ADOM Wiki

[adom-admin]

Uhm, I wouldn't be sure about it. Those things you listed doesn't really forejudge about anything(especially first two things - Creator already published parts of source code on blog, so I think that "Source Code" section was designed for parts or examples of source code). I'm also not convinced if releasing JADE as open source really would be good thing.

Anyway, no matter if JADE will be open source or not, I'm going to send TB some postcards when it will be released.

Thanks ;-)

Right now I do not think that JADE will be open source... for much the same reasons as ADOM (and BTW I by now have bought a commercial license for IntelliJ).

IMHO that's not that much of a problem with JADE as the public APIs for it will be available and it thus will be possible to program extensions (e.g. items, monsters, a new UI interface (without having to know all the sources). I polan to release limited snippets of the sources to illustrate concepts and show how some things are done.

As for JADE I'd like to keep the complete sources closed to (a) prevent stupid variants I personally do not like and (b) keep some secrets better protected. Right now I'm checking various means of code obfuscation for Java to achieve that just.

We'll see.

[coppro]

I think that variant management is not really dependent on whether or not something is open-sourced or not. Generally, a good open-source project doesn't get a lot of variants - contributions head back into the main code tree. Doubly so for a program which can have variants written (to an extent) without needing to edit the sources.

I'm not saying that open-source is necessary - there's no particular reason why JADE would need it. But Java obfuscation seems to be relatively difficult (don't have a lot of experience in the issue, but I know how Java works well enough) - it's quite easy to see whether a particular function is called for instance. And if you try and obfuscate code via tricks like reflection, you end up sacrificing speed. And most code will eventually be cracked.

You could always do it half-and-half - make the fundamental engine OSS, but keep code to any easter eggs or storyline components or mechanics private. But that would make it easier to decompile and lead to likely more variants.

Personally, I'd like to look at the source, because I would like to see how a Roguelike is written in a language other than C, and that has more than 7D to be written in (there are a few others, but I don't trust them for some inexplicable reason. I trust The Creator though.). I'm a curious student!

[Epythic]

I think that variant management is not really dependent on whether or not something is open-sourced or not. Generally, a good open-source project doesn't get a lot of variants - contributions head back into the main code tree. Doubly so for a program which can have variants written (to an extent) without needing to edit the sources.

I'm not saying that open-source is necessary - there's no particular reason why JADE would need it. But Java obfuscation seems to be relatively difficult (don't have a lot of experience in the issue, but I know how Java works well enough) - it's quite easy to see whether a particular function is called for instance. And if you try and obfuscate code via tricks like reflection, you end up sacrificing speed. And most code will eventually be cracked.

You could always do it half-and-half - make the fundamental engine OSS, but keep code to any easter eggs or storyline components or mechanics private. But that would make it easier to decompile and lead to likely more variants.

Trust me, Obfuscation works. If you need an example, do a google search for "wurmonline" (no i dont play it, but i once tried to get their sources ), download the .jar and decompile it (there are a few online decompilers). then, look at what you get.

Basically, it renames all your variables and performs some other tricks that even make it faster. If you have some hundreds of thousands of lines of code and neither comments nor names - you are out of luck.

That does *not* mean that you cant edit it. For example, cheating should be easier than if it was written in C.

I'm a curious student!

Me too.

[Epythic]

Thanks ;-)

Right now I do not think that JADE will be open source... for much the same reasons as ADOM (and BTW I by now have bought a commercial license for IntelliJ).

IMHO that's not that much of a problem with JADE as the public APIs for it will be available and it thus will be possible to program extensions (e.g. items, monsters, a new UI interface (without having to know all the sources). I polan to release limited snippets of the sources to illustrate concepts and show how some things are done.

As for JADE I'd like to keep the complete sources closed to (a) prevent stupid variants I personally do not like and (b) keep some secrets better protected. Right now I'm checking various means of code obfuscation for Java to achieve that just.

We'll see.

For some reason I did not read that post.

While I accept your decision (your code, your choice), I again say that I dont like it, this time directly to you. Not that anyone will listen to me...

[Grey]

I for one am quite glad JADE will be closed source. I don't want all the secrets of the game laid bare, and I don't want idiots making poor variations. Also means we won't have fools posting bug reports about variants etc.

[F50]

while I am not really opposed to closed-source, I think the decision to close-source it is based upon misconceptions of open source software.

I don't want all the secrets of the game laid bare

"laid bare" is hardly necessary. A minor attempt to obfuscate code in a separate file prevents people from casually looking in there are finding stuff out, and combined with the goodwill of a community there isn't much to worry about from regular people like you and me. I have heard that people have used code-diving to figure out some stuff pertaining to the scroll of omnipotence as well as some GB information, even though it is closed source. Naturally, few people have, but mostly out of a respect for the Creator.

