You know, folks, I don't quite understand this decline in Usenet.
Okay, I know about providers cutting service, and the obvious impact
on binary traffic.
But how does that translate to a decline in these comp.* discussion
groups?
I mean, all those who are not here any more, why are they not here any more?
You'd think that comp.* people can find their way to an NNTP server.
Maybe they were here for the sake of the newbies all along? So now that hordes
of drooling masses can't find their way to your favorite comp.* group, you go
away? What? (And that after years of complaining about the damn September
that won't end!)
So what are all those people doing to get their Usenet fix? I can't believe
someone would rather, for instance, log into some obscure discussion website,
written by user interface morons.
It looks to me like the good old Usenet is back. People with a clue should be
flooding back in here. Spread the word?
Kaz Kylheku <k...@kylheku.com> writes:
> You know, folks, I don't quite understand this decline in Usenet.
> Okay, I know about providers cutting service, and the obvious impact
> on binary traffic.
> But how does that translate to a decline in these comp.* discussion
> groups?
> I mean, all those who are not here any more, why are they not here any more?
> You'd think that comp.* people can find their way to an NNTP server.
> Maybe they were here for the sake of the newbies all along? So now that hordes
> of drooling masses can't find their way to your favorite comp.* group, you go
> away? What? (And that after years of complaining about the damn September
> that won't end!)
> So what are all those people doing to get their Usenet fix? I can't believe
> someone would rather, for instance, log into some obscure discussion website,
> written by user interface morons.
> It looks to me like the good old Usenet is back. People with a clue should be
> flooding back in here. Spread the word?
I think it's the reverse. Usenet, in absolute numbers didn't decline,
but on the contrary, benefited from a big increase.
Of course, in relative numbers it became insignificant, compared to the
porn eDonkey and BitTorrent traffic, or even as you mentionned, to the
web forum ads traffic.
But you cannot expect a whole planet to be hackers and geeks.
> So what are all those people doing to get their Usenet fix? I can't believe
> someone would rather, for instance, log into some obscure discussion website,
> written by user interface morons.
I think people migrated to blogs and forums of different kinds.
Blogs make sense if you make a large post, but discussion usually sucks.
I agree that traditional forums have inferior UI (IMHO), but many find them "ok".
But there are specialized kinds of forums
-- stackexchange (stackoverflow) is great for questions-and-answers discussions
-- reddit is great for article discussions
Neither is particularly suited for lengthy discussions among people who know each other like we have on USENET. But I guess people either do not feel a need for such discussions, want wider audience, or are clueless.
To summarize, USENET is mostly a form of socialization for nerds, and demand for this product is vanishing.
> You know, folks, I don't quite understand this decline in Usenet.
I don't either. I kind of see why, for instance, blogs have become a big thing, because they're a fairly good way of people publishing article-style content (and combined with something like instapaper, they can be really compelling). But the whole web-forum thing (in any form: attached to blogs or not) is just crap, and its especially crap compared to usenet: even the rather rudimentary newsreader I have now beats crap out of any web-forum thing I've seen as an interface.
I've heard people say that the reason for the decline is that people providing news servers are very vulnerable to content appearing over which they have no control, but for which they then can get sued, and have been sued in various cases. That's clearly an abuse of the system -- there ought to be some kind of "common carrier" (is that the right term) defence -- but I think they did get sued in various cases, and lost. So any kind of entity worth suing (such as a university) probably doesn't want to run a news server any more. But there are still people who run perfectly adequate ones, who are easy to find, so that argument seems a bit hollow really. Also, the web forum places are presumably also vulnerable to getting-sued-for-content-they-did-not-originate and some of them (Facebook, google) should be very attractive targets indeed for the suing people. Perhaps they are just better at suppressing content they don't like.
I guess another reason might be that, because institutions don't run their own news servers now, they also can't have institution-local newsgroups which get used as noticeboards &c. So they need another solution for that, which probably means some kind of web forum thing (probably run by google or someone, but password controlled, so it looks secure though of course google get to mine all your information). That means that students &c don't get exposed to usenet, thus hastening the decline.
