[Jezz_X]

And back to the original question

It really doesn't matter if its "good or bad for the XBMC development process" its what the people who give their free time for no pay want to do.

And the day you start forcing people to work on stuff in their free time they don't want to do is the day they all start leaving so in the end it wont be any different

[Robotica]

And back to the original question

It really doesn't matter if its "good or bad for the XBMC development process" its what the people who give their free time for no pay want to do.

And the day you start forcing people to work on stuff in their free time they don't want to do is the day they all start leaving so in the end it wont be any different

This is not the original question; this is you opinion about the freely donated efforts from the devs. This has nothing to do with the multi-platform discussion. And how can you talk about forcing devs when XBMC is GPL and non-commercial?

The original question is if those efforts would result in more/better features if the focus is on less platforms. TOPfs threw some arguments that that´s not the case.

[Jezz_X]

Right and my answer is they would probably leave anyway and you wouldn't get anymore work done on the one platform because it would still be the same amount of devs doing it that there is today.

And Again I was saying that you can't force them so the whole topic is pointless because its not going to happen

[topfs2]

Say we would focus on windows, I would surely leave. Not meaning to pick on people using windows but I have no reason for xbmc on windows. Heck, even if it was only on linux (which I use) I might leave. My main motivation when I began coding for xbmc (and dropped my own project) was that a multi platform htpc software was needed, so you can choose the OS as you wish but have the same interface with the same features. i.e. the OS should not matter for the application.

[Robotica]

funny... As a regular user, I chose XBMC for it´s features, interface (which is also a feature) and active community. I think, the success of the community (and thus the features) is partly due to being multi-platform and the architecture. This resulted in a great XBMC GPL-software product.

But I rather don´t deal with the OS, as long as everything keeps running. Since Linux offers great and free specialized builds (remeber the successtory of linux-based NAS devices?), I chose Linux as OS. It also saved me around 100 bucks compared to Windows7 or OSX. Generalizing, we can conclude devs motivation and users needs are contradicting.

But what I don´t understand is that devs care about more then one platform. It´s not like your have a multiboot device under your tv just for the fun. At the tv you rather use your HTPC...

[topfs2]

You hit on the point there, you'd rather not deal with the OS. I consider you being fully able to choose the platform as such. What if you had to use frebsd to use XBMC, then you'd have to learn freebsd just to use xbmc, makes no sense.

Besides. 90% is easily platform agnostic in xbmc and have always been. Even 90% of the features are platform agnostic so the whole extra resources for doing it multiplatform are probably just a few % but what we gain are multiple of extra devs. And being multiplatform gets more buzz which in turn even get more devs and resources. I really believe that the resources needed for xbmc being platform is say 50% but what we gain by being it is perhaps 500% of resources, thus we are actually progressing faster.

[rausch101]

@Robotica

Trying to make the argument that it would be a good idea to regress and go back to a single-platform project, which it appears is essentially what you are saying, is an awful idea. XBMC started out as a single platform on XBOX. It was the natural and correct progression to move to multiple platforms. Even if you say that's an incorrect assumption I've made about what you're saying, there's still glaringly obvious reasons not to move to a single platform.

The most obvious reason to move to (or stay at, in this case) a multi-platform development is to increase the potential users. For a community-driven project like this, more users means more contributions. PLENTY of XBMC components (like Addons, skins, web interface, scrapers, etc.) are inherently platform agnostic and so the benefit from people in the community who only work on one platform apply to all of the other platforms. Someone who uses a Windows machine makes an addon, and then it gets applied to all platforms. This addon wouldn't even have existed if XBMC wasn't on Windows because the addon creator wouldn't have XBMC installed. No one would deny that if XBMC was only for one platform, it'd probably have fewer bugs, some more fleshed out features, but the overriding downside is that it'd have FAR less features and FAR less reach.

That's all completely aside from the fact that the devs can do whatever they want to do in their free time. It just so happens that they're smart and their project roadmap/development process is spot on.


