From: Matthew J. Dovey <matthew.dovey@ou...> - 2003-12-03 11:12:35
Chuck Bearden said:
> What really struck me is how inapplicable Turner's criticisms=20
> were to the library open-source community. Of the five points (
> 1. Too many developers "scratch the same itch".
> 2. Open Source developers love a good feud.
> 3. Open Source developers often scratch the wrong itch.
> 4. In the Open Source Community, you're either "with us or=20
> against us".
> 5. The Open Source Community has a huge chip on its=20
> ), the only one I've seen any evidence for in the oss4lib=20
> world is #5, though I haven't noticed that it rises to the=20
> level of a chip on our collective shoulder. In any case, we=20
> have sound reasons for wanting open and transparent=20
> standards, which militate against vendors who embrace and extend.
To be honest, I've not seen much of no. 5 on this list either (there's
been a few "have you any experience of running your library entirely on
Linux" mails but all very factual rather than strongly anti-Microsoft
without checking the facts...)
On the other hand, I've encountered both 4 and 5 from small pockets
within my institution...
Richard Crawford said:
>It does seem, on the other hand, that there are a relatively few=20
>number of itches to scratch, and quite a few people scratching=20
>those itches. I haven't seen too many OS ILS solutions,=20
>but I've sometimes felt overwhelmed by the number of projects=20
>designed to translate MARC records into XML.
Being sort of responsible for one of these, I'll be a little defensive.
Actually the project we developed was to provide a Java/XML toolkit to
Z39.50 that you could use just knowing XML and Java (or JSP) rather than
Z39.50, ISO2709 etc (see http://www.jafer.org). Of course we needed code
to convert the binary formats coming from the Z39.50 server (not just
MARC, but also GRS.1 etc.) to XML. At the time we were working on that
bit (2000-2001) there weren't any converters so we wrote our own. Had we
been starting now, we'd probably have picked up Bas Peters Marc4J and
I suspect that a number of different groups needed MARC to XML
conversion at roughly the same time (as tends to happen), noted that
there wasn't anything (as we did), or decided not to rely on projects
just starting (it is difficult to predict with a newly started
OpenSource project how quickly it will release something stable or that
it will even get started - the number of projects on sourcforge which
have been in the design phase for years is sometimes depressing,
especially if you were hoping to use it!), rolled their own and then
This is acutally quite common in Open Source, and close commercial
source. However, close commercial source has natural pruning process (if
it isn't making money, is losing against the competitors the product
gets dropped from production). OpenSource products on the other hand
never really die...
On Wed, Dec 03, 2003 at 11:15:22AM -0000, Matthew J. Dovey wrote:
> This is acutally quite common in Open Source, and close commercial
> source. However, close commercial source has natural pruning process (if
> it isn't making money, is losing against the competitors the product
> gets dropped from production). OpenSource products on the other hand
> never really die...
You have a point, but I think it is usually possible to get an idea
of whether a particular open-source project is vital or moribund,
e.g. from frequency of file updates, or traffic on associated mailing
lists. There is a lot of open-source "detritus" lying about on SF and
Freshmeat, and you have to pick your way carefully.