Castle director and Tematic boss Peter Wild has said RISC OS must become open source to compete in modern markets. “”Make [RISC OS] open source and encourage users to contribute to it, use it and drive its ongoing development. This is the only scenario I can see in which the solution might prosper – short of someone investing a few million quid to fund further development.”
It’s the fact that RISC OS runs on ARM processors that’s more key to its future than open sourcing the code. Open sourcing it would help other people try it – but then in the end people end up comparing it modern day operating systems (like the summary) and it doesn’t compare.
RISC OS isn’t a modern OS, it’s not trying to be. It’s more of an embedded system. Small, lightweight, focused.
Just my opinion.
You’re right..the fact that it runs on ARM processors is key; both towards it’s appeal, and how an open sourced OS wouldn’t immediatly mean clones all over the place and people freeloading. Opensourcing the OS would mean the potential of more developpers enhancing it, but it could also mean the dedicated programmers currently employed as devs could become less productive, as they would have to streamline an OSS proces instead of programming themselves, for a potentially small programmer audience.
Opensourcing the OS would also mean two other things:
-a handful of people would start running RISC OS on their PDAs
-it would make it a lot easier for a new company to start producing computers running RISC OS.
And yet, still, if Castle and the other company go bankrupt, there won’t be anyone left making ARM computers except for PDAs, so that would be the end of RISC OS still.
And yet, still, if Castle and the other company go bankrupt, there won’t be anyone left making ARM computers except for PDAs
I guess you meant “ARM computers runing RISC OS”? Because there are plenty of cellphones, wireless access ports, and network attached storage devices running on ARM processors.
I don’t think a phone, an access point or a NAS device could be really described as ‘computer’
If a PDA is a “computer”, why isn’t a Treo 6xx or 7xx?
People who used to balance the books were called ‘computers’, what’s your point? If it’s electronic and can calculate even one sum, it’s a computer.
If it’s electronic and can calculate even one sum, it’s a computer.
If I had said ‘personal computer’ instead of just ‘computer’, even more confusion would’ve ensued.
Either way, I meant the kind that stands on or below your desk, and runs desktop/CLI applications.
If I had said ‘personal computer’ instead of just ‘computer’, even more confusion would’ve ensued.
Either way, I meant the kind that stands on or below your desk, and runs desktop/CLI applications.
I have one of these
http://www.embeddedarm.com/epc/ts7200-spec-h.html
standing on my desk running netbsd right now.
At home I have a slug
http://www.nslu2-linux.org/
running Linux
These machines are both faster than the VAX 11/780 and the PDP 11/70 that were the first machines I ran Unix on, and neither is more powerful that then Xscale processor in the TREO 650, which is capable of running Linux as well.
I think we have outgrown your distinctions.
Why do you call it freeloading? It is just exercising the rights given to you by one of the OSI compatible licenses. That is in no way freeloading.
Also i dont see how opensourcing it could be bad for Castle, since they could continue to sell their pc’s with RISC OS (another right given to you by OSS licenses, namely to charge for the software), and the development or RISC OS would go much faster! Neither do i see how the OSS software development process could slow development down. Look at for example Microsoft which uses the current RISC OS development model (proprietary), and yet Vista is still years late.
Opensourcing RISC OS would be a win-win situation, both for RISC OS and the opensource community.
Why do you call it freeloading? It is just exercising the rights given to you by one of the OSI compatible licenses. That is in no way freeloading.
By freeloading I meant people taking the effort of Castle ltd/ROS, slapping it onto something else, meaning Castle ltd/ROS get no return on their investments. Which would mean no incentive to do active development.
Neither do i see how the OSS software development process could slow development down.
If they currently have 1 or 2 programmers developping the OS (seems accurate, based on linked articles), and one person is needed to ‘interface’ with the audience, set up access to a CVS, talk to interested developpers..and if no actual developments come from it (since skilled kernelhackers are few and far between), then development would slow down.
Pointing at Vista is a bad example; there’s more wrong with Microsoft than being closed source. As a counter-example I could point to Unununium; an open-source OS with no visible development for the past months.
“By freeloading I meant people taking the effort of Castle ltd/ROS, slapping it onto something else, meaning Castle ltd/ROS get no return on their investments. Which would mean no incentive to do active development.”
By freeloading, you presumably believe that the only licences of an open source nature are permissive ones. Consider for a moment the non-permissive ones, used successfully by a number of companies (including MySQL, but I don’t hold them up as a shining example of anything in particular): the copyright holders could offer RISC OS under the GPL and a commercial licence, reserving things like trademarks for brand identity purposes – think Red Hat vs. Fedora!
As for the pace of development, and with no offense intended to the small development staff of the RISC OS commercial scene, there are probably more qualified people out there who already know the code than there are working on it right now. Even if these people start competing with the current custodians, if the pie grows fivefold and Castle’s share is halved, everyone still wins.
Any software on the market today must open Source or die. We are seeing the slow rotting death of Windows, We have seen the death of SGI, all proprietary companies must either open source its code or it will displaced by more innovastive and open source code. The days of proprietary code are gone. Open Source is the only future. How do they say it, Resistance is Futile.
Must dissagree with you. I’d say we are seeing a maturing coexistance between opensource and closed.
Business’s like IBM do very well in both. Hopefully so can Novell & others.
I’m sure customers of QNX are quite happy to pay top $ for a quality product with complete support.
The rate of change and improvement and evolution in open source is so great that the only reason OSS and proprietary software “coexist” is because there are so many proprietary software programs and companies. Not even OSS software can or could obliterate them in a day.
IMHO Pegasos users (as me) I think they will welcome Risc OS for their machines.
The pegasos II:
http://www.pegasosppc.com/pegasos.php
Risc OS is a good small footprint OS with good features and it evolved from the first Acorn-Archimedes Risc OS to became more mature and usable.
(A little example on how it evolved thru times:
If I remember well first versions of Risc OS were only co-operative multitasking. Now it achieved also Preemptive multitasking. A great advance.)
Well maybe it could be ported to PPC if the source became available (although it may be written substantially in ARM assembly which would make porting hard) but as it is it’s useless on a Pegasos…
“If I remember well first versions of Risc OS were only co-operative multitasking. Now it achieved also Preemptive multitasking.”
No it didn’t. There was a pre-emptive multitasking extension, but it was a dirty hack on top of the window manager. RISC OS is still a co-operative multitasking microcomputer operating system, even though it was running on workstation-level hardware when first introduced.
—
No it didn’t. There was a pre-emptive multitasking extension, but it was a dirty hack on top of the window manager. RISC OS is still a co-operative multitasking microcomputer operating system, even though it was running on workstation-level hardware when first introduced.
—
Well, it is a pity, it doesn’t have, but even co-operative multitasking is good if the OS works fine.
This however ties all system with more limits of usage.
It is frequently mandated by telcos and service providers in their specifications for STBs [that the operating system] is open source and the Telco will have access to a system it can work with an modify independently of the hardware manufacturer. In emerging markets without clearly defined standards this is considered essential for future proofing.
If telcos and service providers need to be able to modify the operating system, you can always offer them sourcecode access on a case by case basis. There is absolutely no need to make the whole operating system open source. Customer specifications should not always be taken literally. It is important to ask for the motive behind each specification and understand what the customer wants to achieve.
Of course, if your operating system is offering absolutely no real-world advantage compared to free open source solutions, your best shot is to go open source too and pray for a miracle… But I would think RISC OS is actually better than that. With no private equity available, that probably does not matter however.