Questions about this so called.. open source thing.

Chaser The Wolf · **Joined:** Sun Dec 06, 2009 3:49 am **Posts:** 1

i have read several post about it saying "where can i download the code", many reply its not available yet, here are my questions about it:

1: Why tell everyone its open source and not have a code released yet?
2: I have read your waiting for the staff/developers to get back, why not just use an SVN and update it when the developers get back(If there's important codes in there, just replace it with ##IMPORTANT PRIVATE CODE##, so no one can use that part).
3: If there's a release date on it, whats the time(i hear its this month, is it true?)
4: dang it i forgot the 4th question >_>

btw, this is my first post, im not sure if this is where i need to post this, if it is incorrectly place, please move the post.

Tai'lahr · **Posted:** Sun Dec 06, 2009 4:32 am

Shorah and welcome to the forum, Chaser. Yes, your post is in the right section.

Chaser The Wolf wrote:

1: Why tell everyone its open source and not have a code released yet?

Cyan hasn't said that it is open source, just that it will be, and judging from your second question, I think you see what the hold up is.

Chaser The Wolf wrote:

2: I have read your waiting for the staff/developers to get back, why not just use an SVN and update it when the developers get back(If there's important codes in there, just replace it with ##IMPORTANT PRIVATE CODE##, so no one can use that part).

Something like that has already been suggested here and it appears it may be taken under serious consideration re this post by RAWA.

kaelisebonrai · **Joined:** Sun Jan 20, 2008 2:14 pm **Posts:** 902

Tai, as much as this community is willing to wait, the rest of the world is finding it pretty strange (to say the least), that Cyan first mentioned making the toolset open source some point last year, they then mentioned making the whole thing open source not too long after that... and all the while there's no information on the license, or any note of exactly what the progress is. I know the FreeGameDev (not freeware, it was an open source game dev community) community was finding the lack of information highly suspicious, and it certainly does look like it never will be. We might know better, but, the trouble is, when Uru goes open source, this isn't just about us, anymore.

Quote:

So, Cyan has decided to give make Myst Online available to the fans by releasing the source code for the servers, client and tools for Myst Online as an open source project. We will also host a data server with the data for Myst Online. MORE is still possible but only with the help from fans.

This certainly does look as if they already /have/ done so, so, really, the question about "why isn't there any source code available if they say it is open source?" Is certainly valid.

JWPlatt · **Posted:** Sun Dec 06, 2009 3:27 pm

kaelisebonrai wrote:

the trouble is, when Uru goes open source, this isn't just about us, anymore.

Bingo. "This isn't just about us anymore" is a message I've been trying to push since forever because the ramifications still have not seemed to really sink in with many folks. But it's only "trouble" for those who are in this for themselves and the hacker mystique of an extremely limited set of skilled people. For those interested in a larger vision of worldwide contribution and helping Cyan potentially make a bigger success of Plasma and Uru, this is everything but "trouble." It's the biggest opportunity the community has ever had.

kaelisebonrai · **Joined:** Sun Jan 20, 2008 2:14 pm **Posts:** 902

I'm not sure, JWPlatt, but, are you suggesting the "hacker"-types are the ones that find this "trouble"? Since... I find the opposite is the case, but, granted, I could be interpreting what you're saying horribly wrong. However, don't get me wrong, either, its not about Cyan, either. This is bigger than the community we know right now.

A number of the current hackers, as far as I can gather, are well aware of what this means, I find those outside of that group are far less clued in on what this is really about. This isn't purely about this community, or Uru, either. Plasma will be more than just Uru, or Cyan games, anymore, too. Its going to be a bit of a bumpy ride for some people, but, its going to be pretty good for everyone, should this community hold onto its identity, though it'll more than likely become but a subsection in the plasma-engine community that will eventually spring up, with a bit of luck. The trick is to not let this, the Uru community lose its identity as such. But, put simply, Uru will not be the only driving force behind the source code, I'd hope. It'll simply be just one game built on an engine used for many things.

Keep that in mind,.

By the by, JWPlatt, my "trouble" was referring to the way people still believe this is just about the uru community. This has further reaching consequences than just our community.

JWPlatt · **Posted:** Sun Dec 06, 2009 7:16 pm

It seemed you were writing that it was trouble. Thanks for clarifying.

kaelisebonrai wrote:

Uru will not be the only driving force behind the source code... It'll simply be just one game built on an engine used for many things.

Right. You get it. And that could come back around to benefit the Uru community. And more importantly, there's increased opportunity for Cyan to exploit its expertise with the engine and Plasma content by offering services beyond their own games and IP. MQO could just be the first of many such projects.

Nalates · **Posted:** Sun Dec 06, 2009 8:52 pm

I suggest not making too many assumptions about any part of the Uru community and what they are or are not thinking. We tend to talk Uru here because that is what we have in common. That does not mean fan thoughts are not encompassing outside ideas (UberThend?). I also have a belief that only the new, innocent or thick skinned will hazard voicing exotic ideas here.

