ProtoCiv: porting Freeciv to Python

[Servé Lau]

As somebody once said, the purpose of paying wages is not to motivate
people to work. It's to motivate them to come in on Monday mornings.

Think of the boring bits as commercial software's secret weapon ;-)

I agree, most open source is not finished or crappy implemented. At some
point most people lose interest especially when it's not successful.

Yes, there are exceptions.

 

[Brandon J. Van Every]

So why waste their time even asking if they would consider
aligning their goals with yours? That seems like the ridiculous
part- OF COURSE hobbyists are not going to want to take
commercial concerns into account, THAT'S THE POINT.

Reminds me again of the joke about the 2 economists:

Two economists are walking down the street. Near the gutter, a $20 bill
flutters. One economist looks down, looks at the other, and exclaims, "Hey!
There's a $20 bill on the ground by the gutter!" The second economist says,
"Nonsense, if there was someone would have picked it up already" and they
keep right on walking.

In other words, 6 months ago I had no reason to assume that hobbyist labor
would be completely useless to me. I knew they wouldn't have the same
agendas as myself, but I thought they might have sufficient overlap to get
something of value to me accomplished. I was mistaken. Linux hobbyist game
developers of GPL flavor are of no use whatsoever to Windows commercial game
developers of BSD flavor, even if both are doing open source and might
benefit from common tools. That may seem obvious to you, but it wasn't to
me, and I'd question why it should seem obvious to you at first glance. I
looked at a *lot* of open source projects in the name of avoiding Not
Invented Here. In hindsight, I should have gotten back to my own Ocean Mars
code a lot sooner.

 

[Brandon J. Van Every]

And guess what bits of a project OSS programmers often tend not to
bother finishing?

As somebody once said, the purpose of paying wages is not to motivate
people to work. It's to motivate them to come in on Monday mornings.

Think of the boring bits as commercial software's secret weapon ;-)

Some boring bits are easily accomplished, and tractable through incremental,
communal effort. Others take too much pain to bother with. I think in
theory, commercial developers could alleviate themselves of certain boring
problems through BSD open source projects. A good example would be
http://nebuladevice.sourceforge.net . Of course, that project would never
have gotten off the ground without the up front contribution of Radon Labs,
but since they did decide to do that, it's a viable 3D engine for solving
various people's problems.

In practice, there's all kinds of "Not Invented Here" and "Make Myself
Invaluable" out there. I see a lot of political reasons why human beings
will never cooperate to save humanity much labor. Even when projects *are*
BSD licensed, like Python for instance, you can get people refusing to
*market* a product properly for predictable political reasons. Distaste for
commercial developers, "suits," other people's problems, etc. Consequently,
relevance is diminished. When relevance is diminished, the available
solutions become fractured and things don't work well together. Redundant
labor is performed because people gotta be different, they gotta try to make
everyone else march to *their* drum instead of being more inclusive.

--
Cheers, www.indiegamedesign.com
Brandon Van Every Seattle, WA

20% of the world is real.
80% is gobbledygook we make up inside our own heads.

 

[Brandon J. Van Every]

Then you don't own the second edition, which came out a year
before that "12 years ago" comment you made. Had you said
14 years ago then things would be different.

Oh please **** OFF!!! 12 years, 13 years, 14 years, awhile ago! Why are
you wasting our time on this nonsense??? "Computer Graphics, Principles &
Practice," 2nd. ed, 1990. ISBN 0-201-12110-7. Acquired my copy at Cornell
U., spring semester 1991. NO GODDAMN TEXTURE MAPPING TREATISES. In those
days, you pulled your "advanced" papers on texture mapping, and unless you
owned very expensive 3D HW you implemented them by hand. PITA. Never
bothered. Still can't stand texture mapping to this day. What a bunch of
inelegant ugly flat looking crap!

A slight timing
error perhaps, but properly you should have used the publication
date of your book.