On 2011-12-31, Pascal J. Bourguignon <p...@informatimago.com> wrote:
> So my point is that we still have the same number of hackers roaming the
> newsgroups. Only now there are billions of non-geeks also using the
> web.
Do we, though? I have the sense that a good number of the ones I remember
seem to have disappeared. Maybe infrastructure/connectivity issues really are a
tipping point for some people.
On 2011-12-31, Tim Bradshaw <t...@tfeb.org> wrote:
> I guess another reason might be that, because institutions don't run > their own news servers now, they also can't have institution-local > newsgroups which get used as noticeboards &c. So they need another
I was thinking of this also. And the obvious problem there is that if you lose
your local news server (likely because /that/ news server has lost its upstream
provider), then you may really have nowhere to go other than Google Groups,
which is crap. The reason would be: no outgoing port 119 access through your
draconian corporate firewall to get to an alternative server. Oops!
People may be in need of a proxying solutions that go over port 80 and use the
HTTP protocol (to defeat firewalls that do allow port 80, but also parse for
HTTP).
In general, maybe Usenet needs to start supporting HTTP transport: it could be
time for NNTP over HTTP.
> People may be in need of a proxying solutions that go over port 80 and use the
> HTTP protocol (to defeat firewalls that do allow port 80, but also parse for
> HTTP).
There's at least one (hem, the one I'm using) whch does this already, it turns out (I'm not using the port-80 server, but I will try next time I am trapped somewhere with restrictions (most of the time I now use my own little 3g/wifi gadget for access from my laptop at work, which doesn't suffer from this).
Alex Mizrahi <alex.mizr...@gmail.com> writes:
> But there are specialized kinds of forums
> -- stackexchange (stackoverflow) is great for questions-and-answers
It's good, but it's not a suitable replacement for USENET. Despite the
numerical reputation system, it seems to celebrate the clueless and the
careless, as opposed to good USENET group like comp.lang.lisp, which
makes a newbie's entry a more trepidatious affair, but in time teaches
him how to behave, who to trust with which topics, and how to contribute
in a valuable way.
By contrast, StackOverflow -- the StackExchange site I visit most
frequently -- is for many a speed game: A lazy or indifferent poster
submits an ill-formed question, and those hungry for reputation race to
post the first responses, rewarding the question's poster with undue
effort and attention in trade for recognition. Note, though, that once a
few answers arrive, and especially once an answer is accepted, the
attention given to the question falls off rapidly. A question answered
poorly will rarely be corrected, and, even if a hero attempts to set the
record straight, he's effectively arguing with /the person who asked the
question/, who likely isn't fit to resolve the argument.
Voting on answers is meant to help here, but I wonder if people who
search questions and read the answers pay more attention to the accepted
answer or to the answer that eventually receives the most
votes. Unfortunately, there's no way (that I know of) for the crowd to
usurp the question author's decision as to which answer is "best."
Perhaps the "community Wiki" conversion handles this situation.
I've listened to every one of Joel Spolsky and Jeff Atwood's podcasts on
the development of the site and platform, and I know that their stated
goals for the platform included creating "living documents" that sift
the best information to the top and keep it accurate over time. Lately I
see more and more noise in the contributions, and it doesn't feel nearly
as worthy of archiving as CLL has in the eleven years or so that I've
been reading it.
> Neither is particularly suited for lengthy discussions among people
> who know each other like we have on USENET. But I guess people either
> do not feel a need for such discussions, want wider audience, or are
> clueless.
One of the beautiful things about USENET (and NNTP) is its granting of
experiential control to each user. By that I mean that the client
software is decoupled from the hosted discussion, and that each user is
free to choose his own tools for reading and participating in the
discussion. The information created lives apart from those
client-related decisions, as evidenced by the DejaNews/Google Groups
archive surviving behind a Web front-end, even if few of those
constituent contributions were made using any Web-related technology.
Wanting that power of customization and being skilled enough to use it
is no longer common among computer users; that skill was once a
prerequisite, but it's arguably the success of software evolution that
most people can /use/ computers today without also needing to understand
how they work or how to program them. The open and empowering
opportunity offered by the USENET /access model/ is now seen as a
burden. It's competing against ever-slicker Web-fronted systems that
/each collect a maintain a disparate set of information/. (Twitter is a
not-quite-as-open counterexample, with a variety of client software
fronting a common information pool.) It's as if dropping one newsreader
for a newer alternative not only provided you with a nicer interface,
but also forced you to now participate in a wholly separate USENET.
