Boost logo

Boost :

From: Dietmar Kuehl (dietmar_kuehl_at_[hidden])
Date: 2002-03-31 10:39:43


Hi
Carl Daniel wrote:

> From: "Beman Dawes" <bdawes_at_[hidden]>
>>The rationale is that (1) we want tight control on who and what gets
>>posted, not just to avoid spam, but also to avoid a lot of other unhelpful
>>posts,

This is exactly what moderated newsgroups are for (I hear you scream
about the delay introduced but moderation but bear with me for a moment
as this is not required).

 (2) we would be better off to build experience with newsgroups which
>>we control (and can change quickly at will) before going public (if indeed
>>we ever want to go public),

I don't think forums work this way: They get established and basically
are either successful and stay the way they are or they disappear
siltently. Also, once a newsgroup is created the handling of the
articles and what exactly is considered on topic or off-topic is subject
to variation. That is, I think it is OK to go "public" right away: After
all, the forum is public in a sense anyway.

>> (3) we want to hold administrative work to a minimum,

Once a newsgroup is created, there is no administrative work. Well, if
it is a moderated forum, administration of the moderation stuff is
needed. But even this is not that much (well, John Potter does the work
for comp.lang.c++.moderated such that I can't really tell how much it is
exactly...).

>> (4) a hope that spammers are less likely to harvest email
>>addresses from a private, controlled membership, newgroup than a public
>>one.

For moderated newsgroups the messages are channeled through a script.
This script can easily be used to mangle the know e-mail addresses
exposed (eg. the "From:", "Reply-To:", ... etc. headers).

>>I don't have any personal experience running a newsgroup, so the above is
>>based a bit on speculation rather than experience.

I am a moderator of comp.lang.c++.moderated and I implemented at least
parts of the initial moderation scripts. I don't think that any of the
raised objections really holds...

> I agree 110%. I think the only alternative would be a public moderated newsgroup,

> which obviously entails establishing a moderation group, delays postings, and does

> nothing about email address harvesting.

False in all accounts! Nothing requires the "moderator" of a newsgroup
to be a human person. Almost certainly, there will be an administrator
who is responsible for having things work smoothly but this is already
the case anyway (however, with the creation of a newsgroup I can imagine
that a new administrator would be added to deal with the newsgroup
related stuff).

Concering posting delays, this is not required at all: Submissions to
newsgroups can be handled by a script which eg. checks that the "From"
header indicates a registered person and if this is the case just sends
the article forward. In de.comp.lang.iso-c++ we use a variation on this
scheme adding a little bit more control: Unknown poster's articles are
sent to a moderator for manual processing (in this case indeed adding a
moderation delay) as are article containing specific keywords (like eg.
"ActiveX" which is, by itself, off-topic in this newsgroup). Articles of
known and "well-behaved" posters are approved directly without human
intervention. That is, at this point there is no moderation delay beyond
sending an e-mail to some remote host.

Since people also mentioned a propagation delay (which is, however, on
all servers I ever used in the area of minutes), just note that it is
trivial to sent approved article not only to the newssystem but also to
corresponding mailing lists: In this case the propagation delay is the
same as the current delay. Of course, the system can be made such that
it does not matter whether an article is posted to the newsgroup or to
the mailing list...

Concerning e-mail harvesting, I don't think that this is a big problem:

I'm posting regularily since several years and the majority of spam I
get can be tracked to other activities rather than my news activity.
However, the scripts handling the moderation process can easily mangle
all known headers and/or all occurances of things looking like an e-mail
address. No big deal.

I think that a public newsgroup is the way to go. On the other hand,
nothing prevents us from having multiple forums anyway and public
newsgroup, eg. comp.lang.c++.boost, is probably a good idea even if
we stick with mailing lists, too.

Regards,
   Dietmar

-- 
<mailto:dietmar_kuehl_at_[hidden]> <http://www.dietmar-kuehl.de/>
Phaidros eaSE - Easy Software Engineering: <http://www.phaidros.com/>

Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk