|| ||dev-talk <wesnoth-dev-AT-gna.org>, Ivan Illarionov <ivan.illarionov-AT-gmail.com>|
|| ||Re: Wesnoth refactoring and future direction plan|
|| ||Tue, 06 Jan 2009 14:28:27 -0500|
|| ||dev-talk <wesnoth-dev-AT-gna.org>|
|| ||Article, Thread
Quoting Ivan Illarionov <firstname.lastname@example.org>:
> On Mon, Jan 5, 2009 at 7:43 PM, Eric S. Raymond <email@example.com> wrote:
>> This looks like a decent start on a plan, and very much the sort of thing
>> that can be done on a branch. I'd like to see a round of questions from
>> other core devs, and *answers* to those questions, before we create
>> the branch
>> and go forward from there.
> I will answer any questions in detail. The main reason I didn't
> answered on some previous questions before is because I wasn't
> prepared and didn't want to piss people off (I pissed them off
> anyway on IRC though -- again, sorry for that).
> My honest answer to some questions on first days could be something like:
Let me try to explain something very very clearly to you, since you
don't seem to understand it. It's very important that you understand
this before moving further (in life, actually, as this is related to a
basic concept in human communication).
In the above referenced email, someone is trying to convince Linus
that he should change a project he founded to use C++. Linus is in a
position where he can go on a profanity-laced tirade against C++ if he
wants. He is in control of the project. He makes the decisions. He
doesn't have to convince anyone of anything. He can just say "this is
the way it will be" and that is the way it will be.
However, you are in the opposite circumstance. You are in the
circumstance of trying to convince the Wesnoth developers to make
changes to Wesnoth. You have no power to get us to do anything. The
only way you can effect change (other than by forking, which you're
welcome to do at any time) is if you convince us of something. To
convince anyone of anything, you will have to be polite, courteous,
patient, and forthcoming. By being rude and obnoxious, you only
convince us to ignore you completely, and you won't get anywhere.
This is the case for anything in life: if you want to convince someone
in a position of power to do something, you will have to be polite and
diplomatic. I doubt that Linus convinced anyone of anything in the
above tirade. He probably made people who already think C is an ideal
language for such development to think "well said!" and further
alienated people who would prefer C++. This was probably his exact
intent: his aim isn't to convince anyone of anything, it's simply to
put an end to people complaining or arguing or disagreeing.
Trust me, if you sent an email to the Linux Kernel Developers Mailing
List entitled "Linux refactoring and future direction", you would
receive MUCH more hostile responses then you have received here. I
think we have been incredibly patient with you, but our patience is
going to run out.
You have irritated our developers with pointless posts and ranting and
so forth. I suggest that you go away and not post here again until you
actually have a completed patch that is ready to go into Wesnoth and
does something useful.
Finally, to Linus's points, one must understand that Linus works with
very low level development. Linux is a very low level program. Git
isn't quite as low level, but is still fairly low level. Personally, I
think that using C++ in the kernel would be rather insane, for the
reasons he mentions. Using it on git might be a little more
legitimate, but I think C is a good choice for that also.
Wesnoth is a much higher level application. I don't think that C is
appropriate. I think that C++ is a very good choice. You disagree? I
don't really care. I've written plenty of good quality working
software in C++ that is used by many thousands if not millions of
people. I know what I'm doing. So do Wesnoth's other developers. We
don't need anyone coming along telling us how to engineer software.
> I'm glad that I didn't answer, and, now, when I'm prepared and ready
> to explain and justify my thoughts in a friendly way, I will answer
> any questions. And, BTW, Linus has some very good
> points there that could explain why I use plain Python/C API, for
If you want to develop for Wesnoth and have your changes accepted you
are going to have to develop in Wesnoth's way using its style. It
doesn't really matter how much you disagree with it. We're not going
to change our minds on your say so.
Basically at this stage I would say your best option would be to go
off and program your own game in Python and show us how much better
you can do than Wesnoth. However, if you really want to persist with
this "project" you are welcome to. Please only post once you have
complete working patches that do something useful and are ready to
show us. We've wasted enough valuable developer time on all of this
> I very want to find the agreement on all questions and make this game better.
> Wesnoth-dev mailing list
to post comments)