One more point: Much lisp Lisp, USENET is the dog that everyone likes to
kick, and the would-be defenders are nowhere to be found. Apparently
there was a phase when a lot of people must have /wanted/ to get
involved in USENET (my guess is around 1998 through 2002), but had
trouble wrestling with that freedom I described earlier, and also
perhaps had trouble socially with navigating long discussion threads. I
expect that the latter is again partly due to poor software choices, and
with inadequate skill in using the available software, but people
suffered regardless. Those experienced are now codified as exaggerated
"wisdom" that USENET was horrible and needs to die, as if the whole
system was just a barely-acceptable resting stop on the way to the Web
really solving the problem.
The aforementioned Joel Spolsky and Jeff Atwood promulgate this
opinion. As much as I adore the StackExchange system, if they think that
the StackOverflow site is a fitting replacement for USENET as I know it,
they are sadly mistaken.
Kaz Kylheku <k...@kylheku.com> writes:
> You know, folks, I don't quite understand this decline in Usenet.
> Okay, I know about providers cutting service, and the obvious impact
> on binary traffic.
> But how does that translate to a decline in these comp.* discussion
> groups?
> I mean, all those who are not here any more, why are they not here any
> more?
I think that quite a lot (my estimate would be: about 20-30%) of those
people who once posted frequently at c.l.l. are still lurking here (or
at least take a glance from time to time).
However, those people are usually silent, first because they are no
newbies anymore who are in urgent need of answers, second, because
Usenet is no more the medium of choice for newbies who need to be
answered, and third, because thought- and discussion/flamewar-provoking
opinions are posted much more rarely than in earlier times (especially
those times, when Erik was still around).
> You'd think that comp.* people can find their way to an NNTP server.
>> You know, folks, I don't quite understand this decline in Usenet.
>> Okay, I know about providers cutting service, and the obvious impact
>> on binary traffic.
>> But how does that translate to a decline in these comp.* discussion
>> groups?
>> I mean, all those who are not here any more, why are they not here any
>> more?
> I think that quite a lot (my estimate would be: about 20-30%) of those
> people who once posted frequently at c.l.l. are still lurking here (or
> at least take a glance from time to time).
As a Usenet user since the early 80's, comp.lang.lisp seems as active to me now as in its heyday, and much stronger than a fallow period a number of years back.
What I see is a big drop of available groups. Many old groups have faded, which is to be expected, but they've not been replaced by new ones. For example, I see no group with "agile" in the title. All that discussion, and there's a lot, is happening in Yahoo Groups and elsewhere.
Nicolas Neuss <lastn...@scipolis.de> writes:
> I think that quite a lot (my estimate would be: about 20-30%) of those
> people who once posted frequently at c.l.l. are still lurking here (or
> at least take a glance from time to time).
> However, those people are usually silent, first because they are no
> newbies anymore who are in urgent need of answers, second, because
> Usenet is no more the medium of choice for newbies who need to be
> answered, and third, because thought- and
> discussion/flamewar-provoking opinions are posted much more rarely
> than in earlier times (especially those times, when Erik was still
> around).
I have been reading comp.lang.lisp for years on end without ever posting
here. I can imagine other people do to. So tonight I thought I might as
well make a first post to let you know I've been priming my lisp
knowledge from you for a long time now :-)
Peter Stiernström writes:
> Nicolas Neuss <lastn...@scipolis.de> writes:
>> I think that quite a lot (my estimate would be: about 20-30%) of those
>> people who once posted frequently at c.l.l. are still lurking here (or
>> at least take a glance from time to time).
>> However, those people are usually silent, first because they are no
>> newbies anymore who are in urgent need of answers, second, because
>> Usenet is no more the medium of choice for newbies who need to be
>> answered, and third, because thought- and
>> discussion/flamewar-provoking opinions are posted much more rarely
>> than in earlier times (especially those times, when Erik was still
>> around).
