Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Miko19]

This discussion has went on for quite some time and most people still didn't understood what I meant....

Let me say it differently: What I originally wanted to say is that Toady keeps on adding new features, which elongates the release, while he does NOT fix any bugs because he's occupied with adding content. I never said that I want him to rush with giving 33.01 to the public, but many of you (newcomers) have misinterpreted my words as "another nab dat ask for stop wrok and release game". Instead of solving problems with the game he keeps on getting features instead of fixing up what's in it already, delaying the release AND letting a lot of bugs in, which will require fixing after a release instead of being taken care of right now. How many dev logs have you seen about fixing bugs and how many about new content? There's my concern.

Cheers.

PS:

Guys, you're missing the point.

I very occasionally suggest things that I think might be interesting. But I don't demand them. I simply front ideas, because Toady has a suggestions forum set up.

This isn't a commercial computer game. It's art. Art you can interact with, art you can download, but art nonetheless. Because of that, we are not Toady's muse, nor should we try to be. We are watching a man sculpting, from the marble of his mind, the greatest computer game ever made. We watch, we throw some cash in the donations box, we appreciate, but...I do not criticize, except for obvious bugs, because it is not my game. I can't complain about what I didn't paid for, in this case.

It doesn't matter how big and expanded your sculpture\painting\anything is if it's lacking in polish and full of incomplete elements.

[MrWiggles]

What do you mean he doesnt fix any bugs?

Out of all the release for DF, minus the very first one, they've all contained bug fixes. In fact, if you take a count of the release, and their primary contents, you'll notice that ~50% of all release, are just bug fix release. The last 6 releases were just bug releases.

The more amusing stories from we hear from the Dev Log, is Toady coming across bugs, and then quashing them. The period, we seem to be in now, is him going over the newly added content, quashing the more obvious, and bugs and cleaning up the features.

[Starver]

[...] he does NOT fix any bugs because he's occupied with adding content.

It doesn't matter how big and expanded your sculpture\painting\anything is if it's lacking in polish and full of incomplete elements.

I read a mutual contradiction in that first complaint and the second statement.  Maybe I got the intent wrong, but it does seem that way to me.


It appears to me that we're getting more things filled in, the way this is going, than a pure bug-fix crusade[1] merely on the existing code would achieve.  A number of those bug-fixes would also be on features marked down to be deprecated and superseded by new (maybe initially buggy, but at least now worth sorting out to attain the originally perceived standards) versions that would wipe and render obsolete all the hard work spent to for only mildly broken (awkward to solve, but not fatal in use) aspects of the game.

It's a YMMV issue, naturally.  As is obvious from this thread.


(I also echoed that ninjaesque MrWiggles post, but no need add my own talk about that aspect of the in camera bug fixing process.)


[1] We already had a bit of one of those along the .31 run, not even counting the "bugfix the latest added features" effort which we all expected and got anyway.

[Jelle]

In my opinion Toady is definatly adding to much in one go. I would think it's best as soon as one new feature has been implemented fully to go ahead and put it live for testing.
There's simply so much you can tweak after some good feedback, especially if it has been implemented recently so it's still fresh in the mind.
Also if it turns out there's a huge flaw in something that has been released it can be detected before building on it further.

But I don't think opinions about his developement schedule matter much, it's their project and they do things their way.

I kind of wish a new version would be released soon now though, it sounds as if it'll be aweome I can't wait to try it out.

[MrWiggles]

In my opinion Toady is definatly adding to much in one go. I would think it's best as soon as one new feature has been implemented fully to go ahead and put it live for testing.
There's simply so much you can tweak after some good feedback, especially if it has been implemented recently so it's still fresh in the mind.
Also if it turns out there's a huge flaw in something that has been released it can be detected before building on it further.

But I don't think opinions about his developement schedule matter much, it's their project and they do things their way.

I kind of wish a new version would be released soon now though, it sounds as if it'll be aweome I can't wait to try it out.

This assumes that Toady doesnt let anyone test as he goes on. I dont think this is true, at the very least, Three Toes plays the dailies in some capacity. I'm sure there are others, probably from the community, that also get to play daily builds.

