[STICKY] The Team OSCAR General Discussion Thread

An Open Content Development Team For Myst Online: Uru Live Again And Compatible Fan Shards

Moderator: OSCAR Managers

Nye_Sigismund
Member
Posts: 64
Joined: Wed Sep 29, 2010 12:59 pm

[STICKY] The Team OSCAR General Discussion Thread

Post by Nye_Sigismund »

Hey folks,

About two weeks back I floated the idea of forming a project development team for creating content for MOULa and compatible fan shards. The infrastructure was the biggest concern to me then, as it was obvious to me that that would be the tricky part; managing a large number of assets amongst a distributed team, with those assets being of large file size.

Since then, I’ve been pondering over the viability of the idea. There have been successful and productive teams in the past – the DRA being a recent example – but there also have been a lot of failures, and in general the idea of teamwork is the antithesis to the individual content projects that have been the norm for development in the last few years. However, I’ve never thought that that way was a good thing for the community – it’s slow, unproductive, requires duplication of work and usually leads to half-finished projects. As that’s not good for the community, I want to experiment in the avenue of teamwork, to see what the community can really achieve if everyone who is willing to contribute puts their back into it, as part of a team that can draw parallels to the methodology of open source programming teams.

The Proposal

I propose to organise the creation of a content development team for Myst Online: Uru Live Again and compatible fan shards. The suggested name for this team is Team OSCAR – OSCAR standing for Open Source Creation and Restoration. OSCAR would have the following mission statement:

“To create the best possible content for Myst Online: Uru Live Again and compatible fan shards, in an open way that facilitates cooperation and conversation within the community.”


This mission statement holds a few crucial concepts that define what OSCAR is. Most importantly, it defines the team as a team focussed on MOULa and compatible fan shards, as opposed to e.g. PotS development, or branched versions of the game which modify the core engine to such an extent as to break compatibility. It also specifies that the methodology OSCAR uses should always be open to everyone in the community to get on board with – the community being everyone, not restricted to cliques, political groupings, etc.

OSCAR would work on the foundation that an open development process is far easier to stay interested in and be attracted to from an outside perspective, both from Uru developers and – in the future – external developers who find their way to Uru development. By an open development process, I mean one that is founded on the methodology of open source development – access to the assets with a license to use them how you wish, a development structure that facilitates everyone getting involved, etc.

OSCAR, as a content development team, stands separate from the architecture of either OpenUru or the Guild of Writers. As distributed content development has not been a priority for the modding community, this is hardly surprising. I propose, however, that OSCAR will take an active role in defining what needs to be done in the area of distributed content development, both in the methodology and the supporting structure that will be needed to facilitate future teams.

To summarise – OSCAR learns from the past, takes what we have in the present and will work towards a future where what we currently have is far better.

Does anyone have anything to contribute to this? I’m putting this here as Veralun, MOUL forum moderator, pointed out “[The MOULa Open Source Discussion sub forum is not] the right place to decide or discuss which team or what system is the best.” This may be more suited to the GoW forum, and to be honest I'd be happy to talk this over with the GoW too. But I want to discuss this on neutral ground - here - before I waltz in to their forum and say "Hey guys, let's try this different way of doing things!"
Last edited by Nye_Sigismund on Wed May 25, 2011 4:11 pm, edited 2 times in total.
Reason: Discussion does not have three "i"'s, Huw!
Huw Dawson
Team Member
Team OSCAR
User avatar
Mac_Fife
Member
Posts: 1239
Joined: Fri Dec 19, 2008 12:38 am
Location: Scotland
Contact:

Re: Team Creation Proposal: Team OSCAR

Post by Mac_Fife »

It may be worth bringing this two year old thread to the attention of readers: viewtopic.php?f=12&t=142
Once you get into that a bit, you'll see some pertinent discussion on this subject has already taken place, both here and at the GoW.

