Future AGS development

[timofonic]

[/////

[Calin Leafshade]

It's like FLOSS only less useful for dental hygiene.

[monkey0506]

Tim, it seems to me that you seem to believe that if AGS were made to be full F/OSS-compliant that the entirety of the Linux/open-source community will suddenly flood to make AGS the single most powerful game development engine of all time ever.

The fallacy in this seems to me that they have not thus far shown strong interest in the engine (such as porting the engine to Linux, which has been done but has been maintained by dedicated individuals and not a huge community of developers), and that if that was really the end-goal of the open-source community as a whole, then why haven't they just written their own anyway?

Something you need to keep in mind is that for more than a decade this program has been exclusively programmed and maintained by Chris Jones. The only reason it is now being opened at all is because he has a life outside of the program, but does not want to see it be held back by his own inability to dedicate time to it due to other obligations. In essence, he is allowing us to get our grubby hands into his lovechild's life as an act of kindness, rather than having a strong inner desire to see AGS become a F/OSS-compliant project.

CJ understands well enough what open-sourcing means, but he wants to maintain a level of control over his project that certain licenses would not necessarily allow. His goal is to keep the project alive. The number of people who have already contributed to the editor, along with the number of script modules, templates, plugins, etc., and those who contribute freely of their own time to helping others in learning to use and make games with AGS, and so forth, should all go to show how AGS can be maintained by the community, while allowing CJ to have the final authoritative control that he deserves and expects after more than a decade of selflessly developing the program for anyone to use.

[SSH]

Well, too be fair, ScummVM have said they would definitely consider supporting AGS games if the file format was made open source.

[monkey0506]

Which certainly wouldn't be a bad thing, but historically speaking it's never been a priority of CJ's, and considering the number of people who have, and are currently still developing with AGS, I don't think that not having it is a real deterrent from the engine.

I'm not against F/OSS, I was just trying to explain that for the time being it's not the priority.

[Snarky]

What ScummVM support gives you is instant cross-platform compatibility, which at least in my opinion would be a major bonus for AGS.

[juncmodule]

Well, too be fair, ScummVM have said they would definitely consider supporting AGS games if the file format was made open source.

This matters to me more than anything about the editor itself belonging to some particular license. Whatever is comfortable for CJ and whatever keeps the engine alive is good enough for me.

I left AGS after creating Keptosh and have never seriously approached another project with it. I love AGS, but I feel that with 10 years of no true cross platform support for games, I will never again use AGS. Of course, this breaks my heart, I love the community here. After 8 years of lurking around these forums with no intention of using the editor, I think it says a lot about the strength of the community. I want that back. I want to make games with AGS again. The one thing that would bring me back faster than anything would be ScummVM support (not that anyone wants me back ).

I feel that the editor being cross platform or having a specific license outside of what Chris has already done is just silly. If you are a game developer and you are too lazy to run wine or VirtualBox, or better yet, just run it in Windows natively, then you are not serious about developing games with AGS.

Thanks,
-junc

[timofonic]

///

[Calin Leafshade]

Well herein lies the problem.

the whole reason that AGS is going open source is that CJ doesn't have the time anymore to maintain the engine to the level that he would like but running an open source community takes far more time than the current situation. So that would be counter-productive for his purposes.

Secondly the AGS engine is 10 years old. CJ knows it like the back of his hand i'm sure but *no one* else knows anything about the source AFAIK so no one else could effectively lead an FOSS community.

The fact of the matter is that it will take a while for people to get up to speed with the engine source code to a level that they could effectively manage a FOSS dev community.

[monkey0506]

I was essentially trying to point out what Calin just said with my prior post. I am not in any way against F/OSS. I'm a fan of it really.

That being said though, it's not like AGS could be converted into a F/OSS project overnight anyway. As you've already pointed out, both the editor and the engine alike have been engineered with Windows in mind. The number of users who have ever come here asking for support for other systems has always been relatively low.

That's not to say that there haven't been any, or that there might be some who simply haven't bothered since they knew of the low level of support for it. That being said though, CJ has always tried very valiantly to prioritize his efforts in developing AGS. If nobody says anything, or if at least he can't get a significant number of people to show genuine support, then why would he spend his time focusing on cross-platform compatibility in favour of improving the editor and engine as they currently stand with new features and bug fixes? It's simply not a practical endeavour.

CJ is not "hostile" toward open-source, he simply started developing the program himself and he enjoyed keeping it that way. Just because you create something doesn't mean you have to run around telling everyone exactly how you did it or else you're a greedy, selfish prat. For over a decade CJ willingly gave this program completely free of charge to the world. Wanting to have control over the source seems a reasonable trade-off to me.

As has been pointed out CJ is opening the source slowly because he wants to see AGS continue to develop rather than stagnate as his personal availability to commit time to the project has lessened in recent years.

Seeing as you're so committed to seeing AGS go fully F/OSS-compliant, my suggestion to you then would be to start making changes yourself. We here are not against the idea, again, it is only a matter of priority. If making AGS F/OSS is your primary priority in being here, then do so. As for the rest of us, we will put forward the effort that we deem necessary to helping AGS progress.

I wish you the best of luck in your efforts, and although I'm not likely going to commit myself to helping you rewrite the entire editor and engine from scratch, you do have my support in doing so.

[Sslaxx]

That being said though, it's not like AGS could be converted into a F/OSS project overnight anyway. As you've already pointed out, both the editor and the engine alike have been engineered with Windows in mind. The number of users who have ever come here asking for support for other systems has always been relatively low.

