RISC OS cannot be saved by simply open sourcing it, said RISCOS Ltd boss Paul Middleton. He added that Castle’s wish to open source RISC OS “would have no effect on RISCOS Ltd’s products”. Paul told subscribers over the weekend: “Making RISC OS open source, or free as some people want, is not the panacea for suddenly taking RISC OS forward. No operating system can be developed solely for free.”
It may not help (which i doubt btw) but it will surely get riscos some new users and attention. I dont get his point about that they cant earn money when they opensource it, i always thought RiscOS was sold bundled with Hardware?
“To develop RISC OS requires customers, developers and cash. RISCOS Ltd’s customers are the end users. We don’t have a second source of income from major developers who will pay megabucks for specific support that can then subsidise the home users.”
Clearly this guy has no concept of open source.
There was a saying “build it and they will come”, with open source this changes to, ” let them come and build it themselves.”
Open source programmers do it for their own pleasure, as a hobby, etc etc etc
Open sourcing something != automatic development
Never seen dead projects on sourceforge? This is what mr Middleton refers to, and an open-source project would still need to be maintained, probably by someone on a pay-roll.
Of course, the question is how much exactly of ‘free’ development open sourcing would lead to. I think there are enough developpers in RISC OS land to push it forward, so this is a wasted opportunity.
What’s worse is that Castle sort of reached out a hand towards ROL by saying they could/would open source parts of their version with the subtext of improved integration, and ROL slapped that hand away: “he added that Castle’s wish to open source RISC OS would have no effect on RISCOS Ltd’s products”.
I think the concern from ROL is that they want to keep paying the paychecks of their developers. Replacing them with hobbyists is a way to immediately get rid of their jobs, not keep it going.
That’s a valid concern, their product is the OS, and the fear that if they open source it most users will start downloading it for free rather than continuing to buy from them is a valid concern. Open sourcing Netscape may have lead to the production of one of the best web browsers ever, but it certainly didn’t boost Netscape’s profits.
There’s definitely some linguistic trickery, I think he’s definitely saying “RISC OS” when what he should be saying is “my firm.” From the point of view of the RISC OS community, opensourcing the OS may well be a good thing, and everything you say is right. They’re thinking of this from the point of view of a corporation, though, and companies aren’t generally known for intentionally nailing the lids on their own coffins.
“No operating system can be developed solely for free.”
Please someone drop a mail to the LKML, they must be all mad.
Please someone drop a mail to the LKML, they must be all mad.
No, he has a point, at least in the literal sense. In general, the biggest contributions to the Linux kernel come from people that are employed in such a way that a majority of the time for their employer is spent writing those contributions. So, in reality, most of the work isn’t “free.”
We all know that whole Linux thing didn’t work out.
</sarcasm>
GJC
He took care of Linux’s case. Linux’s continuing success is largely due to the contributions made by corporations who have adopted Linux as an OS and now pay people to cut code for it. Like he pointed out, these companies can do this because they’re funding it with separate revenue streams – IBM, for example, ships Linux with its hardware, and the money they spend on Linux development is figured into the hardware pricing. This made great sense for them because they were already paying to develop an entire OS (AIX) and then shipping it with their hardware – by moving to GNU/Linux, they reduced the cost of doing that because they are no longer trying to maintain a Unix all by themselves.
RISCOS Ltd doesn’t sell any hardware that I know of, so they really can’t do the same thing.
…Sure. But keeping it closed-source doesn’t seem to be doing anyone any favours, now does it?
I think Castle’s desire to keep RiscOS closed isn’t a bad one. Open source needs mindshare to work, and frankly, almost nobody is interested in RiscOS. Given the nature of the RiscOS market, a commercial product is probably the only way to keep it developed.
How so? opensource it, and I’ll put money on it, the first move would be to migrate it to x86, that’ll solve the hardware issue in regards to underpowered, over priced hardware.
Then once thas been done, it will encourage those in the userland to port their software once the likes of GTK and Qt are ported; its a snow ball effect.
Its like putting a new product out in a super market, for the first month it might not sell, but once people learn, and get to know the product, there will be a sudden upsurge.
