20100721|StepMania project reboot: opinions wanted

[chrisdanford]

StepMania has been stagnating. The code base is unwieldy and theming has become so complicated that it's turned off many potential themers.

My ideas for getting back on track:
- Fork StepMania from the build used in ITG PC/Mac (mid 2006). This is probably the best-tested build in StepMania's history.
- Port the 3.9a theme (SMMAX re-skinned with some SM 4.0 graphics) to this build. We'll use the same base that many 3.9 themes use so that they're easier to port to this new build.
- Remove esoteric features and theme elements. This will reduce bloat and make things more themer-friendly.
- Add support to StepMania Share for uploading NoteSkins and Themes. This will require changes to the theme and NoteSkin directory structure so that different users' uploads don't collide.

Quick stab at a build - Windows StepMania CVS 2010-07-19
|
icon=stepmania

[Xyton]

Well as far as I can tell, there are still a lot of people that prefer to run off the 3.95CVS build that (I believe) is the build used by ITG, so if nothing else i should button up that schism nicely. ;)

I've never really looked at StepMania from the code side of things, but the only feature in 4.0 that I would be sad to loose would be the improved course editing / information from within the song editor (i.e. pick a song, hit C, and pick a course) -- and even that seems to be in the 3.95 build (3.9 alpha 23) in a rudimentary form. Also, native Intel support for Mac is nice, but I'd imagine that would happen in the new builds, too.

-EDIT: Time signature support was pretty slick, too, if only for the music nerd in me. :)


So yeah. I doubt I could help with the coding, but I'm totally here for moral support! ;D

[chrisdanford]

http://www.stepmania.com/forums/showthread.php?t=16173 seems to say that it was forked August 2005, which is about a year before ITG PC/Mac. I'll ask Vyhd because OpenITG is aparently based on 3.95.

[tweak]

Although I can totally see where this is coming from, I feel backwards is really not the way to go, especially with OpenITG being a nice stable build that appended some nice features and stability.

My thoughts on this as a user/basic themer:

Many issues seem to stem with people just losing faith it seems. With the amount of time between some releases, large changes being made to the theme engine, some strange noteskin choices, and a rather heavy default theme. Some forks like SM-SSC address many of these issues.

As to making it more "themer friendly", one of the best things you could do is to provide documentation to get people started. Keeping things simple in the default theme is a great idea, but why remove/simplify things for everyone? As people like shake, AJ, Midi, Deluxe, and your own team with the Pro series is that the current build and forks can do a lot with great results. But, at the end of the day, i find my reliance in these few people to be rather overwhelming, as besides them and articles they have created, there is more or less no way to know what i'm doing. Documentation and tutorials straight from the horses mouth would help get things moving.

As I recall, from joining the community aggregate around the late 3.9 alphas, many prominent themers dropped off/moved on before final was released. Only a couple notables such as AJ and k// and some recent efforts really continued the push after 3.9 final.

And as for pulling new themers in, interest is the only thing seriously bogging it down as a whole is lack of interest. As the ITG community has showed us time and time again, many are willing to at least try to theme, and the xml based "3.95" cvs/OITG isn't exactly a far cry from the LUA system currently used. Many basic things can be done similar just by changing the syntax.

As I previously stated, going backwards probably isn't the best route. Although I'm sure some people would love SMMAX brought up to date, I believe it would hardly push many make the jump to SM4. I feel many would continue the path they are on, i.e. they use what they like, and what works for them. Since Pro2 is supposedly finalized now would probably be the best time to work on reinvigorating the community. Embrace your pad players, your keyboard players, get StepMix finished, and give the community a real reason to want to use SM4.

[john_reactor]

I was trying to learn themeing since early 3.9 alfas. It was hard. Mostly because there was no documentation. So then 4.0 came up with its XML themeing system, all my hard earned knowledge became obsolete. You can imagine that I wasn`t happy because of it.. Lack of documentation, low performance and slow development of 4.0 didn`t help to change of my attitude.
My money is on SM-SSC now. It`s stable, it`s fast, it`s documented. I only don`t like it`s default theme, but i can live with that.
So, in my opinion, if you want to revive 4.0 you must go the way that ssc team took.
Or if it is "politically" possible dump 4.0 and make SM-SSC official version.
Don`t hate me ;)

[eighty5cacao]

Not being an "expert," I am generally neutral as to what happens to the official StepMania, provided that sm-ssc continues to be maintained for people who need the SVN features.

If the StepMania project is rebooted from an old version, then it should probably be renamed slightly (StepMania Reloaded?), and the version number should start over at 1.0 or other number deemed suitable.

[akstylish]

So are you suggesting that we abandon the current 4.0 alpha and restart?

[chrisdanford]

I'm pretty sure that the amount of effort required to achieve a quality release from the 4.0 trunk is more than I'll ever have time for. I'm just looking to get something official, stable, fast, compabile, and well tested out there so that there's no reason to stick to something as ancient as 3.9.

If there are specific things that themers can't live without (support for "default.lua" instead of just default.xml), I'm interested in trying to make that work.

[tweak]

If you really are intent on going that far backwards, there is little reason to move from sm-ssc and oitg.

[chrisdanford]

Starting from the OpenITG code would be great, but the legal state of the theme makes that part of it unusable.

Maybe this is more trouble than it's worth and things should continue on as-is. People will have 3.9 if they want something stable, OITG if they don't mind something that has to live "underground", and sm-ssc if they want something bleeding-edge but a little more complicated.