Guild of Writers

It seems to me that a lot of people underestimate the work involved in updating pyprp. Commonly saying things to the effect of "if only we had the source, we could be running fan ages in moul tomorrow". As I've said before, the limiting factor there is not a lack of knowledge on the file format for moul, (try any moderately recent plasma related tool besides pyprp and see if it will open your moul files) but rather the amount of work required to find a way to work around blender's interface, (I'm given to understand that this is much less of a problem in blender 2.5, but that will require a complete rewrite of pyprp. (time for the mythical 2.0 to appear )) and the general lack of a working environment to test the output in. Thus, what we really need are working standalone moul servers for testing on. Naturally, I'm all for the source, but I'm not about to stop cyan from giving us *something*. However, it's really the servers we need, not the plugin.

*raises a hand* I have a question. What does Cyan mean by "ancient version of 3DS Max"?

I have 3DS Max 8. I would love the binary, assuming it will run on 8. So, y'know, while everyone else wants source, I just want to get started NOW PLEASE.

But I'm also well-aware that most people don't have Max, so it's less useful for them. I would be delighted if we got both binaries and source.

1- Server binaries (which should mean zero work for Cyan)
... We need to set up a test environment.
2- Plug-in binaries (which should mean little work for Cyan)
... we don't want to know where Writers get their 3DSmax)
3- Server sources (which should mean little work for Cyan)
4- Plug-in source (as soon as Cyan can release them)

Hmm, revise my preference to what chacal just said.

I generally agree, Chacal, but I would put the server sources move up the list... I doubt it means "zero" work to release the server binaries, and I agree that since PhysX, Bink and so on are not part of the server, their sources should be very little work. I would even say if releasing the server sources would take, say, about three times as long as the server binaries, don't waste your time with the latter. Binaries need documentation to be usable, while source can explain a lot in itself if (and only if) it is commented fairly reasonable.

(I'm given to understand that this is much less of a problem in blender 2.5, but that will require a complete rewrite of pyprp. (time for the mythical 2.0 to appear ))

Yes, I think that's the way to go. If I only could get my head around Blender.

Chacal and Trylon: I hear ya. But because Mark said:

the next goal should be to get one or more of the fan created content on the MOULagain servers. To me, that would show the most promising thing about the entire project.

we gotta leave servers for later. So, I put you down for plug-in binaries.

Also, here's the fun part: we'll need to reach a consensus in order to have any affect on their decision. So, if you're absolutely sure your choice is the right one, try to convince the other people to switch to your side. Think of this as an American jury, except instead of 12 angry men, you have everyone that posts on this thread - and we can't leave the room until we all agree. If we're split on this, then this debate won't do any good - they'd declare us a 'hung jury.'

P.S. Nadnerb - one more post, and you'll reach 1000!

chacal sorted 4 points.
when we gotta leave the servers, only two points remain.
binaries or source.

I consider this as easy........ the source code.

Robert The Rebuilder wrote:(...) because Mark said:

the next goal should be to get one or more of the fan created content on the MOULagain servers. To me, that would show the most promising thing about the entire project.

we gotta leave servers for later. So, I put you down for plug-in binaries.

To be honnest that quote has me a little worried: while it does make some sense, it also seems to me like Cyan still needs some motivation to move the project forward.. Do they not believe in us? Have they not seen what we have done for the past couple years? I might be reading too much into this though.
At any rate it gives us a clear direction of work: polish our Ages to motivate Cyan some more!

Chacal wrote:1- Server binaries (which should mean zero work for Cyan)
... We need to set up a test environment.
2- Plug-in binaries (which should mean little work for Cyan)
... we don't want to know where Writers get their 3DSmax)
3- Server sources (which should mean little work for Cyan)
3- Plug-in source (as soon as Cyan can release them)

4- Client source.

That's going to be a looong party then, I doubt the Uru community or even the guys doing msot of the hacking ever agreed in it's long history...

the next goal should be to get one or more of the fan created content on the MOULagain servers. To me, that would show the most promising thing about the entire project.

Well, he then has to answer the question how we are supposed to make any MOUL-compatible age without the slightest possibility to test it? Of course we can test it in POTS, but that's only of limited value. A simple proof-of-concept might work, but everything more complex has to be tested with the actual engine we are writing for. The value of a plugin, even with the sources, is very limited if the output can not be tested by us.

4- Client source.

You mean, one day we might actually have the source for the full toolchain, maybe finally stopping to write silly work-arounds in the server and Python code to fix those nasty bugs hard-coded in the client? That must be paradise

Aloys wrote:Do they not believe in us? Have they not seen what we have done for the past couple years?

I think they do believe in us, and they're impressed with what we've done - heck, Chogon served as a judge in two of the RAD contests - so they'd like to see our work in MOULagain soon.