[tei-council] TEI P5 1.9.0
James Cummings
James.Cummings at oucs.ox.ac.uk
Wed Jan 26 13:36:32 EST 2011
On 26/01/11 18:05, Sebastian Rahtz wrote:
> This is why I am raising the subject - what do we expect our constituents to do
> when we say "lookee, folks, a new release!!!"?
In my mind they (should) expect the following things:
1) The website text to have changed to new text (with the old one
preserved in Vault/P5/) i.e. releases/doc/
2) The website releases/xml/ to have been updated
3) The downloadable packages on sourceforge to have changed
4) Roma to use the new version of the files by default. (i.e. if
you don't specify a source)
5) The release to be submitted to oXygen so that whenever that is
next packaged up they will get the most recent version.
Obviously, those of us who continually use the version in
SourceForge Subversion don't really notice as much of a
difference at this point.
I don't view the the TEI schema debian packages as that necessary
(I can have a cron job do 'make install' if I need to). I suppose
the same is true of xslt stylesheets.
I do like the tei-oxygen package because (once sebastian
repackages it) I don't have to think to install it, it just comes
as a regular upgrade. Whereas if I've downloaded a standalone
package from oxygen, I've got to manually go upgrade it when it
nags me to do so.
> my personal use of an installed P5 is to validate files, I know where
> I can find tei_all.rnc or teilite.rnc or whatever I want. How this relates
> to Windows folk, lord knows :-}
Presumably they want to do the same thing, if they don't use oxygen.
> I suspect many of our users rely on oXygen to have stuff in the right place;
> and indeed our distribution of zip files was set up talking to the oXy folk,
> to make sure it worked.
I would be tempted to suggest the following:
If packaging up the tei-p5-xsl2 and tei-p5-schema are able to be
automated, then fine lets keep doing so. (I prefer the idea of
deprecating P4, by only providing support for it via the website
and SF.) Other tei packages I am less sure we should be providing
other than in SF subversion.
But, in wanting to encourage more community involvement. I think
if we decide this we should then go to the community and say "are
there any volunteers to do this?" and if the community wants to
support it, then they can. If they don't want to, then it might
not be necessary. (e.g. if Martin wants to provide an easy
windows installation of TEI xsl and schema, then great, but it
should be him as an individual (or an institutional activity)
rather than the TEI-C providing this.)
But this stems from a desire to simplify what we do provide, but
make sure we do that well.
My two pence,
-James
--
Dr James Cummings, Research Technologies Service
OUCS, University of Oxford
More information about the tei-council
mailing list