlds, although the old Mianframe listserv list servers actually did try to spread the load around various sites.
What one might want is to enermerate some desirable features of any new system.
- distributed
- scaleable
- more than one copy
- geographicaly distributed so that each article does not have to cross oceans many times
- Spam Proof (or at least resitant)
- uncensorable
- having long retention
- searchable
perhaps by taking the good points of USENET, the web and IRC one could come up with a beter discusion method?
Usenet is arguably as successful as it ever was, and it's still a great place to find answers, but the total number of thoughtful participants in the Net's work and play zones has exploded in the last half decade, and there's no way Usenet (or any other single resource) can accommodate them. Dozens or hundreds of sub-communities have formed on an ad hoc basis, and some of them (like Slashdot) have been lucky enough to prosper and become famous as Usenet once was.I don't think we can solve this problem (if it is a problem) by making new rules. But if we can work to ensure that discussion communities are ARCHIVED in a stable and spider-accessible way, then we can always rely on the ever growing search engine competition to keep those discussions (and access to the communities hosting them) available to Netizens.
I noticed one comment in this thread that "Usenet II is just the old Usenet with a few more rules". That's pretty much dead on: Usenet II is supposed to be the old Usenet, the one people are bemoaning the loss of, with just enough new rules to keep it that way.But really, the most interesting result of the Usenet II experiment is the realization that the existing Usenet newsgroups are still viable, still working.
And there's a number of small disconnected Usenets, like Usenet II, that are oriented towards specific goals. Some of them are a single server, some a few dozen. Internal company newsgroups are all over the place.
The technology is still useful.
In fact for extended discussions, I don't think that there's any question but that Usenet technology is head and shoulders over any web-based or email-based system. The existing newsreaders are excellent, and I generally read mailing lists by redirecting them to a local newsgroup.
So why try and reinvent the wheel?
Instead of trying to create new tools, use the old ones in new ways.
sab39 posted:Mailing lists seem to be very popular for free software development but I find them obstructive - the process of subscribing is time-consuming, and then for sanity's sake you need to create a folder for each list you subscribe to and set up a filter rule, and then set your mailer to sort that folder threaded,Try these procmail rules to catch most mailing list software (mail will be sorted into mbox folders in ~/inboxes/):
# Mailing list rules:0:
* ^Sender: owner-\/[^@]+
inboxes/$MATCH
:0:
* ^X-BeenThere: \/[^@]+
inboxes/$MATCH
:0:
* ^Delivered-To: mailing list \/[^@]+
inboxes/$MATCH
:0:
* X-Mailing-List: <\/[^@]+
inboxes/$MATCH
:0:
* X-Loop: \/[^@]+
inboxes/$MATCH
Then add the following lines to your .muttrc (be sure that you've made ~/inboxes/ first):
mailboxes `echo ~/inboxes/*`I use this to automatically include new mailing lists that I find myself subscribed to. It doesn't work for all of them, but it sure does for a lot of them. I also keep my mailer in threaded mode all the time, so i never need to add a hook for that.lists `cd ~/inboxes && echo *`
It's true, I never would have had to install all this nonsense to make slrn or trn do this. I also find that killfiling is desperately needed. All the same, I've been able to make mailing lists work fairly automagically.
FOAF updates: Trust rankings are now exported, making the data available to other users and websites. An external FOAF URI has been added, allowing users to link to an additional FOAF file.
Keep up with the latest Advogato features by reading the Advogato status blog.
If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!
del.icio.us
Digg
Google bookmark
reddit
Simpy
StumbleUpon
Furl
Newsvine
Technorati
Tailrank