EDIT: Well, it looks like topfs2 was typing pretty much the same thing anyway...

[topfs2]

EDIT: Well, it looks like topfs2 was typing pretty much the same thing anyway...

Yup, I'm the fastest gun in the west! I think you explained it far better than me though

[Robotica]

Interesting view.

You hit on the point there, you'd rather not deal with the OS. I consider you being fully able to choose the platform as such. What if you had to use frebsd to use XBMC, then you'd have to learn freebsd just to use xbmc, makes no sense.

I would not choose XBMC as my mediaplayer for my HTPC. No users = no devs

Besides. 90% is easily platform agnostic in xbmc and have always been. Even 90% of the features are platform agnostic so the whole extra resources for doing it multiplatform are probably just a few % but what we gain are multiple of extra devs. And being multiplatform gets more buzz which in turn even get more devs and resources. I really believe that the resources needed for xbmc being platform is say 50% but what we gain by being it is perhaps 500% of resources, thus we are actually progressing faster.

Most devs XBMC gains are Python coders, regex devs, and skinners. Occasionally, a agnostic C-programmer. This and users create the buzz for 100%. I think it would draw a better picture about communities´ attitude towards the multi-platform discussion in relation to their freely donated efforts if some of the Python coders and skinners can give their motivation. Or a guy like jfcaroll, who is creating an agnostic Addon Manager.

@rausch101:
I was informing about this subject and then started to discuss. That´s what this forum is all about. But don´t get me wrong. topfs2 already had some good arguments earlier so please don´t suggest that I already draw my conclusion that one platform is better. Now you stress again about the devs free time, it comes to my mind that devs need to manage many platforms. All that time could also be spent on features and running a Foundation to protect the efforts.

[rausch101]

Now you stress again about the devs free time, it comes to my mind that devs need to manage many platforms. All that time could also be spent on features and running a Foundation to protect the efforts.

I'm not a dev, I'm under no stress.

Yes all that time dedicated to developing on multiple platforms could be spent on more features on a single platform (or the Foundation?) as I said. OR, since they've moved to a multi-platform project, they can just have all the new developers and users (that otherwise wouldn't have existed if XBMC had still been single platform) make and test new features for them. This is the same analogy as population growth in a blossoming civilization. Yeah, you have overhead and other things to worry about as the population grows, but more gets done with more people.

[Robotica]

Yes all that time dedicated to developing on multiple platforms could be spent on more features on a single platform (or the Foundation?) as I said. OR, since they've moved to a multi-platform project, they can just have all the new developers and users (that otherwise wouldn't have existed if XBMC had still been single platform) make and test new features for them.

Yes, overhead is a good way of calling that non-dev stuff about managing this software project. But most attracted devs are python, regex and skinners and they have nothing to do with the agnostic platform besides creating products for a broader audience. But those guys might as well go to another, newer, better platform if available. Maybe Boxee will become a great success by creating a great eco-system were those devs can even make a buck. (yes, I know that is not the motivation but no one would say no to some extra bucks)

This is the same analogy as population growth in a blossoming civilization. Yeah, you have overhead and other things to worry about as the population grows, but more gets done with more people.

This is so true. So the actual question is: What is extracting all those devs? Only the forum to find out... This is the reason I asked for some other users/devs opinions.

[topfs2]

So for you FreeBSD makes no sense, for some it does. Why choose a platform when you can work on both and please both communities.

How to choose the platform? Linux has the powers of being opensource, works perfect on embedded target much better for embedded. Windows has more potential desktop users. Can't really choose one of them, so better support both. Then you might aswell add the others.

I'm going to paint a picture aswell. Say that windows and linux users are equally many, and potential developers aswell.
The extra amount of resources to handle both platforms is probably 10% but we just doubled our users and thus word of mouth and marketing. We have probably doubled our developers aswell (perhaps even more since we have even more PR), this means that for 10% more resources taken you gained 100%. This means you have 90% more resources for platform agnostic features, i.e. you win on having multiple platforms.

