Print

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

[Retro]

It's been said a few times, but it's worth repeating that this thread is not a repository for bashing Impaler, regardless of what he has done. If you want to talk constructively, do so. Talking about the 'scumminess' of his 'idiotic/egotistical' plans is unnecessary, and as Foot mentioned, not classy. I'd like to think we're above that.

[Mr Frog]

It's been said a few times, but it's worth repeating that this thread is not a repository for bashing Impaler, regardless of what he has done. If you want to talk constructively, do so. Talking about the 'scumminess' of his 'idiotic/egotistical' plans is unnecessary, and as Foot mentioned, not classy. I'd like to think we're above that.

Sorry, I can't seem to help myself. I'm not trying to insult anybody, really.

[Sabin Stargem]

I think that having Toady team up with someone to put together an Interface version of Dwarf Fortress would be a good idea.  If I understood it correctly, the D# split was about overhauling the graphical capabilities that the game would have, so having another project to do the same for the interface might be good.   Granted, upcoming content will ask for the Interface to be reworked many times, but I think that means we get a chance to polish it up nicely.  So long as Dwarf Fortress is getting new features, the opportunity to rethink the Interface many times could lend itself to perfecting things.

That said, I am not actually sure what the D# split was for, so I took an enlightened guess from what I read on the forums.  Could anyone tell me more about it?

[MrWiggles]

The least I want from the UI is constancy. Right now, there no uniform control scheme for the various menus. I dont think a total UI right is warranted right now.

[zwei]

I hereby apologize Mr. Wiggles for quiting him on something that he did not say.

Problem #1:
The game is still in ALPHA.  Namely all the current UI are, effectively, place-holders.  Creating a more user-friendly UI (which would likely lock down some features) would result in Toady potentially needing to tear-down the whole thing later down the road in order to add more features.]

Sit back and think really hard.

Alpha (aside of murdering psychopaths) is basically testing version. You test implementation, ideas, mechanics and gui.

There is no excuse not to test new, friendly, gui approaches, because that is one of things that are supposed to happen while game is in alpha/beta.

Many features nowadays are placeholders: caravans for example. They will be torn down and redone later on, including scrapping many related things in game. But they exist and are sources of comments (hey, let us import sand! hey, trading goblin rags is overpowered! etc ...) and serve purporse just fine.

As far as scrapping goes, it is not exactly good way to look at it: sure, it will be eventually obsoleted and redone. But it will be done again using experience (and hopefully) and existing code. Besides, has it crossed anyones mind that with a bit of thinking, gui can be done well enough not to require completelly rewrite every times new feature is added?

Aplha is opportunity to improve and refine gui, not excuse for dark ages. That is whole point of being in aplha: write stuff, see how it works, if not, scrap and redo it. People need to overcome enviromentalist coding mentality (not a line wasted!).

Right now, game is really showing what happens when gui is treated as "i'll do it later" thing. Impaler was afraid that Later means Never. Since there will always be yet another feature to add, Later can indeed mean Never, especially if people support this course of action.

---

While are we talking about player contributions compare "V"isualize function with pretty much any visualizer tool.

Notice how ingame visualizer is quite ... crappy. Experimental feature that should have been removed long time ago for all confusion it does. Look at existing, opensource, licence unecumbered and fairly advanced (as advanced as it can get) visualizers. Stonesense is basically what happens if someone focuses on their feature and does not stray. Why is it not intengrated already?

Dwarf therapist is another example for gui that is not even in game yet, yet desperatelly needed.

I can only imagine what would happen to stocks menu (DQL - Dwarf Querry Language!) or building material selection and all the screens when players got power to help out with them ...

[Deathworks]

Hi!

Zwei: While there is some merit to exploring UI options, I don't see it as so important a part of alpha development. While re-doing stuff is necessary as you illustrated quite correctly, especially the UI is a place where you have a lot of polish. As far as I can tell, while the UI does have some influence on the internals of the game itself, those influences only make up a comparatively small part of the work involved with the UI, the rest being aspects that can be changed/replaced without influencing the core of the game. Things like the caravans, however, for the most part involve internal mechanisms of the game and thus yield important results about the engine. This is why I believe them to be of a higher priority than the UI.

And I would be grateful if you did not insist on a GUI. Not everyone here believes that it is the ultima ratio for the game. Thank you.

Deathworks

[frightlever]

There is something that will give other projects a massive advantage over DF: They don't have to worry much about the underlying simulation. Material characteristics and interactions, realistic body structures, and practically everything that Toady spent the last year doing is not presented to the player in any significant way, and therefore might as well not exist.
Imagine how much of combat they could replace with a hidden HP meter, and a wounds screen that has a certain chance of removing a limb on a critical hit. Most people wouldn't notice the difference.

Not to mention they could hard-code all the improvements that Toady is trying to lovingly coax out of the simulation. Heck, hard-coding animal-husbandry alone would give them an excellent selling point. After all, you can't even milk cows in DF amirite?

I suggested something similar on a Space Empires 5 board once. Barely escaped with my life. The trouble with that game (also coded by a single guy, Aaron Hall at Malfador Machinations) was that all the combats were thoroughly simulated during the computers turn, so by late game in a large map you could be waiting over an hour for turns to process.

"Why not just average out these combats?" I asked. Wow! The you-know-what hit the fan. I figured you could still play through the important battles at your discretion, but leave the AI vs AI fights to be decided on a virtual coin-toss weighted with the relative strengths of the sides. Unacceptable to the hardcore fans. Space Empires 4 was popular for about 8 years with a thriving community. Space Empires 5 more or less sank without trace within two years.