Personally, I always imagined that a single individual will never be as adept in all the skills required to build a "good age" (however you measure that) as a collaboration would be. However, there are those who feel that that they are capable (and I won't argue with that) and don't like the idea that their work may be held up waiting on someone else, say, completing a set of textures that they'd promised.
Mac_Fife
OpenUru.org wiki wrangler
User avatar
Szark
Member
Posts: 44
Joined: Sat Dec 13, 2008 1:41 am
Location: UK

Re: Team Creation Proposal: Team OSCAR

Post by Szark »

I like the philosophy of this Nye. As a wanabee age maker wanting to make ages, as a team, has be difficult at best. Theremin and I have for sometime had plans for two linked ages to start with. So much we got a wiki set up, back-story in progress, concepts going, invited others to contribute. All was well for a few months until we realised we would need coders and more experienced modellers. This idea was born at the same time as DRA, in fact a number of them could have worked on this project if the DRA idea hadn't come first. I am not bitter, saddened as I really think the concept has merit for MOULaOS, but no not bitter. I know a lot, not all, but enough of the DRA project to have learnt from it, the good and the bad. And because we couldn't get coders and modellers to even take a look the project, it is now gathering dust.

The thing is we always said that the project wasn't ours, yes we made a start but we were totally open to a rewrite or to scrap it all together. We wanted anyone who took part to take ownership of the project. But in the same token to relinquish their rights to an idea used in the project. Once an idea goes in it becomes everyone’s. Also I leant that these projects need some sort of management, DUCK Yep someone to oversee, to keep to flow of information going back and forth between coding, modelling , texturing and sound side of things.

Another thing that stood in the way of progress was the time for the OS news to come and having to save the age as Max 7. I will be one of the Blender mob. LOL

So yes I like this idea very much. I personally get more from working as a team than on my own. I had more fun when the GoMe paper was made by a group than at the end when it was just an handful, with me doing the majority at times. Burnout time. If I had to make an age all by myself I wouldn't, not even with the best plug-ins. Why because IMHO you can't beat creative group collaboration, just like Cyan did/do. Bringing all the talents together to make ages worthy of MOULa is a good start.

Saying all this I have nothing against solo age makers, if you can do it all and you are happy then I commend you but for me it is team all the way.

So where do I sign up?
Image
Nye_Sigismund
Member
Posts: 64
Joined: Wed Sep 29, 2010 12:59 pm

Re: Team Creation Proposal: Team OSCAR

Post by Nye_Sigismund »

@Mac - Obviously, a huge amount of discussion has already been done. That topic had some good discussion which I'll definitely read through. As it was made clear in that topic - a good team needs a solid leadership, and even though there's been lots of thought about it, nobody's really tried in public, beyond a few non-starters and SR. I'm not a fan of some top-down leadership style - but what I'm looking at with OSCAR is setting up a team that has a common vision and goals.

One other thing I'll add is that as it stands, I can quite happily trundle along with content development by myself. The trick to OSCAR is launching it, then doing work myself. In a race, you don't wait for everyone else to react to the starting pistol. :P Work on Llantern has provided me with a basic skillset, so work can begin as soon as I'm happy with starting OSCAR.

@Szark, I've not properly got things sorted for OSCAR yet - I've got university work to do. Over the next few days I'll be putting up some draft documents on how I think development could happen, and how contributions can be made. :P

On the Max - Blender issue hinted at in your post, one of the advantages to team projects is that, if done properly, only the people rigging the age to work in Plasma will need Max. Everyone else will be able to contribute from their native environment. I do intend to personally be developing in Max, and until PyPRP2 gets to a 100% MOULa compatible situation I think that's probably the best platform.

EDIT: Actually, an important point just occured to me -

On sharing assets, there are three ways I can see - Dropbox, filesharing websites, and a proper repo with version control. Dropbox is really tempting, especially as it has version control, but I'm converned that the 2GB free limit may be an issue. Does the OU server, internet etc have the capacity to withstand an asset repository and version control system?
Huw Dawson
Team Member
Team OSCAR
User avatar
Mac_Fife
Member
Posts: 1239
Joined: Fri Dec 19, 2008 12:38 am
Location: Scotland
Contact:

Re: Team Creation Proposal: Team OSCAR

Post by Mac_Fife »

Nye_Sigismund wrote:Does the OU server, internet etc have the capacity to withstand an asset repository and version control system?
At the risk of speaking out of turn on behalf of rarified, the Foundry server isn't on a hosting service; it's part of rarified's business platform that he's graciously "donated" to OpenUru. So while he's no doubt got some budget on disk space, I don't think it's anything like as limiting as many hosting packages are. Subversion and Mercurial repos exist there now.
Mac_Fife
OpenUru.org wiki wrangler
User avatar
rarified
Member
Posts: 1061
Joined: Tue Dec 16, 2008 10:48 pm
Location: Colorado, US

Re: Team Creation Proposal: Team OSCAR

Post by rarified »

Mac_Fife has described the Foundry fairly accurately. It is a server I run from home as part of my business interests, as well as providing OpenUru support. Thus far OpenUru activities have not had any significant impact on it.

