Hotfixes and stability vs cutting-edge functionality

Currently there is stable version and Experimental. Looks like a lot of people prefer Experimental for new features and (surprise!) bug fixing. Frankly I don’t even sure if anyone reports bugs for stable version.

At the time Calcysm DDA develoved for almost 7 years. It become relatively popular among opensource projects. That is very relatively, yes, but still. It attracted enough contributors to have manpower to at least try diffierent strategies in development.

Despite for contirbutors it may be more attractive to implement new feautres, some (and a lot of) players prefer stability over new features. Also modding for unstable version is complicated and unstable. There is no stable platform for modding.

Maybe it time to think of different strategies for new versions?

Of course there is no man power for LTS releases like in Ubuntu, but a at certain exent some contributors might be intrested in supporting stable version.

Other opensource projecs do fixes in stable versions.
Dungeon Crawl Stone Soup, for example:

1 Like

I haven’t really heard of any dev wanting to work on anything but experimental, so that’ll probably not happen. Just play the latest stable.

Basically any tagged release (including stable releases and daily pre-releases) is a stable platform for modding - do not backport any new features there, so modders can target any specific version and do not be afraid that something will break for that version in the future.

Despite the number of contributors, we do not have any resources for enterprise level QA, still we are doing just fine with what we have - there are talented people who contribute to CI and tests and also hundreds of players reporting bugs they have found.

People are also preferring to play experimental versions, so they rarely report bugs for stable versions. Also any bugs existing in stable mean that people failed to properly test stable release candidates (we actually do not expect them to find everything). There is some sense in backporting bugfixes for game breaking things (like crashes) to most recent stable, but it should be something really important.

Point is - in most cases usual player should not prefer unstable version over stable. It is just how it works usually.
I think if it happesns- then something is not right.

I am talking about enterprise quality. But it is more the question of scale. Maybe game eventually will come to wider auditory. In this case release strategy should be re-reviewed.

It is in the case of wider auditory is a goal.

It kinda isn’t.

I think it can work in reverse.

Players gets your game if it is good for them. That includes not only number of features but a lot of overall things, like accessibility and polishing. And release strategy have relation to all of that.

You can claim big AAA games for having “no soul” and becoming slot machines and all of that but they nail accessibility part professionally (in most cases).
Marketing will not cover lack of this features (again in most cases).

Good non AAA example of that is RimWorld that is made in inspiration of Dwarf Fortress, but actually has modern UI with mouse support. And I can’t say that it is ruined in some way.

Do we have mods in active development that is aimed for stable version exactly?

I really have no idea - I do not use out-of-repo mods right now.

Something could be wrong, but it is not up to devs to prevent people from using whatever version they want.

Wait. There is stable version that is preparing for months. There is a lot manhours put into stabilizing. And most players ingoring this work it and prefer unstable.


definitely yes.

I mean you are putting effort into it.

I really recommend to investigate this case.

Investigate what exactly? People can play whatever they want.

Reasons why exactly unstable still preferred after development of stable versions started to be at least regular.

Yes, but for whom are you making stable then? Why making stable then?

I mean, maybe it is better to at least ask players something about it?

Well, maybe it just habit becuse of delayting of 0.D.

Looks like it is the case for Stone Soup also. Bit still. It is strange.

Overall it is more like question about do you really want move development on next stage. Or are going to stay at current stage till the end of the project.

Because people want to play with the new stuff? It’s not a big mystery.


If it makes you feel better, consider any stable release a glorified translation milestone.


I just looking for bigger games as example. It is definitely not the case for even indie games. There is definitely not all people move to unstable versions there.

Looks like it not the problem. It is more about development strategy and postioning of the game. And it is quirks of mods and small projects.

My position is that existence of regular stable version is step forward from small project to bigger and better one.
if you are going to stay on small scope forever- then obviously there is no point to discuss it further.

You are making some weird accusations - we are already doing regular stable releases.

Weird? Maybe. Point is if you are making stable - you making it to be played. If most of player base ignoring it - then you are somewhat wasting time.

But looks like our discussion is going in circles. Let’s probably wait another opinions about that.

About Dungeon Crawl Stone Soup, I am not sure now that it is good example for comparsion about stable vs unstable.

Stone Soup actually developed in much slower and strict pace. There is no cases like “3 monsters every week in unstable” or “3 new spells per month”.

You are starting from a wrong assumption that development team should care about which version players are using.

1 Like

We’ve discussed why we do stable releases in the past, and you seem to be ignoring those reasons in favor of just making assertions about other people’s goals.

Just to briefly restate the reasons.
Complete translations are only available for stable releases, so players who want to play in e.g. German can only play stable releases.
We want a reasonably stable and promoted version of the game for people that are risk averse or new to the game to start with before they jump onto the experimental treadmill.
We use stable releases to focus attention on persistent bugs to keep them from accumulating.
Stable releases provide a focal point for advertising the game, both attracting new players and reminding old players that the game existavand is still being updated rapidly.

Getting longtime players to play stable releases for a long time after a release is not one of my goals. As ZhilkinSerg says, it is literally not something we care about at all.

Having said that I have put a significant amount of time backporting fixes for the 0.E release and i believe ZhilkinSerg did a build that is available. Personally I’m willing to back port about two weeks worth of additional bugfixes to a stable release, but the difficulty of backporting more fixes and the perceived value of doing so drops off rapidly.

If you’re interested in contributing in this area, the #1 thing that is needed is identifying which bugfixes are available and sensible to backport.


Sorry, buy you are definitely misremembering something. We do not discussed reasons of developing stable releases at all. And actually we do not discussed anything normally for very long time.

Considerign reasons- I agree with that.
Actually if there will be backport for at least some fixes it would be great. It is exactly thing I’ve proposing here.

Real thing that worrying me is actually that I actually saw the case then player tried to report something to stable, but his issue was closed exactly because it is for stable.
It is completely reverse of what I’ve seen in entreprise. Here is the only thing that users report is for stable releases.

I may be interested in this area. But from my point of view in my banned state it will be more mockery of contribution, rather than proper contribution. After I will find the fix for backporting- I could not even report it on github.
Yes I can easily bypass that with new account but it will kill whole point of my ban. I’d rather avoid that.

We either do it properly or do not do it at all.