Riscos ?


sebt3

homebrew player (P. & C.)
Joined
Sep 9, 2008
Messages
4,885
Age
44
Location
France
Website
sebt3.openpandora.org
One hop to the big website.
http://www.osnews.com/comments/22096
and one hop to the "proof" :
http://www.powerdeveloper.org/forums/viewtopic.php?p=12655#12655
 
sebt3 said:
One hop to the big website.
http://www.osnews.com/comments/22096
and one hop to the "proof" :
http://www.powerdeveloper.org/forums/viewtopic.php?p=12655#12655
I think I will ask the same question that the Fedora feature reviewers ask for feature submission:
How does the OpenPandora project benefit from this?

I really want to know; how would we use this OS on the Pandora?
 
Last edited by a moderator:
dflemstr said:
I really want to know; how would we use this OS on the Pandora?

Glad you asked.
Ya see, many game consoles benefit from a single, unified hardware / OS platform.

But by running a fractured, unholy conglomeration of:

Angstrom
ReactOS
Haiku
Gobolinux
Ubuntu
Maemo
Mer
RiscOS
Gentoo
Fedora
Android

We will be able to keep the developers occupied with packaging and compatibility issues, which will prevent them from actually producing any software.
 
Last edited by a moderator:
lulzfish said:
We will be able to keep the developers occupied with packaging and compatibility issues, which will prevent them from actually producing any software.
Oh, I see... Now, does this only consume the time that would otherwise be used for software production, or can we count on even more free time being consumed? Also, how do we start the time-black-hole to begin with; if developers don't have any software to distribute, they probably won't do it, correct? So there has to be some grace period before the operating systems are distributed for any kind of counter-productivity-system to establish itself, right?

And how do we deal with the portable PND system that probably will be used on all platforms? It might reduce the required time for packaging by up to 90%, which is something that we clearly don't want. It will of course only be usable on Linux-based systems, so Haiku etc will need specialized packages, but the amount of time saved is still big enough to be significant.

We should creaet some counter-productivity systems before all hell breaks lose. Perhaps inserting random "sleep 10"'s in rpmbuild could help?
 
Last edited by a moderator:
Really, we're all aware of what the default OS is for Pandora. While other distros/OS solutions will be made available they're never realistically going to get the support of the built-in OS.

Developers will develop for the built-in OS and nobody should expect them to develop for alternate solutions, unless they have any particular interest in doing so.

Alternative OSes on Pandora will only ever be a hobbiest activity. That being said, I'm looking forward to playing with Haiku and AROS if/when they get to any kind of usable state - but purely out of curiosity. I'll boot straight back into Pandaora's built-in OS whenever I actually want to do something other than 'play' with an OS.
 
lulzfish said:
But by running a fractured, unholy conglomeration of:

Angstrom
ReactOS
Haiku
Gobolinux
Ubuntu
Maemo
Mer
RiscOS
Gentoo
Fedora
Android

We will be able to keep the developers occupied with packaging and compatibility issues, which will prevent them from actually producing any software.
Absolutely correct.

So lets just settle on Angstrom. That can be our unified platform/OS. ;) And since the driver support is superior to RiscOS at the current time, it seems like the obvious way to go!
 
Last edited by a moderator:
dflemstr said:
I think I will ask the same question that the Fedora feature reviewers ask for feature submission:
How does the OpenPandora project benefit from this?

I really want to know; how would we use this OS on the Pandora?
We (the people on this forum, already waiting for the Pandora to ship) will not benefit at all from this, but surely that's not the point? A RiscOS port would open the Pandora up to a whole new market (a market currently starved for compatible hardware) which has to be a good thing for the OpenPandora team. New markets = more sales which is good for us all in the long run.

I really don't buy into the fragmentation claim. OpenPandora are providing and maintaining a standard, supported core OS for us all to use and hack onto, most people will want to customise this with different skins/launchers/window managers, but this doesn't affect biranry compatibility (using packages PND's) at all. Anyone who chooses a different OS will understand what they are doing enough to realise what this means. Most alternatives are Linux based anyway which means that they will probably be reasonably binary compatible.

