Barriers to Fan Development of MO:RE

[Robert The Rebuilder]

A theoretical project for fans to develop the MORE code base has several obstacles to overcome, which I've listed below. All of these are pure speculation on my part and do not come from Cyan.


1) Finding people to work on the code.

This isn't as hard as I thought - see the other thread for the list of people who've "signed up" so far.


2) Getting Cyan to release the source code.

Beyond being the intellectual property of Cyan, the source code is the money maker for them as well. If they hand it over to select members of the fan community, anyone can change it and make their own game, theoretically. Having a non-disclosure agreement in place would somewhat mitigate the risk by leveraging penalties. But once the code is out there in the public, they cannot take it back.

[Note: going open source would effectively end Cyan's involvement in MORE's development, and with URU in general.]

One solution might be to make sure no fan developer has all the code. Distribution would be limited to just a handful of fan developers - maybe one developer per portion. It can be broken into discrete sets, according to purpose, e.g. the server and the client code. Both of them would use the Plasma engine code, but this itself does not need to be released to the community (as badly as we want it). It can be distributed as just libraries and headers to allow the server/client code to be compiled.

But this brings up another issue...


3) Packaging Plasma.

If it isn't in a standalone form already - where it is self contained and located separate from the applications - then there's major work to be done by Cyan before this can move forward.

There's some evidence that it already exists in this form, considering all the titles they have that uses versions of this engine: realMyst, Uru, Channelwood, Myst V, MOUL, Cosmic Osmo.

There's also evidence to the contrary. Cyan hasn't released the source code to their 3D Studio MAX plug-in because of code they want to protect - which can only be Plasma code. If it already existed in a standalone form, then there theoretically shouldn't be any work to be done on the plug-in code.

Let's give them the benefit of the doubt. Let's say they can distribute Plasma as headers and libraries to the select fan developers so that they can compile the client and server code. This reveals another problem...


4) Protecting Plasma.

The fan developer has been given Cyan's powerful game engine in a form that can be easily incorporated into their own title. Other than the NDA, what is to stop them from producing such a game?

They could introduce a hardware solution where, at run time, Plasma checks for the presence of some "dongle" before continuing execution. This would only be in the development version of MO:RE; the widely distributed release would not include this. But this again requires Cyan's time and resources to implement this, both of which they're lacking at the moment. They would just have to trust us in the meantime.


5) Access to Cyan's servers.

To allow the fan community to test MO:RE, Cyan would need to make their servers accessible: to install the latest version of the server code, to query the databases, to read the debug logs, etc.

This isn't that hard a problem. Secure remote-access solutions exist, e.g. SSH, VPN. Again, this would be up to Cyan's willingness to put this in place.


---

These are just a few issues. Can anyone think of others? Or do you want to discuss some of these in more detail?

[Zardoz]

Sorry to drop this one in from left field, but there are legal issues involved in using "free"labor for a commercial product. Under the Fair Labor Standards Act, employers must pay wages to "employees," and simply offering to work for free does not absolve the employer of this legal obligation. There are ways to arrange volunteer or unpaid intern positions with companies, but they must satisfy strict conditions under that law. I raise this not to squash this idea but to signal the need to consider such an issue before too much effort is expended.

[Robert The Rebuilder]

That's a good point, Zardoz. I wonder what they did in Tweek's case?

[BAD]

Moved to Off Topic.

[Aloys]

Just one thing I want to quickly address:
These days people keep mentionning how Plasma is precious, unique, valuable, ahead of its time etc..
Plasma is essentially a 5 years old engine. Which in software history means that the dinosaurs where still rulling the earth when they (Cyan/Headspin) started making it.. And as much as I love Uru, Plasma is in many regards 'outdated'. It's not a bad engine, but in these days where Open Source projects (OSS) are littering the software landscape you can absolutely set up an engine that would be very similar to Uru in terms of features for free. It wouldn't be easy, or quick, but these days you can get something able to run a game very similar and even better than Uru in some regards only with open source solutions.
Just having a quick look at the most obvious/important features:

