Proposal: Feature Lock at 1.3, 1.4, or 1.5 (Whichever works)

[Maeglin Dubh]

I propose that we pinpoint all the features, settings, WML issues, attack types, and things of that nature, that we'd like to change, and have it all done for version 1.5 at the latest. At that point, feature development would freeze (to a point, not completely) to allow for increased content under a stable system. I was reading the thread on holy damage, and I was thinking of the Imperial Era and the Medieval Human faction I had hashed out, and how the clerics would handle their attacks, and how I would proceed with their development when I wasn't even sure what attacks to use. I think it would be easier for campaign and era developers when they know that mainline isn't changing for a while, and they can freely develop within the current constraints of Mainline Wesnoth.

If I haven't articulated my idea properly, I apologize, but feel free to discuss and/or attack.

Just remember. I have a lightsaber.

[Yogibear]

Well, i consider this to be a good idea, but i fear the nature of open source projects in general and wesnoth in particular makes it real difficult to go this way. The problem is actually resources: Proposing a plan at best works for short term goals (like weeks, up to a few months maybe). Even then there is always the possibility of someone becoming unavailable all of a sudden. If that person is vital for implementing a certain feature, the plan would have to be modified. And again you have to deal with uncertainty.

Take the OpenGL graphics engine as an example. It is a technical feature rather than important for content but it demonstrates the point pretty good. This idea has been around for more than a year now. But there is very few devs that can implement it. Ayin once started but then left the project. It is a lot of work to do and you can't say when it will be done. Actually, you can't even say if it ever will be done.

So for me a content maker has two alternatives: Either stick to the current stable release or adapt to changes in the development release all the time.
However, i think the risk of having to rework lots of what you did, becomes drastically smaller with every new stable release. Most of the time, features are added. Modifications are there as well but after 1.2 i don't expect to see many major changes in mainline.

What we can do, to further minimize risks, is to publish a feature list of current ideas, that we intent to implement. I don't think we can fix it like you proposed but it should at least give you an imagination what is about to come.

[Jetrel]

Sweet idea, but grossly disconnected from reality.

If behemoths like microsoft (and adobe, and apple, and all software developers out there) have to axe planned features because they're just not getting done, we're probably not going to be able to make any guarantees either.

These companies promise these features, and then break those promises. The least we can do is have the presence of mind to not promise what we might not be able to deliver.

Believe me, I feel your pain.


What we could do, is each make reasonable statements about what we're going to try to do, and only make such statements if we have some clarity on the issue.

[qk]

I think that say maybe top ten feature request is good, sometime is for me as programmer hard know what feature is important for gameplay and some isn`t much important (i am not good game designer).
But say that to version 1.5 this x features must be implemented is hard, because some features can be very hard to implement for someone who doesn`t know it good. Say this x features is important and is good if implemented to f.e 1.3 sound for me like good idea.

[torangan]

Just a question - how does your proposal differ from keeping 1.0.x and 1.2.x servers up for indefinite time? Those (will) have a stable stable set of features and you can develop within their constraints, expecting your campaign to work in later versions as well.
Of course, the latest development release might contain new exiting stuff you'd like to use or bugs aren't getting fixed anymore. The first problem will never disappear, there'll always be a new feature which would be usefull to have. The second is no real issue either - if you're interested enough become the new maintainer of this particular stable line and create a new version. We'd just have to make that official then. It's the same as with GCC or the Linux kernel. They've got only one stable series in official active development. But older series are maintained by different people for some time and get non-intrusive but important bug fixes.
I see no reason why we shouldn't allow someone new to become the new maintainer of 1.0.x if there'd be interest. Maybe we'd need some work to allow easy forwarding of requests to physically different servers if we don't want the load of old versions on the mainline servers but that's all. If need be we discontinue support for the older stable versions but the next version has different servers hardcoded and still works with all old content. This works even for people who can't get a new version from the net. They should have no need to access the servers either and if they get campaigns in any other way, they'll work with their old version still.

[Maeglin Dubh]

I was just putting up the idea to correct a problem I saw both currently, and possibly looming on the horizon with even greater severity. Alternate possibilities for correction of this issue are completely acceptable.

[Eleazar]

Really a "Feature Lock" isn't neccesary to address the "looming issue." The great bulk of work being done on Wesnoth does not invalidate existing campaigns. Adding more terrain, unit animations, new factions, better UI, OpenGL, more features to WML, more setting options, more efficient processing, etc. — only makes existing campaigns better.

Perhaps what you really should be asking is that after a certain point changes should not be made (or only with much deliberation) that would break or unballance current campaigns. It's the same concept as a "string freeze" — only freeze the minimum neccesary.

[Boucman]

NO. WAY.

I mean, seriously, this idea forgets something really important

we, coders, do this for fun. The day we can't add stuff to wesnoth, most of us will simply leave.

feature freeze are good, but they have to be temporary, or we'll have all the coders run away

[Maeglin Dubh]

Ok. If it won't work, it won't work. We definitely don't want to lose our coders.

As I think about most of our current campaign and Era developers, most of them I think are commited enough to roll with the punches, as it were, and keep up with development.