While we have even less information to speculate on about in what other people that may be interested, in a general way we can we know what the gaming community does. I think ODE and PhysX or Havok show how the game community uses and adds to both full open source and proprietary tools.

I can imagine PyPRP becoming something more like the Blue Mars’ building tools. Or Plasma may already be something like that, I doubt it. Or if one considers it to be more of the game server and client skeletons then it may become something more like OpenSim, which is now the basis for several virtual worlds and vw systems. Whatever happens with Plasma it will be interesting.

I’m not sure how JWP sees open source Plasma use helping Cyan. I would think that would mean they would just have more competition as people use Plasma for various games.

Cyan does not seem to be in a place to advance Plasma development, which would be happening out in the OS arena. I only see them developing it as a means to accomplish another task such as MQO. So, it would be a cost spent for efficiency and necessity rather than as its own profit center. The development in PhysX and Havok is fueled by other commercial interest; selling hardware. While OpenCL has sprung up and come to first release and at least one update in about a year and has many adopting it, ODE is taking years to develop single features and mostly being adopted by the open source and independent game developers community.

How do you see Plasma reaching outside the Uru community working in Cyan’s favor, JWP?

kaelisebonrai · **Joined:** Sun Jan 20, 2008 2:14 pm **Posts:** 902

Nalates, I see it being useful to Cyan, as they, being the developer who has used and developed with that engine for many years now, have knowledge of it, and can support it. And, they can offer premium support at a fee. Its been done before with a lot of success, actually. Its quite a common way of generating income from an open source release, and that income also allows for further developement, of their own projects.

It could also be useful to Cyan in the same way that Open Sourcing the Quake 3 engine was to id Software... Over the years they'd simply "lost" (I believe that was the term they used) the source code for their Mac port... so, put simply, the mac port could no longer be supported, at all. So, that left some customers out in the cold. Guess what happened pretty soon after Quake 3 was open sourced? Somebody made a new mac port, and submitted it to id Software, allowing them to (officially) support Mac users once again. Things like this do happen when games are open sourced, programmers who believe in open source will often do rather nice things for developers who support the open source community. id Software had one guy pretty much port every single one of their games for very little, and last I heard, still does port newer games to Linux, and iirc, Mac. (He may even do it for free, but I cannot remember).

It benefits Cyan in other ways, too. Cyan will be able to use our enhancements, if they're sensible about their licensing, for future games, too, saving them cycles, and money. (Note to Cyan: Please, be sensible in your licensing, LGPL or (modern) BSD-licenses will benefit both you and the community greatly, don't squander your chance, here.)

This open sourcing does not spell doom for Cyan at all, not even close, infact, it could be massively beneficial to them, if they go about this sensibly.

EDIT: Perhaps I should also add to this some other notes.

Be aware that Uru-fans may even become a minority in the coming open source. Some people will not know about Uru, or even care about Uru. They will still be a good thing to have.

Exotic opinions will be voiced, and sometimes loudly. And again, this is a good thing.

Be prepared for an influx of new blood, not necessarily "Uru"-based new blood, but, rather Plasma-based new blood.
Keep in mind the balance of "power" (by that I mean dominant opinions of the community) in the future may be vastly different to what is the currently accepted norm. The new will perhaps outnumber the "old guard".

It isn't going to just be about the Uru community anymore. The Uru community will still exist, perhaps, but, there will likely be a greater "Plasma" community. And some of the, how do I put this, "militantly" pro-cyan-to-the-exclusion-of-all-else group, whoever you may interpret or not interpret those to be, will have to get used to this. As its been said a number of times in this thread, it isn't about us anymore.

The landscape of this community is going to change, and I hope people realise this. Its not a bad thing, actually its a good thing. A very good thing, indeed.

EDIT2:

Also, Nalates, Plasma is nothing like SL, and it never will be. It just doesn't work in the same way. You can't necessarily take what you may've learnt about creating content in SL, and make it work in Plasma, without having to mostly start again from scratch. About all that you could use from your knowledge of SL is either textures, to some degree, and modelling in a 3d program, that might've learnt playing with sculpties. And can I note? Sculpties are realy frecking weird, and if you're really great at making sculpties, you still might have some trouble making content for Plasma. I've played a little with both sides of things, and to be honest, there isn't really a one-to-one comparison of content creation for SL or similiar things, and content creation for a game engine like Plasma. They really don't work the same way.

EDIT3: fixed some sentence fragments.