> I have been reading comp.lang.lisp for years on end without ever posting
> here. I can imagine other people do to. So tonight I thought I might as
> well make a first post to let you know I've been priming my lisp
> knowledge from you for a long time now :-)
I second this, I have been reading comp.lang.lisp for years too but I
post not so often. BTW comp.lang.lisp is a great source of information.
> Nicolas Neuss <lastn...@scipolis.de> writes:
> > I think that quite a lot (my estimate would be: about 20-30%) of those
> > people who once posted frequently at c.l.l. are still lurking here (or
> > at least take a glance from time to time).
> > However, those people are usually silent, first because they are no
> > newbies anymore who are in urgent need of answers, second, because
> > Usenet is no more the medium of choice for newbies who need to be
> > answered, and third, because thought- and
> > discussion/flamewar-provoking opinions are posted much more rarely
> > than in earlier times (especially those times, when Erik was still
> > around).
> I have been reading comp.lang.lisp for years on end without ever posting
> here. I can imagine other people do to. So tonight I thought I might as
> well make a first post to let you know I've been priming my lisp
> knowledge from you for a long time now :-)
I am also in the same category as Peter but probably have been here
for a shorter time. I have been going through here sometime in
sequence and sometimes by topic and it has been greatly helpful. I
have yet to ask anything here as i am still a newbie and my question
have mostly answered by old posts and people on #lisp. but as Steve
said earlier, i have been on Stackoverflow and it's in no way a
replacement for USENET.
> What I see is a big drop of available groups. Many old groups have
> faded, which is to be expected, but they've not been replaced by new
> ones. For example, I see no group with "agile" in the title.
Try gmane.
gmane.comp.programming.modeling.agile seems to be a pretty active one.
> All that
> discussion, and there's a lot, is happening in Yahoo Groups and elsewhere.
A lot happens in mailing lists which are not very different from USENET, especially if you use gmane. The only thing which bothers me is a need to subscribe to be able to post. If they would automatically authorize users which come from gmane that would be perfect.
Annual stats for a couple of groups I read are below (from Google, so perhaps not really reliable, both tables remove partial years). Both look as if they reached a peak in 2005-2006 and are now in fairly steep decline (in gnuplot "plot ... with steps" is pretty good for these).
> On 2011-12-31, Tim Bradshaw <t...@tfeb.org> wrote:
> People may be in need of a proxying solutions that go over port 80 and use the
> HTTP protocol (to defeat firewalls that do allow port 80, but also parse for
> HTTP).
I have just started trying out eternal-septmber.org as a free provider of
text-only usenet groups. All they require is to register with an email
address. They accept connection to their nntp server over SSL on port 443,
which I think would get through most firewalls that permit general https
access to web servers. This is my first post using their service, so I'll see
how that goes.
sidney <sid...@sidney.com> writes:
> Kaz Kylheku wrote, On 1/01/12 6:13 AM:
>> On 2011-12-31, Tim Bradshaw <t...@tfeb.org> wrote:
>> People may be in need of a proxying solutions that go over port 80 and use the
>> HTTP protocol (to defeat firewalls that do allow port 80, but also parse for
>> HTTP).
> I have just started trying out eternal-septmber.org as a free provider of
> text-only usenet groups. All they require is to register with an email
> address. They accept connection to their nntp server over SSL on port 443,
> which I think would get through most firewalls that permit general https
> access to web servers. This is my first post using their service, so I'll see
> how that goes.
Perhaps writing about your experience, how to setup your newsreader and
subscribe, etc, on the various blogs and forums, would be good to
attract newbies to usenet?
> > Kaz Kylheku wrote, On 1/01/12 6:13 AM:
> >> On 2011-12-31, Tim Bradshaw <t...@tfeb.org> wrote:
> >> People may be in need of a proxying solutions that go over port 80 and use > >> the
> >> HTTP protocol (to defeat firewalls that do allow port 80, but also parse > >> for
> >> HTTP).
> > I have just started trying out eternal-septmber.org as a free provider of
> > text-only usenet groups. All they require is to register with an email
> > address. They accept connection to their nntp server over SSL on port 443,
> > which I think would get through most firewalls that permit general https
> > access to web servers. This is my first post using their service, so I'll > > see
> > how that goes.
