Browsing this Thread:
1 Anonymous Users
|
Re: Developers
|
||||
|---|---|---|---|---|
|
Home away from home
 
|
Quote:
You can't make a project more interesting to some other group without changing the project. It was already decided what kind of project AROS would be and that is clearly stated on the web page. Now it will either be interesting to some developers or it won't. If it doesn't interest some developers, I don't think changing the project to interest them is the answer. If you do this enough, it's no longer AROS at all. I say leave it the way it is so you'll only get developers interested in AROS.
Posted on: 2004/9/11 22:00
|
|||
|
||||
|
Re: Developers
|
||||
|---|---|---|---|---|
|
Home away from home
|
Quote:
Indeed!
Posted on: 2004/9/11 17:36
|
|||
|
||||
|
Re: Developers
|
||||
|---|---|---|---|---|
|
Home away from home
|
Quote:
I would love to see more people working on AROS and have as much fun doing that as I do. But if this means that I have to program what other people decide I have to program I will fork AROS and probably just work on a private version of the OS. But I don't see why it should be that way. I don't see why somebody couldn't develop a portable binary format for AROS in the way AROS is developed at the moment. What I do fear is that a lot of people try to say to other what they have to do. Quote:
Personally I feel that we need less talk and more programming at the moment. greets, Staf.
Posted on: 2004/9/11 16:12
|
|||
|
||||
|
Re: Developers
|
||||
|---|---|---|---|---|
|
Home away from home
|
well, some of what you say has been discussed already but not perhaps as you are putting it.
I have spoken to Damocles myself a few times about replacing the bounty system/site with something that better reflects whats missing in AROS in general, but also to list "some" possible projects that wouldnt necessarily be part of the main effort. I think having a better list of whats needed would a> encourage more donations because people can actually see what they want, and b> perhaps get a bit more momentum going with regards to what is actually being done. As a developer, i know it can be hard to identify what exactly you could be doing on AROS, other than rakeing through the code actively looking for what needs done. Personally, i think that site would be better for the kind of goals you suggest, while the main AROS goals remain as is, since they are quite definitive as is with regards to what needs done as a whole. (btw are you looking at the old roadmap, or the revised one Tjodden posted on the ML a few months ago?) p.s - i also think the current bounties should be better reviewed by people with a bit more idea about what is necessary/ possible, and "split" into perhaps smaller more managable chunks .... just my opinion though ;)
Posted on: 2004/9/10 19:57
|
|||
|
||||
|
Re: Developers
|
||||
|---|---|---|---|---|
|
Just popping in
 
|
falemagn wrote:
Quote:
No, AROS needs to be made interesting in general. If the team would like to welcome new developers, that is. Quote:
Of course, these goals should be agreed upon. Quote:
So the answer is: "no, we don't want extra developers"? Someone better remove the request on aros.org then :) Quote:
And chaotic. AROS needs direction IMO. First you say that these goals need to be agreed upon, to which I fully agree. But on the other hand, the current developers just randomly add stuff to AROS. How does that rhyme? Is it not possible that developer A does not agree with whatever developer B has added to the codebase? How many of the active developers would actually like to see new developers join the project? If less than six do, then this whole thread is useless. I could understand, but then please stop pretending help is appreciated. Quote:
Yes, I happened to read those pages yesterday, interesting stuff :) <Memory Protection> Quote:
But don't you (and anyone on the AROS team) feel this absolutely has to be a feature AROS should have in the end? Quote:
I guess I'll do that. I'm wondering if it would be possible to agree upon a number of goals. Quote:
I was thinking of interested developers, not end-users. Regards, Brechtw
Posted on: 2004/9/10 19:38
|
|||
|
||||
|
Re: Developers
|
||||
|---|---|---|---|---|
|
Just can't stay away
 
|
Quote:
Well, I do have some ideas, but I don't know when and if I will be able to implement them. I don't even know how feasible they are. That said, these kind of things are openly discussed both on the ML and on the IRC channel, but they're not project's goals, they're personal goals, goals of the specific developers. All the projects goals there are, are explained on the website. There's no more than that. To put it simply: the goal is to recreate a portable operating system compatible with AOS3.1. Beyond that, there's nothing explicitely set in stone, there are only specific developers' ideas and opinions. Quote:
