[CF-Devel] combining multipart images

[Michael Toennies]

Hm, you are mainly right with your mail but imho you misunderstand the role
of OS/content releated to CF:D and the dx client.

I coded the dx client 2 years ago to give CF the power to grow in the
windows world. It was never my wish to lower or
weaken the unix/linux role inside the project. I for my person would always
prefer a unix server instead of a windows server.

The problem is, that some people on the linux/unix only side of this project
(and in the open source community at all imho) fear that they lose something
when the project grow up in other OS, mainly in windows. First, this is not
what i call "open source or open projects". Then better you say "open linux
only source". Second, its a misunderstanding about the dynamic of projects
and development. The trick to make it a win for all ist to arrange a "win to
win situation". More ppl from windows will invoke more gamers. This will
invoke more map makers - means more content back to linux... etc. Only one
example.

This is a fundamental synergy situation - let me note that i am starting a
company where this kind of synergy effects are one of the pillars of the
business idea.

I dropped the dx client, because it fits not 100% in this (and it was not
designed to be open source). The SDL client is the successor.
He gives all OS the same power.

The CF:D project parts like client, server & editor will have same power for
all OS. Means you use it in the same on windows AND on linux. A linux user
must never even touch windows for every part of the system - EVERY part use
a native OS based library. Editor use java, SDL use low layer and direct
hardware and the server is plain C textconsole (this should be native for
all to). And thats the "trick".

Now you can include a feature or idea in every part of the system - and it
will reflect in useful, native way in every OS java and sdl will run.
The system is a true, real linux system. And its a true, real windows
system. CF:D will not in a single way less linux based as the original CF.

Its very simple: i want to serve gamers - not OS. After i have installed the
OS releated stuff, i want never think again about different OS.

Also, there are some "gaming standards" out. They are not defined by
windows - more from consoles. This means there is a "state of art"
what and how you use gfx, sounds, music and interface. A widget client like
the gtk can't serve this standard. Dot. Hand down. SDL can.

If you see CF:D in aciton, you will understand what i mean. Be sure you will
be impressed.

CF:D is not a "split effort". It is THE split in the next generation. CF has
born a child and thats called CF:D.
What happend here is a fundamental part when not the core idea of open
source.
And i honestly don't think there is a way back. When you reinclude all my
stuff in CF - then you have CF:D.
Note, that i use atm the CF cvs server and just diff him to CF:D. You can
even connect to the public test server of CF:D
at damn.informatik.uni-bremen.de with any CF client - it will run fine (but
you will not be able to really play because the difference is inside the
interface).

In fact , i must now code a version test, because people who connect with
older flat clients to the new servers think the server or
his clients are broken because the map is distorted invoked by the different
tile size. Also, i can then do some changes to the
server->client protocol. ATM, all my stuff i coded for holding it compatible
are still working. After i reinsert the code parts who was removed
as i left CF dev team it runs fine. Damn has not crashed a single time in
the days even wild different clients has connected and the whole arch
set was build up from ground.

I think we will release the first 0.9 version including installed CVS at the
end of the week or in the middle of next week.

MichToen
  Ya I was being a pain on purpose.  I actually have no problem personally
running the linux client, but I do know 3 people who currently play a lot
who would not, and 3
  more people who have played a bit  who also would not use linux. That is 6
to 1.  I also know at least 20 other people who would probably take a look
(I'm working on
  them) but again only if it was a) windows client, and b) easy to install.
26 to 1 then. These people are great gamers with not much interest in
operating systems or in figuring out how to install software - they just
like to play games.  One of my friends who is playing asked me about the
Java MapEditor...you see where this goes. This is why I think that it would
be a very great boon to the game to have a windows compatible client.

  I also think it would not be worthwhile to maintain a separate windows
client - so I agree that the best thing would be a single cross-platform
client (GTK, SDL whatever).  I agree so much that I think private windows
clients would probably be a bad thing (I see the effort that Michael
Toennies is putting into C:D that could be going into CF - not that that is
wrong - that's the system, but it is still split effort).  Having a separate
development for other platform clients almost guarantees that they
  will always be either behind the main development stream and always be
  holding things up, or they will go off somewhere, mutate or die.  There
have been three win clients I know of, all out of date now, and this
indicates to me that this approach does not work.

  I certainly don't want progress on the server to halt to support old
clients, but this tough beans attitude towards a windows client is
troublesome since it is alienating to a whole segment of players who have no
alternative. (is there any idea how many people are using DX client?, MIDs
poll shows it is 26% of their players, but I have no way to make an accurate
prediction from that - is there a way to report on this from the
metaserver?) The feeling I get from reading these forums is that a windows
client is something people maybe don't want to have (or possibly have to
deal with anyway).  It probably isn't planned that way, but it amounts to
the same thing.  I know CF was a unix game, and if that is what it is to be
then that's fine, but currently it isn't a unix only game.  I was, in my
snarky way, making the point that you will loose these people to other games
if they cannot play - likely they will not return.

  I certainly don't expect you to do this - develop or maintain other
  clients. I have not yet contributed a speck of code to CF (although I am
trying to work on some stuff),  I wouldn't tell anyone what they should
contribute.   Somebody should do it however, and the only way it will get
done is to make mention of it- often, maybe even make it an priority item.
The fact is that if it is made a priority it has a better chance ofgetting
done (If people want it to be done...) I can understand wanting to have
concrete evidence that work was being done on a windows client before
delaying updates to the server code, but if you are proposing changes that
will cut off a third of your players (again extrapolating from the MIDs
poll - really should get more info on this...), it should be highlighted in
the proposal - contingency plans made for it (a server for windows client?).
I wanted to flush this out into the open, since I think the client is the
most important part of the game, and players are the life blood.  I also
think it is a great game or I wouldn't be bothering sticking my nose into
this.

  Anyway hope I didn't offend.  I know it was just a proposal - I was
responding to the proposal.  (I would like to get away from multipart
monsters too...)
  -TM
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 
     
     http://shadowknight.real-time.com/pipermail/crossfire/attachments/20020508/90e6b4aa/attachment.htm