Other alternatives (like RiscOS) will be used by people buying the Pandora specifically for that OS (or maybe multibooting if possible) and would probably never have bought a Pandora otherwise, therefore they are not fragmenting the standard market, they are just a niche set of customers boosting sales.

Personally, I'll be sticking to Angstrom for quite a while but if the Amiga oriented OS's (AROS or Anubis) get useful I'll be using them too, probably booing from SD so that I don't affect my core, supported, Angstrom distro. Since Anubis is planned to be Linux based I'll be looking to help out getting it fully running on the Pandora hardware and maybe adding in some PND support, but I see this as expanding the Pandora into new markets, not fragmenting the existing one.
 
Last edited by a moderator:
Lozrus said:
Other alternatives (like RiscOS) will be used by people buying the Pandora specifically for that OS (or maybe multibooting if possible) and would probably never have bought a Pandora otherwise, therefore they are not fragmenting the standard market, they are just a niche set of customers boosting sales.

Personally, I'll be sticking to Angstrom for quite a while but if the Amiga oriented OS's (AROS or Anubis) get useful I'll be using them too, probably booing from SD so that I don't affect my core, supported, Angstrom distro. Since Anubis is planned to be Linux based I'll be looking to help out getting it fully running on the Pandora hardware and maybe adding in some PND support, but I see this as expanding the Pandora into new markets, not fragmenting the existing one.
Now let's face it: The time it would take (maybe 100+ manhours) to port a completely new OS to the Pandora wouldn't really be "paid off" by the maybe... 5 people that would buyt he Pandora because of RiscOS support. It would be an interesting project, but very unpractical.

The other OSs are different: There are already Gentoo, Ubuntu, Fedora, etc ports to ARM so the amount of work required would be minimal. Also, those OSes have huge application databases which means that they won't be completely worthless.
Even Haiku (which, btw, is a BeOS remake that is awesomely fast and has some awesome features that no other OS has) has been ported to ARM already (It runs on Gumstix etc) so a recompile is theoretically all that's needed to get it to run, and it's built using gcc so you could even make Linux-applications run on it by introducing a new parallel ABI (Yeah, Haiku is flexible like that).

BTW, you can multiboot with the Pandora (From NAND, SD cards and via Netboot IIRC), no problem.
 
Last edited by a moderator:
dflemstr said:
Now let's face it: The time it would take (maybe 100+ manhours) to port a completely new OS to the Pandora wouldn't really be "paid off" by the maybe... 5 people that would buyt he Pandora because of RiscOS support. It would be an interesting project, but very unpractical.
When you consider that the port is done entirely by a third party at no cost to the Pandora team, even one sale because of it makes the payoff infinite from a mathematical perspective.
 
Last edited by a moderator:
WizardStan said:
even one sale because of it makes the payoff infinite from a mathematical perspective.
You could discuss endlessly about that point :P Division by zero is nasty.

Also, I was more considering the porter's POV. The porter will spend lots of time on the project, and the only payoff he would normally get (if it had been a game that was ported or something) would be the community's appreciation, the fact that the Pandora platform becomes more popular (→ more sales → more economical freedom for the OP team so that they can produce cool products), and the feeling of having helped someone.

In the case of RiscOS, however, only a very few people are going to appreciate the port and that might demotivate someone intending to port the OS by a fair bit.

Or are you assuming that RISC OS Ltd. or Genesi will become interested enough in the Pandora that they will sponsor the port? As it is now, Genesi is interested in porting the OS for their own products, and a significant amount of work would be needed to adapt the port to the Pandora platform.
 
Last edited by a moderator:
I don't think the port to the Beagleboard is being done because of the number of hardware units it'll sell. It's being done because the Beagleboard is there, and because it's interesting work for the coder(s) and because it helps make RISC OS more portable the more platforms it's ported to.