Look at what the *original* title of the thread was. Consider where you've
*taken* it. *What* is your fucking point??? Your point was, and is, and
will continue to be... ***ME!!!***

I give up!!! You have won!!! I am driven screaming from the FUCKING
C.G.D.* NEWSGROUPS, NOT TO RETURN FOR QUITE SOME TIME!!!
AAAAAAAAAAAAAARRRRRRGGHGHGHH!!!!

Wouldn't be the first time, either. Probably not the last. Hard for their
founder to stay away forever. In the meantime, we will be MUTUALLY THANKFUL
for the absence!

 

[JanC]

(e-mail address removed) (R. Alan Monroe) schreef:

"Hello World"? :^)

You know, some early computers couldn't even do things that simple. ;-)

 

[JanC]

Never!!!

Emacs, on the other hand.... <duck>

vi can't lose that competition.

You can always implement vi in emacs if you want an editor, but you can't
implement emacs in vi if you need an OS.

BTW: I don't care about vi vs. emacs, I use SciTE.

 

[Andrew Dalke]

R. Alan Monroe
My favorite program for this is /bin/true. On some machines it is 0
bytes long. It works on Unix and, with a ".bat" extension, on MS
Windows.

Andrew
(e-mail address removed)

 

[Owen Jacobson]

My favorite program for this is /bin/true. On some machines it is 0
bytes long. It works on Unix and, with a ".bat" extension, on MS
Windows.

[root@eidolon root]# uname -a
Linux eidolon.lionsanctuary.net 2.4.20-24.9 #1 Mon Dec 1 11:43:36 EST 2003 i686
athlon i386 GNU/Linux
[root@eidolon root]# ls -l /bin | grep true
-rwxr-xr-x 1 root root 10172 Oct 29 06:44 true
[root@eidolon root]# ldd /bin/true
libc.so.6 => /lib/tls/libc.so.6 (0x42000000)
/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)

Not this machine, apparently.

 

[Dennis Lee Bieber]

JanC fed this fish to the penguins on Thursday 15 January 2004 20:27 pm:

You know, some early computers couldn't even do things that simple.
;-)

.*.. *...

.**. .*.*

.**. **..

.**. **..

.**. ****

..*. ....

.*.* .***

.**. ****

.*** ..*.

.**. **..

.**. .*..



--

 

[Eternal Vigilance]

Congradulations. You finally discovered that people will not do your work for
you
for Free. Open Source projects produce 99% garbage and thats for the projects

that already have a good working 'product'.

If you want it done right, do it yourself. (Either that or get alot better at
talking the
right people into doing the work and then manage them without interfering too
much)

 

[=?iso-8859-1?Q?Fran=E7ois?= Pinard]

[Andrew Dalke]

My favorite program for this is /bin/true. On some machines it is 0
bytes long.

Yet, the AT&T empty file for True is heavily copyrighted. Stretching
the meaning a bit (maybe more than a bit!), it amuses me to think that
nobody may produce an empty file anymore without being indebted to them.

 

[Nick Vargish]

I've also become extremely disillusioned with Open Source developers.
Basically, for a commercially minded developer such as myself, I don't think
they're any value add at all.

"I've become very disillusioned with cats. For a stick-fetching minded
person like myself, I don't think they make good pets at all."

Well, duh!

I'm also starting to downright despise Linuxers. If I see another project
that's "cross-platform, so long as your platform is Linux," I'm going to cut
someone's head off.

And you seemed to want something cross-platform, as long as it runs on
..NET. I don't see your position being any better, from a
cross-platform standpoint, than that of the "Linuxers".

Wait, let's look at the FreeCiv download page:
http://freeciv.org/download.phtml... I even see Win32, Mac OS-X, and
OS/2. That seems pretty damn cross-platform to me.

Rather, it's always the induhvidual's responsibility to
download and install all the needed component libraries.

If their distro doesn't include the components, they will go and get
them, or switch to a different distro. Your preferred platform comes
in only one distro, which means there's no incentive for the provider
to include a wide selection of components. Actually, your provider is
disincentivesed from including components other than the ones they
control.

