I remember about a year back Toady saying something about him doing this huge update because it'd be unfathomably faster than going month to month with releases.
I'm no programmer, but I suspect that the things Toady is coding are all complexly interrelated, and he has to gut the existing code to work on them; it's probably about impossible to just stop and release on a dime. Even if he wanted to do a rush job and release a buggy, suboptimal build, he probably couldn't.
I vaguely remember someone reporting that or toady himself saying somewhere on the forum that he had admitted he bit more than he wanted to, so it is more annoyingly longer release than what he actually wanted to put us through.
Since people are talking about Toady's reasons for the long wait, here's his statement on those reasons:
In general, yeah, I'm not happy with how long the release is taking, though I'm not going to be too critical of myself since it's hard to see how these things are going to work out sometimes. Save compat or not, the material rewrite had to come in at some point, and excluding the rest of the release, that's probably still six months (not even done yet), for little immediate return without additional features. Picking off some of those low-hanging fruit that popped out of the definition rewrite like descriptions could have waited, but there are downsides to a dry six month release as well. The entity rewrite was necessary for the military stuff, which is still the ostensible goal, so the several months going into that is also a chunk (you have to count time spent on entities prior to starting the material/underground stuff). Then there's the underground update, which goes nicely with the military rewrite and also the material rewrite but could also have waited. The combat rewrite was partially necessitated by the material change, since too much wouldn't make sense without it, but the specifics of the health-care rewrite might have waited (though I'm not sure what impact that would have on playability, since I'm not sure if the number of broken limbs won't increase somewhat). Anyway, yeah, there are a few more dependencies in between them, so I'm not going to acknowledge the absolute exclusivity of the groups of changes for this release, but I do acknowledge that it definitely could have come in more than one giant section, but I think more than two or three parts would have had led to some heavy downsides caused by some very, very dry longish-wait releases.
Last month was particular bad, though. Sometimes I just don't have a lot to talk about, because what I was doing was boring as hell for the most part. I have fun writing up the dev log entries like the one I wrote for today, but now that I've talked about it, I have to do it, which is still cool, but there won't be that much more to say that hasn't been said (there's not even a lot that can go spectacularly-and-amusingly wrong during testing here, just plain wrong, which isn't fun). I don't really have any ideas for how to make the log more exciting during these times, especially without seeming like I'm deflecting from the fact that we are in a boring stretch (as it would seem perhaps if it became a cat blog for those stretches). The dreariness will happen again during this release cycle after squads as well, as there are still a lot of routine changes to make.
It would be better to make it faster than more exciting, but I don't know how to do that either, now that we are fully underway.
I'm still ambivalent about save compatibility preservation as I've been handling it. It wasn't fully responsible for the decisions that led to the long release wait, but it was an encouraging factor in piling on a bit more than I needed to pile on. On the other hand, world preservation is only going to become more important to people, especially when they are allowed to get out there themselves beyond running isolated sites, and I'll probably be happy in the long run to have made so many save-breaking changes now once the main elements of the army arc are complete, rather than wrecking worlds that people have invested in. Nobody really cares now, but if people don't care later, we aren't doing this right.
In any case, we're well along now, and all I can do is try to be a little more careful with my scheduling for future releases. For instance, we mentioned a while ago that we were thinking of doing improved sieges or sending out armies plus looking at the top ten on the voting thread plus one thing of our choosing. I don't think that would take a year again if it were all done at once (as if I've ever been right), but after this process we'll probably still stick more or less with that sequence -- but definitely grouped into several releases. There would be no reason not to do all of the voting items independently for example, if I'm remembering them correctly (it does represent a few lost days, since releases themselves take several hours to prepare, especially now that I'm on three partitions over two computers).
There are reasons not to break up improved sieges that much, since a too-incremental approach there, aside from lessening the impact, would also involve way more AI rewriting each time, and improved sieges require a lot of mechanics that aren't in the game yet. My previous inclination would just be to do it all, but if I want to do more releases, I might have to backtrack and do things like the vehicle rewrite as a first goal, which is essentially wasted time until they are used, which leads to complaints and ill-will, just as many elements of this release garner complaints for their apparent lack of utility. The material component of the release never had a chance, I think, in terms of public relations and expectations (broadly speaking, I know many of you were excited about it -- and I have no real way of telling how upset people are. Donations have been okay so far.). You either spend some months working on something with very few immediate benefits and do a release that changes nothing, saying it's good for the project in the long-term, or you explore some of the benefits, which leads to further delays and some new features. It's not so clear-cut, as there are always short-term projects laying around that might be used to spruce up a dead-weight release without delaying it too much, but it is a conflict that's happening and which I have to think about when I'm planning what to do next, which is something that sucks personally, since it screws up the flow of the project, though I think it's good to keep you all happy as well.
In terms of my well-being and interest in the project, I do get irritated by those kind of considerations, but by my own self-observation, which isn't going to be perfect, I don't think any of the annoyances surrounding working on computer games are remotely close to burning me out or making me tired of trying to make a living this way.
[...]
Just to make sure we're all on the same page, I'm speaking of save compatibility being opposed to frequent releases in terms of how my maintenance of save compatibility between versions leads to a piling up of save-break features and then a large messier delay, as happened here and with the Z-axis (the Z-axis wasn't the only save-compat change saved up there).
Author
Topic: Serious Threat! (Read 2016 times)