3.3.0 Hobbit Rogue / Death

[CliffStamp]

You are right that this is a real danger. We'd have to make sure that the population and fecundity numbers were such that this didn't happen - that it just isn't possible to denude the dungeon of large numbers of species.

You can factor in time in the dungeon as well, so it will only really be noticeable to people who do either in extremes, i.e., item/scum or monster/genocide.

[Philip]

About the developement vs maintainership issue, I have a suggestion. I fully support v4 as the devteam variant so maybe we could set up a poll of sorts every month about which new features should be added into angband proper. For example this month we'd chose among other things the ego-item system. I have no idea how many active angbanders there are that check the forum and play v4, but I suggest taking a certain number and features that recieve fewer votes than necessary stay in v4.
I feel that such an approach would work quite similarly as in the age of submitting patches to the maintainer and starting variants.

[Magnate]

About the developement vs maintainership issue, I have a suggestion. I fully support v4 as the devteam variant so maybe we could set up a poll of sorts every month about which new features should be added into angband proper. For example this month we'd chose among other things the ego-item system. I have no idea how many active angbanders there are that check the forum and play v4, but I suggest taking a certain number and features that recieve fewer votes than necessary stay in v4.
I feel that such an approach would work quite similarly as in the age of submitting patches to the maintainer and starting variants.

This is more or less how it's likely to happen. Before v4, we always used to poll about major changes, and have discussion threads asking for and dissecting ideas. So far there isn't anything in v4 that's really mature enough to port across fully (none of ego items, combat or traps are really 'finished') - but when there is, I suspect a poll and discussion will ensue.

(Nomad's new rooms are quite popular and uncontroversial, so those could actually be porrted across - but I'm not sure it really needs a poll, because there's never been any criticism expressed!)

[CliffStamp]

It is of course impossible to gain complete consensus on where the line is, and we have certainly stepped over it several times in the eyes of many players.

No one is going to be perfect, either in coding or requests, all that you can ask for is to respect the other party and try to work out a solution realizing the positions and how demands are fairly absurd just as ignoring each other is also just as absurd.

if what we're maintaining is not considered "Angband proper", anyone can step, realign the code with the one true Angband, and release the next version of Angband proper.

If you decide to be on the dev team of Angband then this is a commitment to actually be on the dev team of Angband not on BeiberBand. If the dev team isn't willing to work with the user base then it really needs to step down from that and continue with the variant they are coding.

The user base however needs to make the same realization and that if they are so adamant about coding and do not respect and allow the dev team a lot of freedom to explore what they want then there will quickly be no dev team outside of variants which by definition do exactly what they want to.

Again, it is just the willingness to work together.

For example, I would really like to see Angband be more immersive in that what you do has effects on what happens in the future. This should ideally be transparent if a user does not actually try to do something and just proceeds at random, but if a user actually wishes they can modify the environment.

Take for example if you play a hobbit and everything you but is all light armor as you are rping. The store owners see this and start offering more light armor. If you just buy armor at random then you don't see any effect. But if you are a half troll and the first thing you do is select all heavy armor, and you always select heavy armor even above light ego's you see the effect in game.

Now I am one person, it is kind of mental to assume that this means it has to be a dev-team priority. But lets assume that this becomes popular, people start to talk about it and want to make Angband have more of these features. If the dev team ignores this then it is hard to see how they can call themselves the dev team for Angband proper as they are ignore the user base.

[Derakon]

Thinking back, I believe that the current dev team has a stronger integration with the community than any previous maintainer has ever had. We do a lot more discussing of potential features before they get implemented, and a lot more analyzing of those features once they are implemented; my memory of the old days is that a maintainer would simply implement changes on his own, and very rarely would the community backlash be sufficient to get him to change his mind. To be fair, the maintainers were also a lot more conservative back in those days, so there were fewer controversial changes going in.

[Magnate]

If you decide to be on the dev team of Angband then this is a commitment to actually be on the dev team of Angband not on BeiberBand. If the dev team isn't willing to work with the user base then it really needs to step down from that and continue with the variant they are coding.