It is fairly well equipped, the disk array that the Foundry uses still has 1.5Tb free, so space currently is not a problem, and I'm happy to host any Uru/MOUL related materials in the repositories (within reason!) I also have a commercial ISP provider, although the upstream speed for the link is currently capped at 5Mb/s, so it's not going to provide bulk transfer for a large number of simultaneous users. There (currently) is no bandwidth usage cap. There are traffic shaping rules in place to give me priority, but as far as I can tell there have been no occasions where they've been used.

I'd suggest you start planning early on for a description of your workflow, and try to identify where and what external resources you think you need or would like. We'll see what can be done to provision them in a shared environment for you.

_R
One of the OpenUru toolsmiths... a bookbinder.
User avatar
Szark
Member
Posts: 44
Joined: Sat Dec 13, 2008 1:41 am
Location: UK

Re: Team Creation Proposal: Team OSCAR

Post by Szark »

Nye_Sigismund wrote:@Szark, I've not properly got things sorted for OSCAR yet - I've got university work to do. Over the next few days I'll be putting up some draft documents on how I think development could happen, and how contributions can be made. :P
LOL Nye it was a figure of speech, no rush. I just thought I would show my support. :D

A good point on resources and bandwidth there Mac and rarified. I would imagine a model repositry would get large fairly quickly. Plus textures possibly, scripts don't take much room.
Image
Nye_Sigismund
Member
Posts: 64
Joined: Wed Sep 29, 2010 12:59 pm

Re: Team Creation Proposal: Team OSCAR

Post by Nye_Sigismund »

I wrote this a few days back - it's basically some proposals on how OSCAR would do things.

Organisation and How Things Are Done

Organisation - Communication

General communication should be done in a forum or message board environment. This is arguably the most important area – all information needs to be placed here promptly. This is because OSCAR is founded on the belief that open communication with the entire community allows for discussion and gets people interested in helping. It's been seen in the past that teams that do not tell people what they want to or are working on have trouble attracting interest.

As forums are a somewhat cumbersome way of discussing things and generally chatting, OSCAR should maintain an IRC room. As with the forum, it's an open area which anyone can choose to hang around in to catch up on the latest information.

Scheduled meetings of interested members of OSCAR take place when there is a need for those meetings to take place, dependent on the project in question. A fan age will almost certainly have a meeting to discuss Stage 1 and both reviews (stages 3 and 5) are glorified project meetings.

Meetings can be conducted via Skype, with additional screen-sharing and remote access being used when needed. Screen-sharing has huge benefits, especially in concept, because it allows for everyone in the meeting to see exactly what is going on. Skype works as a platform as it's dynamic. It can scale from two team members chatting about the weather to an all-team conference call, and also has a handy chat function. Meetings can be organised by anyone. Ideally, the meeting is recorded and made publicly available, and the outcome of the meeting must be published.

It's hard to underscore how critical open communication is. It, tied into the light-hearted and hopefully politics-free nature of OSCAR, is the entire reason that OSCAR was set up in the first place as opposed to continuing on with individual fan development.

Organisation – Roles

There are three roles in OSCAR:

Member: Everyone who wants to be a member.
Contributive Member: Members who have contributed work to OSCAR. This is simply to monitor active developers.
Team Lead: An organiser. Team Leaders don't have special priviliges in projects themselves, but they're the people who organise the stuff that keeps OSCAR running, and settle disputes. The people behind the curtains, but without the curtain. ;)

Organisation – Projects

Projects are divided into five stages.

Stage 1: Creation
A project is defined. If it's an Age, concept art is made. If it's something else – for example, more dance emotions - the project is outlined. A project must go through at least one meeting to be “green-lit” for Stage 2.

