On Fri, 31 Mar 2000, Webmaster Jim wrote:
> On Thu, Mar 30, 2000 at 04:38:15PM -0600, Klaus Weide wrote:
> > Webmaster Jim wrote:
> > > We have provided a URL for the translation teams to pick up the
> > > developing version with a GNU-based subdirectory scheme, but I'm not
> > > aware this has been adopted.
> >
> > Cuuld you explain this sentence? Especially, what is a "GNU-based
> > subdirectory scheme"?
>
> Since I've worked on Lynx, each "version" unpacks into a subdirectory
> called, for example, lynx2-8-3. Thus, it is hard for some automated
> program to notice that a new version has been created. Typical GNU
> software is numbered with the version, as in lynx-2.8.3.7. Each change
> in the rightmost number indicates a new relaese or version. Thus, the
> translation project chose to publish lynx "2.8.3" templates, even
> though the Lynx developers understand that 2.8.3 doesn't exist yet.
Thanks for the explanation.
I still don't really understand why the _directory_ name matters -
the version embedded in the .tar.gz filename should be what matters.
> What I do is unpack the Lynx distribution, rename the subdirectory to
> include the "dev" number, then repack it as a tar.gx file. I also
> include the latest message catalogs I'm aware of.
>
> I'm not concerned that this is being generally ignored.
But it would certainly be better if there were only _one_ scheme,
used consistently. (I mean for the code's version - again, I don't
think directory names should matter.)
I don't find any hint of your re-packaged code on
Linkname: The lynx textual domain
URL: http://www.iro.umontreal.ca/contrib/po/HTML/domain-lynx.html
so it seems that, indeed, it isn't being 'picked up'. You may be going
to all this trouble for nothing.
Klaus
