cygipc (and PostgreSQL) XP problem resolved!

[Charles Wilson]

Elfyn McBratney wrote:

> Kinda, my other side was working for a production company and I was an 
> editor. Uhm, I did  very little work on this and missed a few features 
> here and there in the code. I'm interested in working on it, though.
> 
> Is there still interest in having a working (the other 20%) cygwin 
> daemon?

Absolutely.  And that will be true regardless of whether cygipc is 
accepted as a package for distribution via the mirror system.

cygipc sucks.  (Even the original author, Ludovic Lange, agrees with 
that).  My fondest wish is for it to die.  But it's the best (only) 
thing we have -- because it is fully functional, and cygserver is not.

The fear is/was, as Robert said waaayy back and quoted again today, that 
by making it easy for cygipc to be used by 'the casual cygwin user,' we 
remove any impetus for something better.  E.g. nobody would be 
interested in working on cygserver if cygipc was 'already there'.

So, we decided to keep cygipc available -- but awkward to get to.  It 
was a bit of social engineering -- that failed, as cgf pointed out last 
Tuesday in this thread.

Egor wrote the original cygserver.  Robert worked on it, and added most 
of the existing IPC functionality.  Conrad took over maintainership last 
July and worked on IPC stuff (plus general cleanup, added some debugging 
features, etc).  But he disappeared last September, and nobody has done 
anything since.

cgf asked (plaintively) about the cygserver status in November.  Nobody 
responded.  He asked again in January, to little immediate response -- 
although you (Elfyn) responded in February -- but again, little happened 
(I'm not being critical; life happens)..

So finally, cgf said it looks like our social engineering experiment 
didn't work; perhaps it's time to relent and put cygipc into the 
distribution.  But that DOESN'T mean that cygserver should be abandoned. 
  Remember the mantra: cygipc sucks.  If you want to, and have the time, 
PLEASE help finish cygserver -- so we can send cygipc to its 
long-delayed doom.

Even if we do (temporarily) add cygipc to the cygwin distribution.

--Chuck




--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/