Stage 2: Preliminary Work
The project begins to be worked upon. Foundation work is done. For an Age, this is the “blocking-out, figuring out” stage, where the details decided in Stage 1 are brought to simplistic life and changed if issues arise. Minimal coding and wiring is done at this stage. For other content, like the dance emotions example outlined above, this age is context dependent, but generally speaking this is when the foundations for the work that must be done in Stage 4 are made.

Stage 3: Preliminary Review
Another meeting is scheduled for when the people involved in Stage 2 believe that their work is done. This will review the progress of Stage 2 work, especially in matters arising. The blocked-out age is reviewed, and the dance emotions are analysed. Ages can be sent back to Stage 2 if the foundations are not fully solid, or the project may be scrapped entirely. However, it's far more likely that projects will advance to Stage 4.

Stage 4: Work
This is where the serious stuff happens. As all preliminary work has been done, this resource-intensive section of development is smoothed as much as possible, as things like design revisions will be far less frequent. The animations are made, the music is recorded, the puzzles are wired, the lighting is fitted, and so on.

Stage 5: Final Review
This is the serious review, which determines if a project is ready to be considered “Gold”. Gold content is complete to a high standard, with no bugs of major status or higher, and few minor bugs. Gold content is QA tested, probably by the GoMa, who analyse it, bug test it in various situations, and give it the QA testing that can give shard owners the confidence to use the content.

How Things are Done

All projects are public. Depending on context, the most efficient way of producing that content is decided on. Team OSCAR will utilise a wide variety of tools for this purpose.

This is an area that will need to be ironed-out by more experienced contributers, but a tentative outline of how things are actually produced is below:

AGES:

Age development has three different major roles.

Prop modeller: A 3d artist who makes individual models. For example, a column, a piece of rubble, a book, a linking stand, a statue. These are what can be considered “props”. Props, ideally, will be made with future versatility in mind. For example, the best props will use a variety of texture maps, even if the project they were originally developed for does not use them all.
Environment modeller: A 3d artist who develops the area that props populate. These team members operate in tandem with prop modellers, and often functioning as prop modellers as well.
Programmers: These guys write the code whilst the age is being built by the modellers, and do the complex wiring of the age once the age is model-complete.

In addition to these three roles, you also have the musicians/sfx engineers, writers, etc. These are smaller roles, but still important to making the age complete.

Props and environments can be made in any 3d modelling program – Max, Maya, Blender, whatever.

Wiring the age, for the time being, should be done with the PlasmaMax plugin. Ideally, this should be done in a Max version compatible with Cyan's build PC. However, wiring for development can be done in any Max version, but the understanding is that the age will be ported to Cyan's Max edition before going Gold.

Ages should have a project forum thread, a project on something like JIRA for management and a repo for people to access. The repo should be committed to by everyone – however, the people involved with getting the age together – firstly the environment artists, then the programmers, should be the ones editing the .max file the age itself is contained in.

Props should be distributed in formats that the environment artists can easily utilise Max to import and place in the environment – eg, .OBJ and .FBX files. Ideally, prop creators should be committing to the repo into something like a “prop” folder, which stores all the meshes and textures.

Depending on the project, if might be that several people intend to work on the environment at the same time. Here, allocating everyone a certain area, represented by an individual page file, will be very handy. Communication amongst environment modellers, as they're responsible ultimately for the age as a whole whilst it is being modelled, should make every effort to communicate often with other environment modellers.
Huw Dawson
Team Member
Team OSCAR
Nye_Sigismund
Member
Posts: 64
Joined: Wed Sep 29, 2010 12:59 pm

Re: Team Creation Proposal: Team OSCAR

Post by Nye_Sigismund »

One exam down... This thread is kind of word heavy, but...

Image

All that is needed for the time being is a subforum. I'll send a message the admin's way now.
Huw Dawson
Team Member
Team OSCAR
admin
Site Admin
Posts: 194
Joined: Thu Nov 13, 2008 6:33 pm

Re: Team Creation Proposal: Team OSCAR

Post by admin »

This proposal has been promoted to project status and given a forum. This original topic has been moved from the Open Discussion forum.
Post Reply

Return to “OSCAR - Open Source Creation and Restoration”