You're conflating "open source" with "cross platform". The editor was converted to open source overnight. Making it cross-platform, though, would be a much longer and very involved task. As for the engine, that we shall have to wait (and see) for.

[monkey0506]

Actually I'm not. I said "F/OSS" which actually implies quite a lot of things above and beyond simply having opened the source code.

[timofonic]

 /

[monkey0506]

But there are other options, like merging to a an already successful project with compatible or similar goals.

An example is ScummVM...

The problem with this line of thinking is that as soon as a project like AGS were to "merge" with another project, it immediately stops being AGS. The AGS community has developed into what it is because it is AGS. The AGS community is not the ScummVM community, and vice versa. That's not to say that we couldn't play nicely with them, but this community has their own unique history that is based around AGS, not some other project.

If AGS were merged into another project, then that history would be lost. I've not been a big part in the community myself outside of contributions here (modules, a couple templates, helping others, largely staying out of religious debates 8), etc.), but the AGS community actually extends beyond the realms of the Internet in the way of Mittens, Brittens, and all the other -ittens where AGS members actually meet up in real life for hang-outs and such.

Beyond that there's things like MAGS, OROW, RON, and others that all go to making this community what it is.

Compatibility with ScummVM would be great, I never intended to even suggest that it wasn't, but directly merging the projects would be detrimental to the AGS community.

In any case though, none of this is going anywhere until CJ decides to open the engine's source. From there we might be able to better consider what could be reasonably done to further the cross-platformability of the engine. As someone said, with regard to the editor itself, emulation options would be a reasonable route for those on other operating systems to take, IMO anyway.

[nimh]

Hi. I'm a linux user who just found out about the AGS being open-sourced, and I'm incredibly excited about the whole idea.  I have in the past kept multiple outdated computers boxed up in my basement just so that I could play favorite games that are no longer natively supported.  I realize that emulators can take care of almost all of that nowadays, but for the guarantee that a game will always be playable, nothing can beat running on an open-source engine (or virtual machine).

I've heard the possibility that the file format won't be open-sourced to make it harder to decompile people's games, and I've got some questions and an idea about that.  Assuming that the rest of the engine gets open-sourced, how hard would it be for someone to write a program that simply grabs the data from in-memory, and writes it out in any format they want?  I suppose only CJ knows the answer.  Even without the engine being open-sourced, if no AGS decompiler currently exists, it seems like that's mostly because no determined hacker has decided to make one.

However, I do understand that people have put a lot of effort into games that they have released for AGS, and some are undoubtably worried about their game data becoming suddenly so easy to steal. So, here's a thought on a solution.  Let AGS have a brand new open file format, but also allow AGS to read in the file through a binary plugin, or a binary callout program.  Then AGS could provide a binary plugin to load files from the existing file format.  This way, everyone who has written a game for AGS already will not have to worry about their game data being trivially accessible. And going forward people have a choice. They can distribute their game in a format that will be guaranteed to be playable everywhere, or they can distribute their game in a format that isn't so easy to decompile. Paranoid developers could even write their own encrypted file formats and binary plugins to read them, but if the API to read files changed in a future version of AGS, their games may become unplayable.  Additionally, if there was a conversion program, the the creators of existing games could be contacted for their permission to convert their game to the new open format.  If they agreed, then their game would be converted. If not, then it would remain only playable via the binary plugin.

I realize that I'm suggesting a bunch of development work that I don't plan to do myself, but I had the idea, and thought it would be worth sharing. In the end, I'm grateful that AGS exists at all, and also that it's going open-source.

[Shane 'ProgZmax' Stevens]

If you are a game developer and you are too lazy to run wine or VirtualBox, or better yet, just run it in Windows natively, then you are not serious about developing games with AGS.

Holy condemnations, Batman!  I guess I'd better stop making games straightaway since I have no use for Wine or Linux!  Better yet, I should go back in time to 2002 and stop myself from making my first ags game in windows since I couldn't have been serious about it back then, either.


Seriously, do you re-read what you write before you post?

[Calin Leafshade]

or better yet, just run it in Windows natively,

Do *YOU* re-read the stuff you are replying to?

[monkey0506]

I think that the view might have been embellished a bit when that was first written, but to a degree I do agree with it. For the time being the editor is built on the .NET Framework, and it would be no simple task to remove that while maintaining the current editor as it is. I'm just being realistic when I say that.

Porting the engine and porting the editor do not have to be seen as mutually inclusive endeavours. Once the source for the engine is released (which is already written in native C++), then those who truly wish to do so can work on porting the engine to other systems and so forth. The editor being written in C# doesn't mean they can't port the engine successfully. If anyone ever does want to port the editor, the source for that is already out there, but in the mean-time, there's no practical reason that I can think of that would make running some form of emulation software out of the question. Sure, it may not be considered "optimal" or as user-friendly as a native editor port, but if you want to use AGS for making your game, then why would it be so bad to have to take an extra step to reaching that goal?

[Khris]

/

Another weirdo who removes all his posts? Some people need to chill about being wrong.

[Snarky]

Tim wasn't even wrong as such, he was just very very insistent on The Right Way To Do Open-Source (in his opinion). And ProgZ apparently misread or misunderstood the comment he responded to. To blank your own posts is such an annoying and immature way to react, though, it quite literally erases any merit your argument might have had in the first place.