And what if you don't know whether they're unrealistic or not? Quote:
I'm sorry, but I don't agree with that. Quote:
So, you think that if one of the stated goals is "portable binaries" we'll automagically get more developers? I'm afraid it doesn't work like that. Quote:
Because this is not a personal project, where I or someone else can dictate what is AROS to become. The agreed upon goals are only the ones you can see on the website, whatever is beyon that is purely part of one's own personal goals. I'm sure no one will hit me for implementing portable binary support in AROS, nonetheless this is not something everyone in the team agreed upon, and it's not something everyone will work at. It would be just me working on it, and I don't have the resources to afford taking the responsibility for stating that one of our goals is implementing portable binaries. For instance, we could simply port one of the open source implementations of .NET to AROS, so you'd get portable binaries. But then, is that something worth putting in a goals list? Quote:
The "our" part is wrong there. It's not "our" view, it's just "someone's" view. Quote:
Then so be it.  Seriously, we'd not even be able to mamage much more developers than there are now. The whole developent cycle of AROS is based upon the fact that only an handful of developers work on it at any given time, and on distinct parts. It's worked like this for ages, there's no infrastructure to support much more developers.Quote:
I don't know. I think the first problem to solve would be getting ready to gather more manpower. But you'd face hard times trying to achieve that. People work on AROS because it's fun, they work on it in their spare time, as a hobby. No one would agree in a more formal development cycle, it's purely utopistic. Not at this point in time, at least. Quote:
Actually, it wouldn't be the first at all. There's .net, there is tao's vp before .net, and there are tens of other symilar technologies which date well before those 2. And, as said, it's not an agreed upon goals. On the old AROS site there was mention of portable binaries, as possible future improvements, as something to look at, not certainly as a goal. I think it should still be on the site, under the "ideas" section, or something like that. Quote:
There are technical difficulties, but not something unsormontable. Something needs to be changed at the level of the scheduler, and a proper gfx driver needs to be written. Again, this is no more of a goal than "port AROS to xyz": if someone feels like doing it, he/she is free to do so, but no one will force anyone doing anything, and no one will state it as a goal to reach, simply because it's not seen as a goal by everyone, or at least it has never been discussed. Someday it will happen, that's for sure. Quote:
It will pose difficulties. MP is simply not possible with the AmigaOS API, so forget it for the time being. When it will happen, it will be a completely diffetent OS with legacy AOS support. It will probably still be called AROS, but it would be so much different than it is now that it doesn't really make sense putting it in the goals. And we don't know whether that'll ever happen either. Quote:
No, they're not unrealistic at all, but the AROS site is not my personal blog where I write what I want. I should bring these issues up on the mailing list, ask whether people agree to set them as goals, and discuss on their feasibility, how to do them, and things like those. Feel free to bring yourself those topic on the development mailing list: I tbink you are the best person who could make those points. Quote:
Why would the linux community want to use AROS? I myself wouldn't use AROS, I use linux. When I wouldn't use AROS, how could I even dare to propose it to others?
Posted on: 2004/9/10 19:01
|
|||
|
||||
|
Anonymous
|
Re: Developers
|
|||
|---|---|---|---|---|
|
Personally I still think that when it gets a little farther. We will see more interest from the Amiga comunity. How many of them can devlope is a mystery. I think right now a lot of Classing Amiga users and WinUAE users arn't really looking at AROS because it just doesn't do much. When people can download it and do something. IM, maybe simply Web stuff, audio, video in basic form. More people will see it as an alturnative to some of these other things. Maybe, maybe not. Only time will tell.