...and I don't want idiots making poor variations.

This is very uncommon with open source programs. Most attempts to create forks are ignored.

the four possible outcomes for a fork are:

1. The death of the fork (example: libc/glibc). This is by far the most common outcome; indeed, many forks never receive enough support to ?die?.
2. A re-merging of the fork (example: gcc/egcs); this is where the projects rejoin each other (though one or the other may be the dominant source of the combined effort).
3. The death of the original (example: XFree86/X.org).
4. Successful branching -- both succeed, typically catering to different communities (examples: GNU emacs / xemacs, OpenBSD).

http://www.dwheeler.com/oss_fs_why.html#forking

for a JADE fork, 2 is only possible of the creator agrees with the fork in which case it really isn't a fork. The creator has enough fans that I think 3 is impossible too. Which leaves 4. What is the likelihood that a rogue project would succeed?

Some things going against that possibility.

1. Any such fork would be against the wishes of the creator and would have to overcome significant animosity for doing so.
2. In order for such a fork to succeed there would have to be a dedicated and skilled developer that is willing to undermine such wishes.
3. Such a fork would both have to attempt to be better than JADE (a fork of JADE would not be called JADE) while being controversial enough

Also what would happen if the creator got hit by a bus? Despite being a somewhat ridiculous worry, this is actually a question that often crosses my mind. This question also arises with another free game I know of, Transcendence, which is a 2d (in other words, graphics, not ASCII) space-shooting roguelike. It is relatively near 1.0, and the last thing I want is for that project to die on me, for any reason.


Such a long post and I really didn't want to say much. I don't seriously think that this post would convince anyone of anything, although I hope it does. It would hardly be in the spirit of FOSS to do anything more than ask for a project to become open source, and whatever happens, I do support the creator in making JADE. If I did not, then I shouldn't be here.

[Sradac]

.

This is very uncommon with open source programs. Most attempts to create forks are ignored...

Way untrue. Seen all the crazy and silly and sometimes just plain dumb *band variants out there? Yeah there are some good ones but there are plenty that go the munchkin route a little bit too much. And they dont do things quite right. How the heck do you expect playing as a dragon race to be able to fit into the door of the prancing pony in bree and not have everyone scream their heads off? Its bad enough I can walk around Terinyo as an orc or troll and not have everyone run as soon as they see me.

[F50]

Angband and Nethack encouraged forks I believe. I don't think there was any community protest or disapproval. Also, most of those forks are dead. The only "successful" ones I know of are Slash'em, Zangband, Nethack itself, and ToME.

If you consider the majority of open source programs out there, especially GPLed works, I see few forks. BSD is a fork-conducive license, GPL is not fork-conducive (you have to keep it GPLed, give credit to author, not mix with proprietary code etc).

[DirtyFinger]

New adom Wiki at wikia.com: http://adom.wikia.com/wiki/Main_Page

Since I come from the nwn2 guys and the Epic Character Build Repository, my work there influenced the structure and visuals I use for adom.
I also focused on Classes and Races first.

I haven't figured out yet how to change the color scheme and other gui thingies, but the general framework is there.

Category:Races
|- Human, Troll, etc.

Category:Classes
|- Barbarian, Archer, etc...

Category:Skills
|- Alchemy, Alertness, etc ... all skills belong to this category.
|-Category:Skills that excel at high values
|-Alertness, Backstabbing, etc.
|-Category:Skills that need materials
|-Bridge Building, Fletchery, etc.
|-Category:Skills available to everyone
|-Climbing, First Aid, etc.

Also check out the Attributes section: http://adom.wikia.com/wiki/Attributes
There's probably more stuff you can write in it, but I'd like to brag with the attribute tables which are sortable by clicking on the respective columns.

Races are all done and each race entry contains both the description and the starting attributes ... something no one before seemed to regard as particularily useful, for some reason. Races still need more details, like the mechanics behind the slow Trollish learning ability etc.

Classes are all created although 80% are stubs, meaning that these articles have the attribute modifiers but still lack the description and the power lists.
See the Archer article for a complete example. This is just stuff from the manual. I'd like to see things like race recommendations and maybe basic tactics, playing tips, etc. for each class.

I also gathered some links and put them here: http://adom.wikia.com/wiki/Links
Feel free to add more. Even links that just mention Adom as one of many roguelikes are encouraged in the respective section.

What do you think ?