Re: [debian-devel] Proposal for removal of mICQ package
>--[Craig Dickson]--<crdic@pacbell.net>
> Martin Loschwitz wrote:
> > bug, ok, but not setting such an un-important variable (it _is_
> No, I don't think I agree that EXTRAVERSION is unimportant. From what
> Rudiger wrote in the previous thread about mICQ, this value is
*cough* Rüdiger, that's Ruediger in 7bit-crippled. Thank you for your
consideration.
> transmitted to ICQ servers, and is meant to function equivalently to a
> web browser's User-Agent string. This seems quite sensible to me. If a
> particular client-side bug is only reproducible with a particular
> distro's package, then that would be a very good indication that it's
> caused by something the package maintainer is doing wrong.
Only partially right, you need to request it (and know how), and it's
actually an extension from mICQ to the ICQ protocol. However, it is helpfull
finding bugs, if it's a precompiled binary, then I don't have to ask about
which compiler and settings were used and can even use the very same binary,
if it's not linked against glibc or libiconv, I can tell pretty fast why it
won't convert between different Japanese encodings, etc (the string is
actually larger than the text in the define).
What's broadcasted via the server is only the numerical version, via some
abused timestamps and specific capabilities. Not all clients interpret these
values (the official ICQ clients of course not at all), but mICQ is rather
comprehensive in this regard.
Just in case someone actually wants to know.
Yours, Rüdiger.
--
100 DM = 51 € 13 ¢.
100 € = 195 DM 58 pf.
mailto:ruediger@ruediger-kuhlmann.de
http://www.ruediger-kuhlmann.de/
Reply to: