Warning: This post is containing a rant. If you have weak nerves,
please, do not read it.
Hello,
The porting project needs a porter to port OpenOffice.org on Mac OSX!
You may say, that there is a porting team. I will firmly disagree. There
is a group of Mac OSX "builders", but no porter. The aim of a port is
IMHO not so much in providing binary packages for a given platform as in
making it possible to build these binary packages _out-of-the-box_. This
is not possible for the while. There is a serie of patches that allow to
build scatered all around the place, but OOo is not buildable on Mac OSX
using only build instructions and a fresh cvs checkout.
NAKATA Maho wrote:
My name is NAKATA, Maho, enthusiastic OOo porter.
I would like to ask you for donation of a pentium 4 or Athlon64 (I mean
fast/reliable and small) machine for Windows mingw porting.
Sorry, but I will be a bit personal in this one. Not so long ago, you
were thanking someone for donating you a G5 and Apple developer
subscription. So, please use it for the good of all. Trying to be
porting on hundred platforms at the same time is just leading, as it is
now, to a substandard result. Focus on one and do it in excellence. You
may argue that you are doing it as voluntary work and in your free time.
I will tell you that probably nobody was pointing a loaded gun at you
forcing you to start doing it. You chose to do it and to do it the best
possible is a moral necessity for you. Please, contemplate this deep
blog entry: http://www.advogato.org/person/msevior/diary.html?start=36
Maho's profile:
o FreeBSD ports committer. maintains OOo packages for 2 years.
I'm maintaining 1.0, 1.1 and 2.0 ports and many other ports.
I raised many issues regarding FreeBSD porting, and reduced # of
patches from ~120 to 2 (currently). Has an official porting status
page at http://porting.openoffice.org/freebsd/ and of course, I'm
maintaining directly this page.
o founder of ja.openoffice.org. I was the founder of ja.openoffice.org,
and have been a lead for 1.5 years.
o I made 3 Child work spaces for FreeBSD porting, and one for Mac OS X.
these CWS had already been integrated into master.
o Mac OS X porter. There are several MacOSX active porters here,
and I'm also a Mac OS X porter. I have been publishing MacOSX (Tiger)
packages with patches at ftp://ooopackages.good-day.net/pub/OpenOffice.org/
and also providing CVS repo. for MacOSX.
I'm also do some jobs for fink. Asari-san and I made a
finkinfo for OpenOffice.org and we have been collaborating and trying
to make better packages for MacOSX with fink as well.
It is nice to have a lot of things on the CV, but I will keep contending
that there are no MacOSX porters. It is nice that you are providing
patches in CVS, but tell me honestly, what is preventing you from
cleaning these patches out and pushing them upstream? It may feel
mightily good to have a web-space for uploading packages and a CVS
repository, but this is not making advance the Mac OSX port a single inch.
As Pavel-san said several times, patches sitting in IZ or in other
places will not take legs themselves, they will not create a CWS
themselves and they will not make themselves the QA job. Finally, they
will not push themselves upstream. This is a porter's job and if someone
pretends to be porter, she/he is compeled to do the job well.
Maho's motivation:
o Reduce # of patches as far as possible for Windows+Mingw port.
I'm also interested in committing mingw community.
Maho, what assurance has someone that the patches will be reduced.
Looking at the attitude with Mac OSX port, one has some difficulties to
believe it.
Donation request:
o I live in Tokyo and my home is really small with a lot of machines.
There are one middle tower P4 machine, and four-five laptop computers,
and three extreely noisy middle towers at lab, and one-two
laptop computers. so I would like to have a small, slient and fast :)
Pentium4 or Athlon64 machine.
Would it be better simply to use the G5 that was donated to you to do
well the Mac OSX port and not to try to do everything with suboptimal
results. Understand me, I am far from discouraging anybody from doing
the porting. I just know that time is a scarce resource and doing one
thing well is better than doing too many things with lousy results.
I had subscribed MSDN last year, so you don't need to donate additonal
software.
BTW, for building with MinGW, you do not need MSDN subscription at all.
MinGW is opensource and freely downloadable from http://www.mingw.org
At the end, I would like to stress, that if someone wants to do the real
porting work, you are welcome to drop by at
[EMAIL PROTECTED] and ask for help about CWS creation or
any other difficulty you may encounter. Those who tried it can testify
that Pavel Janik or others (or even me in person) are never tired to
help people who want to work for the good of the community.
The procedure, is simple:
1) Create a CWS
2) Dump your patches into the CWS, indicate which issues are solved by
them and set the CWS ready to QA if you are finished with it.
3) Chose a sensible QA representative that will not wait months before
she/he builds the CWS
4) Once the CWS builds well, the QA rep. (if it is someone in the
community) sets the CWS as "approved by QA" and the ball is in the hands
of RE in Hamburg.
From my own experience, between approved by QA and nominated, there is
rarily more then one week.
Sorry for being too direct, but Janík-san tried in many of his e-mails
hint something and his hints were not really understood, so I decided to
go out of the lurking mode and try to put the stick into the ant-hill.
Cheers
Fridrich
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