> Perhaps writing about your experience, how to setup your newsreader and
> subscribe, etc, on the various blogs and forums, would be good to
> attract newbies to usenet?
I think the problem is much more straightforward:
> Maybe they were here for the sake of the newbies all along? So now that > hordes of drooling masses can't find their way to your favorite comp.*
^^^^^^^^^^^^^^^^^^^^^^^^^
> group, you go away?
Consider the possibility that the "hordes of drooling masses" as Kaz puts it can find their way just fine, but don't like what they see when they get here and choose to take their business elsewhere. Maybe the CL soup just isn't that good any more [1].
RG <rNOSPA...@flownet.com> wrote:
> Consider the possibility that the "hordes of drooling masses" as Kaz > puts it can find their way just fine, but don't like what they see when > they get here and choose to take their business elsewhere. Maybe the CL > soup just isn't that good any more [1].
If you look at traffic stats for unrelated newsgroups (I gave a couple of
samples the other day, but I've since made some better pictures for a bunch
of groups which make the situation much clearer) you can immediately see
that this is not CLL specific, but much more general.
On Sat, 31 Dec 2011 07:14:35 +0000 (UTC), Kaz Kylheku <k...@kylheku.com>
wrote:
> You know, folks, I don't quite understand this decline in Usenet.
I think a part of the decline is through google groups and yahoo groups.
Basically the google interface replaced nntp for a lot of readers, and they
can't differentiate between google groups and usenet anymore. A lot new
groups are created on google groups only (e.g. about varius web and js
stuff). They don't go to usenet anymore (and google controls the view and
gets the advertising revenue). Next to that what others told about
stackoverflow, blogs etc.
sidney <sid...@sidney.com> writes:
> I have just started trying out eternal-septmber.org as a free provider
> of text-only usenet groups. All they require is to register with an
> email address. They accept connection to their nntp server over SSL on
> port 443, which I think would get through most firewalls that permit
> general https access to web servers. This is my first post using their
> service, so I'll see how that goes.
Hi,
I use eternal-september.org since April 2011 via a local leafnode.
Reason was that the German Telekom shut down their news servers.
Sometimes eternal-september.org is down for a couple of hours but
nothing serious for private use. For a free service reliability
is quite good.
On Sat, 31 Dec 2011 17:13:22 +0000 (UTC), Kaz Kylheku <k...@kylheku.com> wrote:
> On 2011-12-31, Tim Bradshaw <t...@tfeb.org> wrote:
>> I guess another reason might be that, because institutions don't run
>> their own news servers now, they also can't have institution-local
>> newsgroups which get used as noticeboards &c. So they need another
> I was thinking of this also. And the obvious problem there is that if
> you lose your local news server (likely because /that/ news server has
> lost its upstream provider), then you may really have nowhere to go
> other than Google Groups, which is crap. The reason would be: no
> outgoing port 119 access through your draconian corporate firewall to
> get to an alternative server. Oops!
> People may be in need of a proxying solutions that go over port 80 and
> use the HTTP protocol (to defeat firewalls that do allow port 80, but
> also parse for HTTP).
> In general, maybe Usenet needs to start supporting HTTP transport: it
> could be time for NNTP over HTTP.
Maybe people move to places behind overly restrictive firewalls. Even
so, it's often easy to find the local administrator and ask for an
exception. I've done it a few times already.
> Usenet needs to be evolved, e.g. to clound. It's not economical to download and to save copies locally, and so not environmentally friend :-)
When you visit a web page you download user interface graphics (CSS, images), layout (HTML) and logic (JS). While with NNTP you download only text content, with UI being a local program.
So if something is not environment friendly, it is the web, not NNTP.
>> Usenet needs to be evolved, e.g. to clound. It's not economical to
>> download and to save copies locally, and so not environmentally friend
>> :-)
> When you visit a web page you download user interface graphics (CSS,
> images), layout (HTML) and logic (JS). While with NNTP you download only
> text content, with UI being a local program.
> So if something is not environment friendly, it is the web, not NNTP.
And the size of the entire usenet text feed is probable less than your average youtube video...