The problem with hard-coding behaviour, and what Impaler may be missing judging from what's been said earlier, is that while you may be able to make a more efficient game by simplifying the calculations, you're going to lose a lot of the character. The procedural story-telling aspect.


The UI consistency annoys me too, never sure whether I should be hitting +/- or arrow keys or number keys or //* so sorting that out would a start. And why require a shifted key to skip up and down z levels? That's just mean.

Talking about having to rip up the UI every time a change is made doesn't seem right. Why would it require anything more than re-jigging some XML? Or, perhaps I should say, why couldn't s UI be designed so that it's that simple.

[Narmio]

While there is some merit to exploring UI options, I don't see it as so important a part of alpha development. While re-doing stuff is necessary as you illustrated quite correctly, especially the UI is a place where you have a lot of polish. As far as I can tell, while the UI does have some influence on the internals of the game itself, those influences only make up a comparatively small part of the work involved with the UI, the rest being aspects that can be changed/replaced without influencing the core of the game. Things like the caravans, however, for the most part involve internal mechanisms of the game and thus yield important results about the engine. This is why I believe them to be of a higher priority than the UI.

Were this a closed alpha of a funded project, I would entirely agree with you.  However, there is another variable to consider.  Let's assume work spent incrementally improving the interface that will be replaced eventually is useless (which it isn't, really, because iterative design and experimentation is of value even if it fails).  Even then, if interface improvements attract more players, then they attract more donations, and it thus improve the project as a whole - even if they themselves are thrown out long before the game is "finished".  The only question is: what interface improvements are sufficiently desirable that they'll improve the game *now* to the point where their long term contribution in players and donations outlays the short term cost of developing something that won't be in the final project?

[Rainforce]

Serious question:  How old is Impaler?

Serious answer: throw a d12 and tell us the result?

[...]
EDIT: Speaking of lines not to cross: interactive fiction (a.k.a. text adventures) has been mentioned elsewhere on the general general discussion forum, and I got this idea for a little interactive fanfiction (basically the player being an adventurer who can visit a dozen of rooms in an abandoned fortress and encounter some classic jokes/bugs we all know and love). Something like that would be alright, wouldn't it?

That would be some nice fanwork XD

[Rose]

Impaler's 28, untill the thirteenth.

just saying.

[Rainforce]

Impaler's 28, untill the thirteenth.

just saying.

I feared that it would be something like that............if you look at it from a certain point of view, <13 would make it a lot more acceptable...

anyway: good response of toady on the other hand, I cant think of a more mature behaviour 

[Jackrabbit]

Impaler's 28, untill the thirteenth.

just saying.

Thank you for putting a stop to this, it's getting ridiculously sour.

[Nexii Malthus]

Impaler's 28, untill the thirteenth.

just saying.

Gotta hate it when people bash age for no reason when the reality is very different.

[Deathworks]

Hi!

Talking about having to rip up the UI every time a change is made doesn't seem right. Why would it require anything more than re-jigging some XML? Or, perhaps I should say, why couldn't s UI be designed so that it's that simple.

If you want something to be very simple, it means that you have to spend a lot of thought and time into its developing. If you want a good user interface, you have theoretically two options:

1. Cast it exactly based on your needs. That is basically what we were suggesting, making a UI that fits the needs of the game perfectly. Putting such an interface together is not too difficult a task as its capabilities are well-defined, but you have to change the UI as much as you change the game. Thus, the ripping apart of the UI.

2. Make a flexible UI core that can be modified easily (your suggestion). Initially setting up the UI would be somewhat similar, maybe a little less than the work needed for the first option and you would have the benefit of editing it more easily. The problem is that before you can set it up, you need to create that flexible UI and if it is to be any good, you better make it very flexible with a wide range of functionality to protect against unexpected requirements. This means you would need to invest A LOT of hard work into the UI right now. And to make matters worse, should something change in the basic design concepts, you may still need to rework parts of it (or, in a milder case, find that some of the code you had developed during those months was completely unnecessary in the end).

This is at least the way I see it and why I think that going by the first design option using only a basic UI for the alpha versions is the best approach for the project.

Were this a closed alpha of a funded project, I would entirely agree with you.  However, there is another variable to consider.  Let's assume work spent incrementally improving the interface that will be replaced eventually is useless (which it isn't, really, because iterative design and experimentation is of value even if it fails).  Even then, if interface improvements attract more players, then they attract more donations, and it thus improve the project as a whole - even if they themselves are thrown out long before the game is "finished".  The only question is: what interface improvements are sufficiently desirable that they'll improve the game *now* to the point where their long term contribution in players and donations outlays the short term cost of developing something that won't be in the final project?

Of course, that question is really the making/breaking point of the UI debate. If the positive effect of the UI was only minimal, it may very well not be worth it. And at that point, I am really glad that I am not the one to make that decision. Sorry, Toady One, but I gladly leave that responsibility to you :) :) :)

Rainforce: Well, I have a lousy track record with lots of unfinished projects. I do have a few ideas for a few famous incidents/bugs to be implemented, so I might at least put together a little something that is good for a few laughs (of course, I am violating the advice of the DM4 by making a piece of interactive fiction based on insider jokes, but that is really the point to it :) :) ).

But no promises there, and I don't think I will really bug Toady One about his opinion on this before there is really at least something to show. (I am also pondering making the source code of the interactive fiction available so that people can expand it/improve it as they see fit).

But it's currently just my sand castle in the sky.

Deathworks

[frightlever]

DW - I love this IF, ever play it? http://ifdb.tads.org/viewgame?id=4glrrfh7wrp9zz7b

Where you planning on writing something totally from scratch or using one of the available languages, like Inform?