EDIT4: Wow, I'm really adding to this more and more. Also regarding ODE vs Physx and Havok... ODE is probably developing and /releasing/ features far faster than Physx and Havok. You do realise that a number of open source libraries are commonly used in the commericial games? Infact, Uru last I remember used a bunch, too. By "Uru" I'm referring to Uru: Complete Chronicles. Neverwinter Nights used them, too... Lots of games use open source libraries. Now, commericial libraries tend to do a few things. One such thing is they tend to hold back on releasing new features until they have enough to warrant an entirely new release of the product, and requiring a new purchase for anyone using the older version, to get this brand new and improved version. That way, they make money, and really, that's what they're there to do, they are a business, after all. Open source libraries tend to just release features as they're done, and tested, mostly. (Depending on if its the stable or development verion of the library, generally) Bugfixes and security updates tend to come along as and when they're done. Another thing to note about the general work ethic of open source developers is the mantra "Release early, release often." its an important thing to note. I'd imagine the development of ODE has taken less time, than that of Havok, of Physx, though I could be mistaken.

ddb174 · **Joined:** Wed Jan 16, 2008 5:27 pm **Posts:** 167

I don't expect people not interested in Uru to join in Plasma development, simply because there are better open-source engines out there already. Plasma's value is really just in running Uru.

Nalates · **Posted:** Mon Dec 07, 2009 9:35 pm

Kaelisebonrai, great post. Thanks. I agree with most of your points. While we probably disagree on the degree or importance of several points for how it can help Cyan, you have it clear and well said. Your Quake example is a good point. Your point on how Cyan writes the licenses is EXCELLENT and the heart of much to come in Uru’s future.

On Plasma and SL I think we see those the same way only on some levels and disagree on others. Plasma and SL are only related at logical levels not implementation levels. I see SL as a good place for total novices to start to learn the problems of 3D modeling for a VW. Those learning game development in more of a classroom type environment (regimented education of some type) sitting through explanations of and learn LOD (level of detail) rendering problems, graphics efficiency, lag causing design and other basics is part of the process. Boring.

In SL a free form learning environment they can experience building things and see their effort at the end result stage and have a reason to now learn why they have to pay attention to LoD, poly counts, texture resolution and all the technical boring stuff. They learn to solve problems they have a reason to solve. I think a large number of Uru fans will never come into age building via the more formal education path where they learn massive information before getting to enjoy results. The formal learning path starting at the beginning to the point they have a 3D model they can walk around in is just too long for gamers. SL shortens it by a considerable margin.

Sculpties are WEIRD. But to make good ones one does have to learn Blender or some 3D modeling tool. If they are at all complex then it is Blender, Maya, 3DMax or something like that. The skills, ideas and principals from those serious tools are transferrable to Uru.

One is highly restricted when building sculpties for SL/OpenSim worlds, far more so than when making a 3D model for Uru. Texturing and UVMap use and learning to bake textures are parts of the sculpty design. Understanding polygon use and vertices is a must for workable sculpties. Those that ignore the limits soon learn what SL types mean by vertex vomit. Understanding LoD and multirez is a must for building stable shapes in SL sculpties. All of that knowledge is useful for age building in Uru.

If one learns to make a sculpty as a step-by-step recipe or ritual then you are right it is not transferable. If one understands why they are arranging vertices in specific numerical relationships, why creases are placed in certain ways with specific numbers of vertices between them… that type of knowledge is transferable.

Depending on what part of the learning process and knowledge one talks about, it is or is not transferrable to Uru.

Only on ODE and PhysX/Havok development speed am I skeptical about what you wrote. nVidia and Intel are building features into the hardware and adding support to PhysX/CUDA/OpenCL and Havok. I am much more aware of what the commercial people are doing than I am ODE. I think that fore knowledge of how hardware will support software gives PhysX/CUDA/OpenCL – Havok people a huge development edge.

Features the ODE folks were working on in 2005 don’t seem to be here yet. OpenCL was a dream in early to mid 2008. By late 2009 most major developers are adopting it and updates are coming out. Many of the hardware additions are driven by needs discovered in the use of PhysX/CUDA/OpenCL. I just can’t see the ODE development matching PhysX/CUDA/OpenCL – Havok development speed.

ODE is used in some high performance games, or at least some parts of it (Call Of Juarez, Elite Heli Squad, etc). When I was looking at deciding which engine was to power my OpenSim I looked at how ODE was being used and who was using it. Several of the games using it were impressive. The OpenSim crowd has adopted it, even if a faction is adding BulletX (PhysX oriented). But the implementation of ODE in OpenSim is NOT impressive. However SL using Havok is not impressive.

So, I’m not sure how much work it will be to make ODE work for Uru or how well it will work. Since PhysX is already implemented and free, I’m not sure there is enough justification to warrant a change. But, since that is part of the open source side of the equation there does not have to be a justification other than someone’s desire to do it. After all OpenSim allows several different physics engines to be used by simply changing a configuration file in the installation. A clean modular design would allow multiple engines to be used in Uru. If PhysX gets pulled out one might clean up the design to make future changes easier.

Questions about this so called.. open source thing.

Who is online