If as a side effect you get another 50 sales for zero effort then that's a nice bonus. RISC OS has its own community of developers and users to do the work and to use it. It'd just be nice if everyone got along. The RISC OS community can worry about Jeffrey's sanity ;)
 
dflemstr said:
In the case of RiscOS, however, only a very few people are going to appreciate the port and that might demotivate someone intending to port the OS by a fair bit.
I'm thinking that a person doing a port should do it because they want it, and not because they'll get any real recognition for it. If others take the port and start using it, that's all bonus: the person doing the port has already received their reward of being able to use their preferred platform in the way they want to.
 
Last edited by a moderator:
WizardStan said:
dflemstr said:
In the case of RiscOS, however, only a very few people are going to appreciate the port and that might demotivate someone intending to port the OS by a fair bit.
I'm thinking that a person doing a port should do it because they want it, and not because they'll get any real recognition for it. If others take the port and start using it, that's all bonus: the person doing the port has already received their reward of being able to use their preferred platform in the way they want to.
I agree, but such a person has yet to present him/herself. There have only been posts about "Wouldn't it be cool if..." and no posts about "I'm totally going to port...". If there had been a programmatically inclined person around that would have uttered his/her intent to port RiscOS, well, then I don't see where the problem is and my previous post would have looked very differently because of that.
 
Last edited by a moderator:
WizardStan said:
dflemstr said:
In the case of RiscOS, however, only a very few people are going to appreciate the port and that might demotivate someone intending to port the OS by a fair bit.
I'm thinking that a person doing a port should do it because they want it, and not because they'll get any real recognition for it. If others take the port and start using it, that's all bonus: the person doing the port has already received their reward of being able to use their preferred platform in the way they want to.

And gaining the knowledge while porting it.

Trooper
 
Last edited by a moderator:
...of course any non-linux based OS will not be able to make use of any hardware graphics acceleration due to the driver issues.
 
dflemstr said:
WizardStan said:
I'm thinking that a person doing a port should do it because they want it, and not because they'll get any real recognition for it. If others take the port and start using it, that's all bonus: the person doing the port has already received their reward of being able to use their preferred platform in the way they want to.
I agree, but such a person has yet to present him/herself. There have only been posts about "Wouldn't it be cool if..." and no posts about "I'm totally going to port...". If there had been a programmatically inclined person around that would have uttered his/her intent to port RiscOS, well, then I don't see where the problem is and my previous post would have looked very differently because of that.
Any RiscOS port will come only if it's actually wanted and someone puts the effort in. I don't think anyone has been petitioning the OpenPandora team or devs here to do it. As I said before, I think RiscOS on Pandora would be something for tha RiscOS community rather than the current Pandora community and so any port would probably come from them too (if the Pandora looks like worthwhile hardware to someone who has the skill and inclination).

I don't know much about RiscOS, but I understand they have a reasonable sized community, a shortage of available hardware and are used to running on far slower processors than the Cortex A8. They have also achieved some impressive results like getting a working port of Firefox, which the Amiga community still haven't managed (despite a huge bounty).
 
Last edited by a moderator:
dflemstr said:
The other OSs are different: There are already Gentoo, Ubuntu, Fedora, etc ports to ARM so the amount of work required would be minimal.
RiscOS is an ARM specific OS. In fact the original ARM processors were developed by Acorn specifically to run RiscOS.
 
Last edited by a moderator:
My bad - I didn't read the links in the opening post and assumed details of the BB port of RISC OS were in there. The BB port is already underway, if you see here: http://www.riscosopen.org/forum/forums/5/topics/166

Indeed, there seems to be some overlap between the Pandora and RISC OS communities - at least three sales already by my counting! Last time I looked, the port had got as far as letting the OS boot to the GUI, and efforts were moving on to getting USB working so actual software could get loaded on to it. As you can see, work has been ongoing since January, though it's only targetting the BB for now and not Pandora specifically.
 
Back
Top