Re: What I would say if someone asked me about no win32 - Mailing list pgsql-advocacy

-----Original Message-----
From: Jan Wieck [mailto:JanWieck@Yahoo.com]
Sent: Thursday, August 14, 2003 1:54 PM
To: Bruce Momjian
Cc: Josh Berkus; Merlin Moncure; Robert Treat;
pgsql-advocacy@postgresql.org
Subject: Re: [pgsql-advocacy] What I would say if someone asked me about
no win32

> It is also partly my fault, as I failed on the first attempt to redo
the
> fork+exec part for Bruce, then got a bit frustrated over things not to

> be discussed publicly and didn't give it much attention, finally I had
a
> ton of other things with higher priority after that. This is not an
> excuse, it's life, sorry.

I have a few observations about this situation.  As an outsider (i.e.
never contributed anything to the project, but has lots of opinions) who
has been trying to break in, I can't really comment on development or
design considerations.  However, I think there are significant barriers
to capable developers who would like to contribute to and/or maintain
the win32 port of postgres, the biggest of which is simply locating the
current port (!).

Here are some of the problems I see:
1. The windows faq has not been updated since November, 2002 and has no
mention of the Peerdirect port (PD) or the status of the official port.
2. There was some confusion regarding the official status of Justin's
cygwin packaged 'alpha' build, which came out with 7.3.1.
3. The move from VS 6.0 to mingw for the build environment will not at
all be popular with windows developers (although I personally understand
the decision).
4. Many windows users are not even aware of the PeerDirect port's
existence.  To find it on the web site you need to follow a strange path
of links to the guides section to get there.  Because 7.2.1 is showing
its age, most people would not find it in the official ftp server.
5. Both PD 7.2.1 and Cygwin 7.3.2 (last version I've tested on win32)
have some serious problems with WAL and other issues which will not give
users the same experience they would have on unix system.
6. There is a certain degree of ambivalence towards the port that is
evident on the hackers list.  This is likely fairly discouraging to
would be donators.  While this is ex parte discussion or opinions of
people who really may not belong on hackers, this is the only place to
get useful information.

Here are some solutions:
1. Make some kind of official statement about the status and direction
of the port on the web page and put it on the front page.  Npgsql, for
example, gets it its own link.  An explanation of why the 7.4 deadline
was missed and the reasoning behind the move from vs 6.0 would be nice.
2. Provide a download link for the current 'official' port's source
along with build instructions *or* provide cvs instructions about how to
pull down the win32 branch (does one exist? Or are they part of the
official source?).

In short, I think if would be tinkerers and testers would have an easier
time figuring out what is really going on, it could help tremendously
and help ease frustration.  I really have a great deal of sympathy and
respect for Bruce and Jan's heroic attempts to complete the work, and
hope their efforts will not be in vain!

Regards,
Merlin