• The renderer itself is obviously outdated, nothing wrong about that, just the natural evolution of software. (Just seeing the loops we have to go through to have the wavesets running give me the creeps. ) A number of free OSS renderers are much more feature-packed than Plasma.
• The sound engine. That's not my speciality but usually that's not the most complex part of the game engine.
• The network aspect is a little more complicated: on one hand it works, on the other hand it could get some serious performance and stability enhancements. We'll never get 200 people in Ae'Gura with this one.
• Avatar/environment Interractivity works pretty well, although it's nothing too complicated. And, heck, redoing the whole camera system and moving from a pan/scan system to an 'orbit around avvie head' system would be worth the work.. (although that's just my totally biased opinion here )
• The physics solution despite its problems works decently too and has some nice features, but again that's nothing that can't be done through one of the several available OSS solutions. (Cyan even did it for Myst5)

Obviously I'm just doing a list of the most 'visible' features here, and I don't know everyhing about Plasma (argh, I still have to read Adam's "plasma 101") there's much more than that involved in a game engine. (not the least of which being to tie all the aforementioned features together in a stable way and debug it) Plasma has some touches here and there that make it unique, but other than that the actual value of Plasma for a random developper is very little.

At this point the only advantage of Plasma for Cyan is that it is a ready-to-use solution for them (and us, to a lesser extend). They have/had a number of people with plenty of experience and know-how regarding its use, and all their software development environment is tailored for it. And that's obviously worth a lot of money. Moving to any new engine would involve a lot of time for them. Which obviously means a lot of money. It's just cheaper and safer to keep Plasma and update it. Finally; and possibly most important of all, there's a strong sentimental attachement to it. They've spent a lot of years developping it and then using it etc.. And that's though to just throw it away.

The fan developer has been given Cyan's powerful game engine in a form that can be easily incorporated into their own title. Other than the NDA, what is to stop them from producing such a game?

Motivation. Or to be more precise: motivation in the long run. That's what all voluntary based projects are about: having the dedication and persistence of turning a simple 'project' into a completed product. How many projects are started vs how many actually see the light of day? It's impossible to get actual numbers but it's safe to say that the ratio is probably around say.. 5%.
Now if we are talking about a fan team producing a game of the size of Uru.. (I'm not even talking about the quality) Well, I'll believe it when I see it. (Some might mention SftsoK: I'd love to see it completed sometime next year; but again I'll believe it when I see it). Having an engine and all the tools is one thing, being able to produce a full game with it is a totally different thing. That's why even with a bunch of powerfull OSS engines available you see only very few large games ever completed.

[dtierce]

It might be possible to minimize the legal impact of volunteer "employment". The individuals could be listed as part-time minimum wage hourly workers who are responsible for reporting their own hours worked via "telecommuting". If such an individual chose to work 100 hours and only reported 1 hour worked, that would be their business.

The only drawback is the vulnerability of Cyan if some such worker decided to report a larger number of hours. Cyan would be legally obligated to pay according to the hours reported.

Would Cyan be obligated to provide a benefit package of some kind in this situation? I hope not. That would be a deal-breaker for them.

[D'Lanor]

So maybe Cyan has broken the law in the past? Hmm, somehow I doubt it. Maybe we should leave all that legal crap to the lawyers.

[Zardoz]

I may not be a lawyer, but I am married to one . . .

And all I'm suggesting is that people be aware of the issue, not that it gets resolved here - certainly not the place for that.

[andylegate]

Point 1: This law is a federal law for those residing in the United States of America, either as citizens or those visitors with working visas. This law does NOT apply to those residing in other countries but instead have to follow those other countries labor laws.
Sub Point A: Many people here do not reside in the US of A: Trylon, D'Lanor, just to name a couple.

Point 2: This law was originally enacted in 1938. A time when a computer was made from synchros and servos and the idea of a internet was not even, well an idea. The law has since been amended many times, mostly to change the minimum wage in the USA.

Point 3: The law clearly reads:

The FLSA applies to "any individual employed by an employer" but not to independent contractors or volunteers because they are not considered "employees" under the FLSA.[8]

Quote from the Wikipedia. Here's a link incase anyone would like to look into the law:

Fair Labor and Standards Act

I asked my ex-wife about this out of curiosity, as she works with many different volunteer orgs her self. She just laughed and said "That's what a volunteer contract is for. It shows the agreement that the volunteer is just that: A volunteer, not an employee, with rights to wages, benifits, etc. The only way the contract fails is if it can be proven that the person agreeing to the contract was forced to sign it in some way, unable to read, or can prove that they were mis-led in what they were signing."

[J'Kla]

As ever a US centric view of the world. Note I am not saying it's a bad law, just pointing out an assumption that for a lot of US citizens the view is one where what is true for them is true for everyone.

Oh by the way add me to your list of overseas contributors.