All in all, there really are no arguments that makes single platform better than multiple for xbmc. I agree that some stuff might be better ironed out if it was single platform but overall it would criple the application if it where single platform.

[Robotica]

Boxee neither Plex will probably become successful but Google TV (which is not multi platform but only Linux) will. The same reasons why Android will beat Iphone:
- Great Eco-system (for content providers, hardware providers, devs, etc)
- Great partners to push technology to users.

For both, users and devs, GoogleTV will be a very interesting platform. So don´t draw conclusion too fast. topfs, I really admire all the work you have done but please stay open minded to not so obvious matters.it´s important to have a clear vision on this stuff/overhead.

I think the GPL-spirit (thank you all XBMC devs!!) will make sure that users get as much possible features for free. This doesn´t necessarily means winning the users, albeit a big advantage. Features, availability, prices, content and partners will win the users and devs (since those are tied together in Open Source software). I don´t believe multi-platform is a key succesfactor in all this; but it´s great to have and if the resources are available keep it forever.

I do not really share the idea that most part of current success is because being multi platform. It already was quite successful on XBOX. Since being multi platform, XBMC became even more successful. I think this success began because XBOX development forced having an agnostic and very creative architecture. But like I said, most attracted devs are xml, json, skinner, regex and python devs; not necessarily agnostic C-devs. I think most of those guys prefer Linux since they can do the same and save 100 bucks for OS.

[rausch101]

This is so true. So the actual question is: What is extracting all those devs? Only the forum to find out... This is the reason I asked for some other users/devs opinions.

I think you completely missed my main point.

Boxee neither Plex will probably become successful but Google TV (which is not multi platform but only Linux) will. The same reasons why Android will beat Iphone:
- Great Eco-system (for content providers, hardware providers, devs, etc)
- Great partners to push technology to users.

Though I'm not sure I agree with your prediction, I'm glad you mentioned Google since that's great evidence of the multi-platform approach working.

The reason Android has been so sucessful and widespread is because Android was created to be on multiple platforms, (platforms in this case being hardware and carrier, rather than OS), whereas the iPhone has just been on one manufacturer and carrier (though not for long). Android has been written for both ARM and x86 architectures as further proof of their inclination to integrate into more platforms. Google's motto on a large variety of applications is precisely multi-platform for the specific reason of appealing to the most number of people possible. No one is going to deny Google's success, and if they're going the multi-platform route then.........

Even though Android deals with multi-platform hardware and we're talking about XBMC's multi-platform software, the principal is one in the same. Multi-platform Android deals with increased user-base to the end of increased revenue while XBMC's multi-platform deals with increased user-base as the end, but the principal is again the same.

The multi-platform approach is a logical progression of any technical project wanting to grow and progress. It makes sense. This is how Windows became successful. This is how Android became successful. This is why "Apps" are written for Android, iOS, Blackberry, etc. and why there's a MAC, PC and Linux version of everything.

For both, users and devs, GoogleTV will be a very interesting platform. So don´t draw conclusion too fast; it´s important to have a clear vision on this stuff/overhead.

The devs have a clear vision. Many of them work full-time in related fields. I absolutely guarantee their project roadmap was very, very well thought out.

I'll try to make this my last post about this

[topfs2]

This will be the last I write since it feels abit like you don't listen to me.

GoogleTV is a platform or a framework (depending on what you define it as) not an application (which xbmc is). IF you are going to compare to GoogleTV you need to compare it against XBMC Live, which is the only thing close to a platform with xbmc. It is based on linux and always will, it would never make sense basing a platform on windows... It might on freebsd but linux is just more mature.

And as stated multiple times here, the developers we gain from being multi platform are both C devs and addon creators. The addon creators don't need to choose their platform they just use whatever they like and all other platforms benefit. This means we would gain addon creators from windows, linux and os x and the addons they create can be benefited from all users no matter their platform. The C devs usually have one selected OS they use and contribute to and thus they are the ones that stands for the extra resources needed to support the multiple platforms.