Other than being frustrating for you, I think the Linuxers have made a
much more rational choice of platforms than you have.

Or requiring Cygwin or MinGW, which real Windows developers don't want to
use.

Where "real Windows developers" == "Brandon Van Every". I see plenty
of open source Windows projects with compilation instructions that
require the use of Cygwin or MinGW.

Whether my Freeciv VS .NET 2003 build makes it into the official Freeciv
source pool remains to be seen. The Freeciv people are, after all, a bunch
of Linuxers.

I'm guessing it won't, especially since rejecting it means they won't
have to continue dealing with someone like you.

Consider me trolled,

Nick

 

[Corey Murtagh]

"Brandon J. Van Every" <[email protected]> writes:

Where "real Windows developers" == "Brandon Van Every". I see plenty
of open source Windows projects with compilation instructions that
require the use of Cygwin or MinGW.

Actually I kind of agree here. I've had plenty of problems with Cygwin
- although less with MinGW. An OSS project that requires a Linux
emulation shim to run on any other platform is not, IMHO, cross-platform
at all.

 

[Erik Max Francis]

Actually I kind of agree here. I've had plenty of problems with
Cygwin
- although less with MinGW. An OSS project that requires a Linux
emulation shim to run on any other platform is not, IMHO,
cross-platform
at all.

Problem here is, Freeciv _does_ have a native win32 binary available.
So the entire premise that Brandon was starting with was totally false.

 

[Nick Vargish]

Actually I kind of agree here. I've had plenty of problems with
Cygwin - although less with MinGW.

It's always performed flawlessly for me. Have you reported the
problems to the Cygwin folks, or tried to fix the issues yourself?

An OSS project that requires a Linux emulation shim to run on any
other platform is not, IMHO, cross-platform at all.

Linux != Gnu, though to some that might seem like splitting hairs.

What other free development environments are there for Win32 that
offers power and flexibility comparable to the Cygwin tools? Visual
Studio is not free (quite the opposite!), and for many OSS developers,
that makes it impractical.

Sorry, I realize this is getting pretty off topic, especially for the
Python group, which is where I'm reading it.

Nick

 

[Chuck Spears]

The conclusion may seem obvious to _you_ but this is no guarantee that

everyone else also possesses this knowledge. OSS is being hailed as
the second coming, and it comes as no surprise therefore that some
people might be deluded into thinking they could harness this power to
cure cancer overnight or land a man on Mars by 2005.

If you can get by his trolling and unbearable arrogance, there are
some kernels of truth in there. I come from a commercial background
as well but instead of trying to exploit the OSS community, ive been
lurking around looking for a project I feel I could contribute to.

Most geeks by their nature, are very independent and abhor order.
You have to have order to get anything done. Thats why most
successful projects have one or at most a few people pulling the
strings because if you don't, the project will flounder. I've
personally based a few of my projects on some OSS projects and they
failed miserably. Because of the bugs and the Authors unwilligness to
address them or even accept them as bugs. You can say "well why didnt
you just fix it yourslef" but I just didnt have the time.

On the other side of the coin, i've used OSS projects like PHP and
Postgres with great results.

The other problem with hobbyist geek programmers is they are just in
it for the fun of it. They get bored when the last 10% of the project
which is mostly bug fixing and rengineering code coes about and
generally abandon it. I've poked into literally over one hundred
sourceforge projects that started out as good ideas and had lots of
activity. i'd come back in 6 months and There would be almost no
activity. With a developer with a commericial background, he might be
more willing to see the project through.

I too get annoyed when an OSS author pulls a massive library into his
project just to get a few functions out of it he could have written
himself. It's really problematic as Brandon has said when you are
using CygWin or Ming because a lot of these libraries dont work on it.

My bread and butter is palm and windows development so I cannot
abandon the platform yet. What i've been trying to do is build up a
nice linux dev environment using ming and python. The python side of
things works great but the C++ side... well.. sucks.

Thankfully, I finally decided to evaluate VMWare and being able to run
linux and windows together has been a godsend so I can hopefully
abandon the cygwin stuff.