The user base however needs to make the same realization and that if they are so adamant about coding and do not respect and allow the dev team a lot of freedom to explore what they want then there will quickly be no dev team outside of variants which by definition do exactly what they want to.

Again, it is just the willingness to work together.

For example, I would really like to see Angband be more immersive in that what you do has effects on what happens in the future. This should ideally be transparent if a user does not actually try to do something and just proceeds at random, but if a user actually wishes they can modify the environment.

Take for example if you play a hobbit and everything you but is all light armor as you are rping. The store owners see this and start offering more light armor. If you just buy armor at random then you don't see any effect. But if you are a half troll and the first thing you do is select all heavy armor, and you always select heavy armor even above light ego's you see the effect in game.

Now I am one person, it is kind of mental to assume that this means it has to be a dev-team priority. But lets assume that this becomes popular, people start to talk about it and want to make Angband have more of these features. If the dev team ignores this then it is hard to see how they can call themselves the dev team for Angband proper as they are ignore the user base.

I think the difficulty I have with this post is that it implies that the devteam is a unified entity and that the userbase is another unified entity. In fact each is made up of individuals with wildly varying views and opinions and ways of acting - and one is a subset of the other.

Even if the userbase were to agree on the (un)popularity of an issue (and in my 3.5 years on the team I cannot recall any such consensus), the fact remains that if nobody actually codes any changes, nothing will get added (or removed). Now this has become 1000x easier with github - anyone can offer a pull request whenever they feel like it, and there is far less excuse for the devteam to 'ignore' such offers (c.f. patches sent to previous maintainers disappearing into black holes) - but the problem is not that we ever ignore pull requests (we don't), but that they don't come nearly as often as the opinions do.

This is the flip side of the v4 issue - we no longer get criticised for changing things people don't want, because all the potentially controversial changes go into v4 instead. But there's no solution (other than pull requests) to changes people *do* want but which none of us have time or inclination to code.

So I'm still of the view that Angband proper is whatever the devteam releases, unless and until someone else releases something that's deemed to be "more proper". (An obvious candidate would be Eddie's 3.0.6-based variant, should he choose to release it.)

[CliffStamp]

I think the difficulty I have with this post is that it implies that the devteam is a unified entity and that the userbase is another unified entity. In fact each is made up of individuals with wildly varying views and opinions and ways of acting - and one is a subset of the other.

It is never the case that relationships are so easily quartered, that is why they are among the most difficult things to maintain. In this case as everything is volunteer it can be among the strongest or weakest depending on how it is approached. If any attempt at force is used then volunteer relationships almost immediately break down for example as it is absurd by nature.

By the way, is anyone compiling Angband on Windows with some free compiler?

This is the flip side of the v4 issue ..

I think that is am amazing thing because you can throw out radical ideas and people can play and experiment. It can be frustrating when it only half works, when you find artifacts which have no finesse score so they get 1 blow, or when you get weapons generated to heavy to lift, etc. or when some really cool feature (rune ID) breaks what was a useful ability (squelch) . But you have to appreciate the fact that you get to fool around with some pretty interesting ideas which are in pre-alpha and keep that in mind, pre-alpha by definition is only partially complete and not tested.

[Magnate]

By the way, is anyone compiling Angband on Windows with some free compiler?

Yes, MinGW ("Minimal GNU for Windows") is the recommended Windows build environment. Angband can also be built on Windows using cygwin (as david3x3x3 posted today), and even Eclipse. We now have a "compiling.txt" which ships with the source which documents what we know.

Blubaron builds using Visual Studio; I can't remember whether his instructions are included (I think they are, but in a different file, under src/win/).

[CliffStamp]

Thanks, I am going to look at a build.

On a rather ironic note, I realized that the exp levels are actually just in the TXT files so I can adjust them as high as I want trivially. I just don't like that if I actually clear pits/vaults that I hit CL 50 far before DL 100 so I actually avoid them as I don't like playing with a max exp (cl) character.

[silsor]

Careful with that "everything in LOS" rule -- it could possibly result in lost artifacts in the edge case. If we want to completely trivialize ID then we could as easily just make it ID everything in a radius-4 circle and have no recharge time to speak of (in the case of the rod).

Oh, like a ball spell?

It breathes identification. -more-