[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

SAMIZDAT on KNOPPIX with database ported from mir: hints

From: boud
Subject: SAMIZDAT on KNOPPIX with database ported from mir: hints
Date: Tue, 16 Jan 2007 00:28:11 +0100 (CET)

hi samizdat-devel,

SAMIZDAT on KNOPPIX with database ported from mir

i created a modified version of knoppix (a CD from which you can boot a computer - the operating system is loaded into RAM, so it
can be used on a non-GNU/linux system without any problems)
containing samizdat-0.6.0 plus some of the hacks i've posted and plus a small part of a mir database ported into the samizdat
sql structure.

The idea is to go as far as possible in giving users power over
the system. The idea is to encourage "ordinary" users to experiment
with the functionality of samizdat in a way that is not possible with
a live site (most people prefer to publish serious things on a live
site, rather than experimenting just as an exercise).

Remastering knoppix in general:

samizdat on knoppix-5.0.1 (slightly old) - a couple of hints:

* i couldn't get postgresql-7.4 (default) to work - the postgresql config files were in strange places and i hacked a bit but failed. i solved the problem by uninstalling (trying to uninstall) 7.4 and i installed postgresql-8.2. In /etc/init.d/postgresql-7.4 i put in exit 0 near the beginning. This was because postgresql (knoppix
uses mostly unstable debian, with some mix of stable and testing)
uses some intelligent way of checking which postgresql versions are
running and i had to persuade it to use 8.2.

* i have a set of sql + shell scripts for the mir -> samizdat conversion.
- they are not user friendly
- they are hardwired in some way
- one script requires clicking/pasting since i was lazy

- they do not preserve all the information from the mir database, they
include things like article title/author/date/text, same for comments
and which article they're attached to, jpg images

IMHO they got the job done sufficiently for the purpose. :)

If someone still thinks s/he might learn something from them after all these warnings, then i could clean them up a bit and distribute them.

On the other hand, it might be better for someone to write a real ruby script
which does all of this much more systematically.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]