From: Kent Landfield (kent@landfield.com)
Date: Fri Jun 01 2001 - 16:11:25 CDT
# The article has to be archived to be spread, that is how Usenet works.
# You archive everything on a server for a single group of users. These
# users then can read it (or not) in their own time. The only reason for
# deleting older articles is when the server gets full.
NO!!!!! An archive is not a news spool directory. Say it again... The article
is not stored in an "archive" to be read by users. It is stored in a news
spool directory. Articles are flooded to the other sites down/upstream via
the news spool directory. Articles are expired based on local policy, from the
news spool directory. "news spool directory" != "archive". Archives are
facilities that store various types of files. The files could be software
totally unrelated to news articles. In most cases, the archives are a mixture
of software files, documentation files, news articles. Notice that news articles
are just one part of the archive ? I write software that I donate to the
community at-large. When I make it available I put in in my web/ftp archive.
When I archive Usenet articles for long term availability I place them in
my web/ftp archive to make them publicly available. Like me, most news admins
don't let the public gain access to our news spool directories.
"news spool directory" != "archive"
# You can't say no to archiving when you post to usenet. This whole
# Archive thing is a farce. Users should be able to ask for a maximum
# duration of archiving by stating a date, to make things easy we can
# define a standard date. Simply saying that it is up to common policy
# when an article is still considered current is so vague I'll call it
# naively stupid.
You seem to be confusing expiring an article from the news spool directory
with a long-term or permanent archive.
There is no reason why people couldn't say no to archiving. It is an easy
thing to deal with from an archivist's perspective. It is not rocket science.
(My apologies to the rocket scientists out there.)
# Example:
# - on avarice a group deletes articles within 10 to 60 days,
A group does not delete articles. The management of the news spool directory
is a local policy issue that is dependent on many items. Diskspace and interest
in the topics come to mind... ;)
# - so you post a article with "Archive: no" and presume it will be
# unavailable to most new readers shortly after those 60 days,
No... It will not be archived and will not be publicly available outside
the standard Usenet newsreading environments. You should not assume expire
periods at any point.
# - but after 30 days disk prices drop dramatically and many servers
# are expanded.
# - the average of article deletion goes up to 60 to 1000 days.
So ?
# Result:
# - your article is now practically archived right here on usenet.
# - and it is common policy because many servers have upgraded.
# What are you gonna do about it?
# - call all newsmaster and ask them to lower retention on that group?
# - sue them all
# - do nothing
#
# See, the current Archive header does absolutely nothing.
Then address that somewhere else in the draft. The Archive header is targeted
towards archives.
# If there was a date involved than it would have been deleted from all
# western servers at your requested date, or sooner.
I have no problem with a date parameter. I have wanted one from the begining
as have others but for the Archive header use...
-- Kent Landfield Phone: 1-817-545-2502 Email: kent@landfield.com http://www.landfield.com/ Search the Usenet FAQ Archive at http://www.faqs.org/faqs/ Search the RFC/FYI/STD/BCP Archive at http://www.faqs.org/rfcs/