|
||||
|
||||
|
Re: Developers
|
||||
|---|---|---|---|---|
|
Just popping in
|
falemagn wrote:
Quote:
I'm sorry, I might not have been clear enough in expressing myself (that happens alot :). I was more thinking of a roadmap like in the current "roadmap" on the AROS site. A better term would be "goals". I agree with you that it's generally not a good idea to set deadlines. I would just like to see a list of features that AROS is supposed to have in the future. I'm sure you yourself have some ideas concerning the (possible) future of AROS. Quote:
No, the "how" is unimportant at this stage. Why is it wrong to set some goals that you would like to see achieved? Of course it's useless to set goals that are clearly unrealistic. Every product design starts from a set of goals, which may or may not be achieved in the end. The "how" is a later step in the design process. Quote:
This is a great example of the chicken-and-egg problem. To attract developers, you need to make AROS interesting. But to make AROS interesting, you need more developers. My goal here is, in the first place, to attract developers. The roadmap is just merely tool to achieve this. However, that doesn't mean the roadmap shouldn't be true. Why are you afraid to set some goals? They are just that, goals. Goals are there to strive after. It's never 100% sure you can actually achieve them. You don't have to promise anything. What's wrong with saying: "this is our view of what AROS should look like. We will try to materialise it, but it's not guarantueed"? I think outsiders are simply not interested in AROS because the current roadmap is just plain boring. To them, AROS is just a reimplementation of a "crappy old OS". Quote:
Then let's try to increase that manpower, shall we? You bring up this "lack of developers" argument each time, but isn't this the very problem we're trying to solve? Quote:
Don't get me wrong. I don't want the roadmap to look like it' might have been written by Bill 'Exciting' McEwen. With "ambitious", I do not mean unrealistic or buzzword-filled. It should be a very sober document listing the key features (which are realistic) that would make AROS a unique OS. Let me give some examples: - Platform-independent binaries: AROS could be the first OS to support platform-independent applications. It's a very realistic goal, as the technology is available (even as open source) - Windows-hosted AROS: AROS runs hosted on Linux, why wouldn't it run on Windows? I don't think there are any technical difficulties here. - Memory Protection: if AROS ever wants to become a modern OS, MP is a necessary feature. This should not pose any difficulties either, provided we forget about MP for legacy Amiga applications (but they would still run). Fabio, can you honestly tell me these 3 goals are unrealistic? "there are not enough developers currently" is not a valid reason, as that is something we are trying to solve and should not be considered as a limiting factor. It's a circular problem, as I mentioned higher in this post. To achieve these goals, more developers are needed. That's why you are opposed to setting these goals, I understand. But I think developers aren't interested, because AROS lacks such goals. So I suggest setting some goals, without any deadlines or obligations. Could it hurt? What other options do you see to attract developers? The only audience AROS is reaching at the moment is the Amiga community. No attempts are being made to reach out to, for example, the Linux community. The fact that AROS can run hosted on Linux is a great advantage here IMHO. Regards, Brecht
Posted on: 2004/9/10 0:08
|
|||
|
||||
|
Re: Developers
|
||||
|---|---|---|---|---|
|
Home away from home
|
I totaly agree with Fabio,
AROS only has a few developers ATM (as much a shame as that is) so no matter how much people want something - it takes time, which unfortunately is hard to find. The developers who are involved AFAIK do spend an awefull lot of time working on AROS. The only "roadmap" is to get it working, which is ambitious enough at this level - anything else is detracting from what needs done, And due to the nature (ie a handfull of people working when they can) time-scales aren't something we can set in stone.
Posted on: 2004/9/9 19:33
|
|||
|
||||
|
Re: Developers
|
||||
|---|---|---|---|---|
|
Just can't stay away
|
Quote:
Sure I would. I just said that the current developers are content with the current pace: there's no other pace we can work at. The more active developers the better, of course. Quote:
The copyright on names doesn't apply to them, as far as I know. Anyway, it would be a matter of changing those names and asking the user to create symlinks, should the need arise. But in any case there's "someone" who would have to face the problem way before we do  Quote:
But I wasn't replying to you Quote:
Well, sorry 'bout that... Quote:
Roadmaps aren't any good if you can't stick to them. Look what happened to AmigaOS4 We're simply humble, and conscious of our limitations. Quote:
Those "features" demand a thorough planning and attentive design. As of now, it's not quite certain how and if MP can be implemented. We will never go the AOS4 route, that's for sure, but in any case it's a so big task which needs so many resources which we don't have at the moment. So it's better not put it in any roadmap, simply because we are not sure we can or will achieve it. As for platform-independent binaries, again the same issue: only an handful of people are devoted to this topic in the team. Or rather, perhaps just me. And I don't have time to look at it currently. If and when I'll have time, I'll look at it and implement it. But it won't be in the roadmap because we are not sure it's something will ever happen. Quote:
I'm glad you think that, I appreciate it a lot, but the lack of manpower should bring you back to reality. It's too unrealistic to think that. Quote:
We're not doing "nothing", we're working on AROS, to make it better, whenever we can. And no, I don't believe a fancy roadmap full of buzzwords and very exciting - but which will never be met is the right thing to do. Sorry to have disappointed you
Posted on: 2004/9/9 18:41
|
|||
|
||||
You can view topic.
You cannot start a new topic.
You cannot reply to posts.
You cannot edit your posts.
You cannot delete your posts.
You cannot add new polls.
You cannot vote in polls.
You cannot attach files to posts.
You cannot post without approval.
You cannot use topic type.
You cannot use HTML syntax.
You cannot use signature.
You cannot create PDF files.
You cannot get print page.
