Print

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

[Footkerchief]

So what I'm hearing is that Toady hasn't expressed any support for the idea, or done any of the (substantial amount of) work to implement it.  Not very worrying.

[G-Flex]

That's correct as far as I know, but Baughn was taking a fairly authoritative tone with his "sneak preview", so I'm not sure anymore.

[forsaken1111]

That's correct as far as I know, but Baughn was taking a fairly authoritative tone with his "sneak preview", so I'm not sure anymore.

Isn't this something that happens pretty often?

[Orkel]

The .09 patch releasing tomorrow (most likely according to devlog) makes me a very happy orkel indeed.

[Baughn]

Ye gods, that's a lot of speculation.. All I was trying to do was show something a little neat. >_>

Look, all I was doing in that email was pointing out ways this *could* be used. It certainly wasn't meant to be a statement of intent.

Considering how annoying this is getting, I also very much doubt it'll be publically available as a console. It's a pity, but if your worries have any weight at all, it's not worth it.

[G-Flex]

I wasn't trying to engage in speculation; I was pretty much going by what you had said yourself.

Look, all I was doing in that email was pointing out ways this *could* be used. It certainly wasn't meant to be a statement of intent.

I'm aware; I was just providing it for context, and so people could see where you were coming from.

[Psieye]

I wasn't trying to engage in speculation; I was pretty much going by what you had said yourself.

However, Baughn WAS speculating, so you were engaging in it.

Mrrrrgh, well it's pretty much been decided so no point continuing I guess. A waste of a good opportunity with valid concerns to weigh against and possible solutions. There are times when the panic button needs to be pressed (and sometimes, mashed) but this incident could have been dealt with more tactfully. Ignore me, I've had enough bad things happen today that I'm sharing the bitterness.

[G-Flex]

I apologize for stirring up any sort of unnecessary argument here, but I figured since Baughn was posting about it, it was a decent enough time to discuss it.



Also, I'm not arguing that any sort of debug console within DF is a bad idea, just that there are issues and implications (some of which aren't always obvious) that need to be considered before jumping into anything of that sort.

[abadidea]

Personally, I would love a scripting language in some form, for extremely custom AI and such. But the discussion over the last page or so has left me extremely confused? 

[arghy]

We need to clone toady so he can work in all hours of the day

[teloft]

We need to clone toady so he can work in all hours of the day

Well, we could give him tons of monny and ask him to quit his other job and perhaps get some more people to help him.

[G-Flex]

There is no other job.

[arghy]

We need a page detailing toadys days so we can assure maximum usage of his time--YOU DONT NEED TO EAT YET TOADY THAT BREAKFAST WAS HUGE.

[Toady One]

I haven't made any decisions.  There are lots of things that have come up in the past where scripting has been a natural extension/solution and I've been trying to find time recently to think about some of the options.

Even the most rudimentary standing production orders implementation is going to have to involve some kind of if-then structure and a way to get at certain game values/sums, and then it's a question of how that's going to work.  My own scripting stuff (which exists for some of the diplomacy but is quite godawful) would be too clunky and would be reinventing the wheel.  If a scripting language is used, it's a matter of whether it is available in the raws or if you'll just be able to enter some simple actions from a manager style screen, from which scripts are then built internally.  I suppose there would be utilities written that grab onto that and stuff scripts into the system even if there aren't scripts in the raws, but it would be limited by the script-DF interface in any case, as far as I know.  It would be a little strange to have universal management scripts that can do all sorts of things and be run any time, in part because of the accessibility issues, and because it's just sort of strange and unnatural to me to have a game that works that way, but that's exactly what a standing production order is.

There are future issues -- take creature special abilities or artifact magic.  It seems like that's inevitably going to involve some kind of scripts, and at that point, there will have to be a measure of interface with the guts of the game.  It's a lot of work to do that, as Footkerchief mentioned, however, it's a lot of work any way it is done, because the magic/abilities have to have effects/targets/etc. and that'll have to be kept up to date and access various parts of the game, pervasively if the typical variety is allowed.  It's similar to an actual API for the graphics/etc., but I don't think it'd be nearly as difficult to maintain, since there would be a fraction of the options, and it doesn't raise any of the other concerns.  In any case, I'd always wanted to have a game where you can do a variety of magic effects with a lot of freedom, but the more power you have, the more like programming it seems.

Random creature generation is another example.  The forgotten beasts are hard-coded now, but I was considering starting with the random dragons out in the raws.  This would involve scripting again.  I don't know if this is at all similar, in terms of the concerns people are bringing up.  It seems like it's a matter of how central something is to the game.  The ability of some people to build water computers doesn't seem to bother most people, because that's sort of difficult and specialized, whereas presumably a script would be easier to use and would have more applications.

[HammerDave]

I have mixed feelings on the idea of scripting in DF.  Part of the attraction of the game for me is the continuous interaction model, where the player has to pay attention and constantly issue orders.  For me the challenge is how long can I keep it going with a low idlers number, processing raw materials into finished goods efficiently, and not running out of booze.  And design a fort for the circumstances of each new embark that is defensible, aesthetically pleasing, and efficient.

On the other hand, the legendary programmer in me wants to see how well I can automate it, or how cool I can mod it.  Could I write a program that plays out the whole 1st year in excruciating detail, as well as a great player can do it?  Could I devise a mod that would randomly resist well-known player strategies to add difficulty while leaving the basics of the game unchanged?  Or squeeze some more FPS out of it by very carefully laying things out in a way that makes it run as efficiently as possible?  (or squeeze some FPS out of it by analyzing Toady's algorithms and suggesting tweaks?) 

Technical note: The approach used by Civ4 seems to work pretty well.  The game core is kept hidden and certain pieces of information and function points are exposed via a formal API.  They use Python which makes reasonably good sense to anyone who is familiar with at least one procedural language.  Scheme might be a little too dense for casual use by people who have only seen procedural before -- though if a functional approach is necessary then by all means use whatever language makes sense.