delorie.com/archives/browse.cgi   search  
Mail Archives: djgpp/1995/06/06/15:26:33

Date: Wed, 7 Jun 1995 02:16:16 +0900 (JST)
From: "Stephen J. Turnbull" <turnbull AT turnbull DOT sk DOT tsukuba DOT ac DOT jp>
To: "Junaid A. Walker" <junaid AT wilma DOT eng DOT monash DOT edu DOT au>
Cc: djgpp AT sun DOT soe DOT clarkson DOT edu
Subject: Re: VOTE: CODE STANDARDS

I strongly considered making this message private.  Unfortunately, since 
a couple of people are persisting in pushing for *other* people to do 
them favors (it's not just Junaid), the point needs to be made publically.

If you need something and can't do it yourself, ask.  Maybe someone will 
do something about it.

If you can help fix it yourself, do.

But don't tell other people what they need to do, especially not in terms
of "even demo coders have standards."  You want to *tell* someone what you
want done first, go talk to Bill Gates.  I'm sure there's a price at which
he'll contract to give you anything you want.  By 1995^H6^H7^H, oh, who
knows....  Maybe it would be better to pay DJ. 

Detailed flame follows:

On 6 Jun 1995, Junaid A. Walker wrote:
> 
> On Tue, 6 Jun 1995, Eli Zaretskii wrote:
> > > 	Junaid wrote;
> > >	Goody, because you wont mind me commenting on the coding
> > > style of go32 and friends. Please dont take offence, just my humble
> > > opinions.
> > 
> > Why was it necessary to send this to the entire DJGPP public?  Wasn't
> 
> 	Well since it concerns djgpp in general then it goes to all
> concerned.
> 
> > private mail to DJ enough?  Critics on the coding style is as close to
> > personal as you can get with programmers, don't you think?
> 
> 	Not really (i was appologetic in the original message).

Right.  Apologetic.  Excuse me while I scuff your shoes and kick your 
kneecaps.  :-)

Nope.  Somehow the smiley doesn't make that funny, either.

> 	Well still doesnt change the facts - the code is basically
> undocumented, and has been that way since day one. Maybe its time
> we started documenting things (emx is a good example). Sure if we
  ~~
Oh, goody, we have a volunteer!!  Right?

> turn a blind eye to the state of the code, we'll all be fatter
> and happier, but
> i think we have to face the fact (esp if he has a newborn) that
> dj will not be with us forever.
> 
> 	Remeber standards work. The unspoken word and anarchy

Huh?  We're talking about djgpp, not alt.religion.scientology.

> disintegrate rapidly. Who cares about dj anyway, he develops djgpp
> for his own personal interest, not for yours and not form mine.

And he owns the copyright.  If DJ doesn't do it, it's not DJGPP.

> His stance is politically correct; i do it for fun, if you want to
> do things better, do it yourself. Enter me, i want to use djgpp
> for real work. Hence i want the greatest milage possible; unfortunately
> there are no roadmaps, and the path has washed away. Hopefully dj

No.  There's the code itself.  There are about 10MB of mailing list
archives, which you can get from delorie.com (and parts from
turnbull.sk.tsukuba.ac.jp).  And there's DJ, Charles, Morten, and many
others who have contributed snippets of code here and there.  You have a
question, post. 

> can guide the djgpp comunity enough that it can stand on its own 
> feet.

Or maybe the community would prefer an early release of V2.

> 	Also as i previous indicated, documentation tends, in the
> long run, to reduce design/test/debug/developer time. Since dj is
> obviously with djgpp for the long run, he might consider this.

Um, a few lines back, DJ was going away?

> As for new comers (like myself, on and off the last few years),
> we should definitely be concerned about this issue. But this stigmatic
> issue has never been discussed before. A tradegy. Just be objective,

Did you grep the archives to find out whether it had been discussed 
before?  There certainly have been complaints....

> and decide whatever you think is most beneficial.
> 
> 	It would be wise to put it to a vote; "does djgpp need some

What do you mean, vote?  Obviously you think you know the answer.  Are 
you just asking for public censure of the DJGPP development team?

> coding standards?". Any self respecting programmer would have to

DJGPP has coding standards.  You may not like them.  (Others don't: I've 
sure seen enough "why do the
go32_dpmi_simulate_interrupts_and_do_many_other_things_so_that_the_name_will_be_real_long()
functions have such long names?  They're hard to type" posts.)  Surely,
they're not optimal from some pie in the sky viewpoint.  Me, I'd like to
have version 2 sometime in the near future.   And POSIX compatibility, 
which is one of the big time-eaters for the DJGPP team AFAIK.

You think DJ remembers off hand the reason for every little tweak in the 
code?  Especially those that were contributed by others?  For a piece of 
software that's going to go away in the near future, the documentation of 
GO32 would be a horrible amount of effort.  Right?  Now, if you've looked 
at V2 and see similar problems, that might be a lot more important.  
But...  Who's gonna do the writing?

As for votes.  huh.  When DJ found that there was no way to run GNU
compilers under DOS, he voted for a solution.  He created GO32.  When I
complained that v 1.12 wouldn't build Ghostscript, I voted for a solution. 
I created the yaseppochi-gumi server and put v 1.11 release software on
it.  Eli Zaretskii voted for a better solution to the same problem.  He
found the bug in GO32's long command line parsing that probably caused it. 
When A.APPLEYARD (among others) found no GCC docs readable by the software
he wanted to use, he voted for a solution.  He created ASCII docs.  Hell,
you just uploaded gdb 4.12 for V2 if I understand correctly. 

Those are the only votes that count.  You know how to vote, apparently.  
So get to it.

Discussion, OK.  Say your message had started with "I need some features 
from GO32 that aren't there, but I can't figure out how to put them into 
GO32.  The documentation is really sparse.  Besides just fixing this 
problem I thought it might be a good idea to put some effort into 
documenting the GO32 source code.  I'm not going to do it all by myself, 
but if a couple other people would like to help, I'd like to coordinate 
the project."  I wouldn't be in this discussion at all.  I'm not really 
competent, nor do I have time, to try to comment the GO32 (or CWSDPMI or 
V2 libs or crt0.o) sources.  But I'd be happy to see others do it, and I 
sure wouldn't have any complaints.  I don't want it bad enough to do it 
myself, so I'd stay shut up.

I do suspect that there aren't enough people who really see the need for 
special extensions to GO32 to make it worthwhile.  Yeah, the people who 
are developing for DPMI 0.9 environments have problems with graphics.  
But they may as well wait for V2 anyway....  Or get the alpha/beta 
versions available at delorie.com.

> agree. Even demo coders have standards (and commented code i must
> add) ....

Good for the demo coders.  Most of the demo coders I know of get paid to 
maintain those standards.

> 	Cheers,
> 	Junaid

ciao.

Steve

- Raw text -


  webmaster     delorie software   privacy  
  Copyright 2019   by DJ Delorie     Updated Jul 2019