 

[Paul Boddie]

[Quoting someone else...]

....by generally clueless analysts and journalists whose day job seems
to focus exclusively on speculation and rumour forwarding, switching
over to hyping Microsoft and other vendors "du jour" when they feel
the need to encourage a "debate".

If you can get by his trolling and unbearable arrogance, there are
some kernels of truth in there. I come from a commercial background
as well but instead of trying to exploit the OSS community, ive been
lurking around looking for a project I feel I could contribute to.

Yes, this is the key to the issue. If you regard open source software
as a huge pile of "free stuff" to plunder, then whilst you may get
some productivity benefits in the short term, the longer term will
most likely bring maintenance issues as the community continues on its
own course and you continue on yours. (If you actually have a course
of developing the code, that is, as opposed to just dropping the
plundered code into some directory and virtually forgetting that you
have it.)

Most geeks by their nature, are very independent and abhor order.

I'm not sure that I equate "open source developer" with "geek" in the
same way that this generalisation demands, but if you mean that most
open source developers who are working on projects in their own time
or under their own motivation won't take orders from newcomers, I
think it's quite obvious why that is.

You have to have order to get anything done.

True, but numerous open source projects have "order" without
conventional forms of management. On the other hand, if you're saying
that a bunch of "elite" 13 year olds on IRC can't scale their project
beyond a slightly larger group of well-trained primates, then you're
making an obvious point.

Thats why most
successful projects have one or at most a few people pulling the
strings because if you don't, the project will flounder. I've
personally based a few of my projects on some OSS projects and they
failed miserably. Because of the bugs and the Authors unwilligness to
address them or even accept them as bugs. You can say "well why didnt
you just fix it yourslef" but I just didnt have the time.

There are lots of reasons why developers can't or won't fix the bugs.
It's quite a common phenomenon in the commercial software world, too,
but the reasons are more likely to be political there. Meanwhile, you
don't need to be a conspiracy theorist to understand developer
motivations around such issues in open source projects; quite often,
developers don't have access to the same environment and can't
reproduce or diagnose the bugs that you're experiencing, for example.

On the other side of the coin, i've used OSS projects like PHP and
Postgres with great results.

The other problem with hobbyist geek programmers is they are just in
it for the fun of it.

How many people with a hobby do it because they don't enjoy it?

They get bored when the last 10% of the project
which is mostly bug fixing and rengineering code coes about and
generally abandon it.

Perhaps many projects need assistance from real-world users in order
to get that last 10% done. Consider cases like The Gimp where the
original developers basically dropped the code before the 1.0 release,
but where interested parties picked it up and finished the job. If
every project suffered from the same symptoms that you describe, there
wouldn't be such a thing as GNOME, Sun's Java Desktop System and so
on. Perhaps JDS would be based on KDE instead, however. ;-)

I've poked into literally over one hundred
sourceforge projects that started out as good ideas and had lots of
activity. i'd come back in 6 months and There would be almost no
activity. With a developer with a commericial background, he might be
more willing to see the project through.

Yes, perhaps because he'd be paid to see it through.

I too get annoyed when an OSS author pulls a massive library into his
project just to get a few functions out of it he could have written
himself. It's really problematic as Brandon has said when you are
using CygWin or Ming because a lot of these libraries dont work on it.

Just because one is developing with open source software doesn't mean
that one is suddenly exempt from normal software engineering
principles. Perhaps that was what the quoted contributor meant by
those "second coming" claims, but most people understand that the
universe doesn't change its rules on the basis of rumours derived from
misinterpretations of claims by Eric S. Raymond.

Paul

 

[Tim Rowe]

[Andrew Dalke]

My favorite program for this is /bin/true. On some machines it is 0
bytes long.

Yet, the AT&T empty file for True is heavily copyrighted. Stretching
the meaning a bit (maybe more than a bit!), it amuses me to think that
nobody may produce an empty file anymore without being indebted to them.

Ah! So /that's/ why most versions contain a revision control comment
(and nothing else).