Unlike Linux, RiscOS is a complete operating system that is easy to use, and couple be the panacea to the ‘alternative to Windows on the x86’; and unlike the *NIX world, there isn’t the issue of competing tool kits, or competing ideologies.
Unlike Linux, RiscOS is a complete operating system that is easy to use, and couple be the panacea to the ‘alternative to Windows on the x86’; and unlike the *NIX world, there isn’t the issue of competing tool kits, or competing ideologies.
I’m not gloating, but I’d put money on Debian having more packages available than are available on RISC OS. Linux is in no way “not a complete operating system”, unless you’re talking about the kernel – but then the RISC OS kernel isn’t a complete OS either.
As for ease of use, I can imagine plenty people whining about not having a start menu and the menus not being at the top of the application window. Personally, I wish KDE menus had an option to work like RISC OS menus.
As for ease of use, I can imagine plenty people whining about not having a start menu and the menus not being at the top of the application window. Personally, I wish KDE menus had an option to work like RISC OS menus.
True, but then again, for me, I went through the New Zealand schooling system which was wall to wall Acorn, Amiga and Apple
I was lucky enough to use Archimedes in school for only a few months. As for Amigas in schools, here, in the UK, IF ONLY!
This is RISC OS Ltd – a different company to Castle, which is also a hardware company (and produce the Iyonix, the successor to the Archimedes and RiscPC). ROSL and Castle are the two ROS licensees/ors.
If RISC OS were open-sourced, I for one would sharpen my C skills specifically to port it to x86.
Better hone your ARM assembly skills as well if you want a chance at understanding the source code.
The problem is that RiscOS is mostly outdated and tied to the hardware… Open-sourcing it would change anything to that…
to survive and ported to different HW. Have you looked lately to http://www.linuxdevices.com? Some sell the A9 (without Silicon Motion) at 180$. Why should I spent 600Euros or more for the A9 home to run RISC OS when there is a cheaper alternative?
AFIAK RISCOS runs under ARM/XSCALE only.. So who will benefit from opensourcing RISCOS? Maybe this will kill completely the R/OS… Developers need real machines – now they are TOO expensive..
If only the (few) companies that build Xscale/ARM h/dare could find an alternative way to reduce the cost, say sell only barebone (m/boad-cpu only NOT complete systems as IYONIX does..) systems..
You’re right: it does only run on ARM processors. There have been suggestions that open sourcing RISC OS might improve the ability of hardware manufacturers to promote it as a set top box platform, so maybe this would be the benefit? However, unlike the hardware manufacturers, RISC OS Ltd doesn’t have this revenue stream as they’re only allowed to develop desktop RISC OS.
It’s interesting that you suggest Castle should offer a motherboard only version of the Iyonix. Very many people requested this (especially people living abroad I believe, to cut down on delivery costs), and eventually Castle did offer this option for a limited time: http://www.iyonix.com/iyonix/news/diy.shtml
However, the price of a complete system has now come down to almost the same price as these were sold at anyway. At any rate, the A9Home is now also much cheaper. http://www.advantage6.com/products/A9home.html
To be honest, whilst I don’t disagree that the hardware is expensive compared to PCs, I’d be surprised if a cheaper RO+ARM machine could really be released. As I understand it (and it does seem plausible to me) much of the expense comes from development costs on relatively small numbers of machines (compared to your average PC). Developers who don’t have the cash can already use an emulated version of course: http://www.virtualacorn.co.uk/index2.htm (although personally, I don’t find this as much fun!)
I asked this summer castle, and got a query about the price of a “barebones” system – about 400 + something GBP , I don’t remember the amount..they wrote me it was a special offer..
Of course the price was right in comparison to “full” IYONIX systeme, but I did avoid to do this..
Having a system with limited hardware compatibilty and software similarity in comparisson to other OSs is a unpleasant situation (I faced the same at the past using my Archimedes A5000)
Looks like you have much more up to date info about this than I do in that case. Thanks for the info; 400 GBP actually does seem like a good deal compared to a complete system, but as you say, it’s still a lot of money, and expensive compared to x86 PCs.
I have to agree the situation of having to use expensive and proprietary hardware is not ideal from a user perspective. However, it’s also one of the reasons I personally like using RISC OS. (Not the fact it’s expensive and proprietary of course, but because ARM processors are an interesting and nice design).