[Brian]

While it's entirely up to Toady, I'd offer the suggestion of making a little scrum team out of himself for the purposes of staying on track. Anyone on any creative venture can easily go off course--Dwarf Fortress would be no exception.

The basic notion of scrumming would be that Toady would plan one interval (typically 2 weeks, maybe a month) with the intention that at the end of the interval there's a new release with some sort of value added (designed, implemented, tested). He could prioritize bugs into the same system as he saw fit.

The advantage to the players is more-frequently improving game. The advantage to Toady is a steady stream of goals that can actually be accomplished. I don't know how he feels emotionally about his current release cycle, but I believe my suggestion would decrease any guilt or anxiety that he feels now (if any).

Long term mini projects (eg 1-6 months) are still possible while scrumming, assuming you can get the system to a working state at the end of each interval. I think Toady can do that.

Again, it's up to Toady. I look at what he's been working on and there's plenty of tangible good that he could have released throughout the year had he planned it that way. I could have neat cities with at least vacant catacombs right now, but instead I'm waiting for vampire hiding. I'm not saying vampire hiding isn't important, but surely I can have upgraded city generation without it.

[zwei]

Problem with scrumming is that it takes considerable resources.

Two week release would mean that he would spend something like 80% time handling that (as he is single dev).

Dunno how much it takes to just get new version out - compile, do smote test, write notes, package, upload to server and mirrors & edit news, post threads, pray it does not have bug that will require new release tomorrow. Half a day seems generous.

---

As for new features vs. bugfixes, my stand is that bugged feature is is some ways worse than feature not implemented at all and while new features are great, old bug workarounds kill the mood.

[Brian]

Problem with scrumming is that it takes considerable resources.

Two week release would mean that he would spend something like 80% time handling that (as he is single dev).

Dunno how much it takes to just get new version out - compile, do smote test, write notes, package, upload to server and mirrors & edit news, post threads, pray it does not have bug that will require new release tomorrow. Half a day seems generous.

---

As for new features vs. bugfixes, my stand is that bugged feature is is some ways worse than feature not implemented at all and while new features are great, old bug workarounds kill the mood.

I wasn't trying to address bug fixes vs new features, but adapting something like scrum would give Toady more flexibility to switch between the two modes without having to merge old code around to fix bugs.

The overhead of adopting scrum is really what you make of it. For a one/two man team it could be an hour a week or less. There would be more ways to implement scrum poorly than well--no argument there.

As for build and test, that makes me wonder what Toady's system is like. I've had the pleasure to work on a couple commercial products for different companies with automated build/test/releases/release notes. While it can be a massive investment to get going, it's been very helpful to my teams.  I can't put exact numbers on things but if Toady can't kick off a build and have completed compilation, test, and smoketest done after he takes an hours' nap, his infrastructure is slowing him down.

[warhammer651]

I don't think he's adding too much. Half of this won't affect the basic mechanics the way 0.41 did when it was released

[malvado]

I've probably not been around too long (2-3 years I guess) but I have to be honest that I'd rather wait another 6 months and see the features he's adding now completed.
I mean , right now the project is turning towards the road of awesomeness , in my opinion just a few small things needs to be fixed (in addition) and the game will be incredible :

*More ways to abandon forts.
*Roadbuilding and Civilization Linking (Build road to humans that never comes with caravans to ensure trade later on etc).

[shadus]

After 8 months of waiting for the new release I got a bit concerned with what's happening.
As you can see from the dev logs, every time some feature is fully finished, Toady decides to add another one, which results in the need of fixing the new one, and along the way he adds another. And another. And he has to fix it again. Then adds another one...

You see where I'm going with this? :\

Anyone else feels the same way as I do, that there's just too much being added instead of finishing up what's there?
I'm all in for new features, but please add them AFTER releasing a new version finally.

I don't mind him adding features the way he has been, looking back its kinda the way we got to where we are now which isn't a bad place.

That being said, be nice to see some regular releases even if they weren't completely bug free, or a nightly build system, or something...  a half year with no release for a alpha/beta product is a long time.

[Sutremaine]

Nah, a year and a half is a long time.

[tmrt]

The wait between 40d and DF2010 was long as well. The result definitely was worth it and I expect something similar to happen yet again.

[FearfulJesuit]

And I suspect we'll have a release by the end of the year.