RISC OS is open for business!

See our announcement here.

Great news!

The old licence had a “feed back changes after x months” clause, with various values of x depending on how much the commercial developer paid in fees. Nothing like that in the new licence. Does that mean if a commercial developer writes a driver for some new feature in a hypothetical new machine, they can keep their new toy (aka the driver source) for themselves?

“feed back changes after x months”

Good point.

Is the entire codebase being relicensed? What about the bsd/CDDL parts?

Why Apache instead of CDDL? Apache includes a one way street to both GPL land and closed source derivatives because the licence essentially boils down to “do whatever the hell you like, just don’t sue me”, whereas CDDL at least requires CDDL code to remain CDDL (but unlike GPL it plays nice with others). Shouldn’t an open source project at least use a licence that keeps it open source?

I am very happy with that, because using a well-known and very liberal licence simplifies explanations to non-RISC OS-knowing people a lot. It also offers new possibilities of using other Open Source licences for combined works without having to ask your friendly lawyer about the consequences.

Unfortunately, it also opens up a lot of potential for future fragmentation, because it allows closed-source products using RISC OS. We’ll see how that will turn out, and I hope sincerely that all parties will keep going along the “one source – one RISC OS from RISC OS Open” way.

Not getting any reply on https://www.riscosdev.com/ – is there anything there yet?

You’re right that fragmentation and non-return of source are the biggest issues we face, and the aspects of Open Source that I’m personally most concerned about. We’re actively looking at ways to encourage co-operation and feeding back source, but would also strongly welcome input from any forum members with experience of this sort of thing.

The last thing we want to see is “R-Comp-OS”, “CJE-OS” and “Elesar-OS” etc. From an RCI perspective, we intend to operate under similar feedback as previously under Castle, as we value the centralised, community work that has been possible thus far. I hope that the other companies will follow our lead. Consider the guantlet thrown down! :)

We’re actively looking at ways to encourage co-operation and feeding back source, but would also strongly welcome input from any forum members with experience of this sort of thing.

The only thing that has worked in the past (to my knowledge) with many Open Source projects is a community that values Open Source and pressures the players to behave.

I.E. if someone provides RISC OS but does not provide a fully buildable source base to go (without a very good reason – it is e.g. OK with me to provide commercial, closed-source add-ons just like it is OK to provide commercial applications) with it should be met with extreme sceptism and the community should largely refuse to do business with players who do not embrace the Open Source idea.

RISC OS is unfortunately in a very weak position to benefit from typical Open Source-big business-sponsor-/partnership (either monetary or man-power), so I am not at all sure how such a scheme will work out – much of the Open Source stuff only exists because corporations around the world have big corporate customers who pay for support. Can’t see much of that happen in RISC OS world where most customers are end-users.

This is all about removing barriers to adoption and contribution for RISC OS. Any potential outcomes around fragmentation or closed source forks would, to be honest, be a nice problem to have, IMO! Anyway, I personally see those as the extremes – there are a lot of significantly more positive outcomes that we should be looking forward to.

By the way, if anyone spots anything in the media picking up on this announcement, please post here so we can all enjoy!

Yes, thanks Steve. It’s worth mentioning that RISC OS Developments have worked with a media PR person to seed the news far and wide, from tech sites, to BBC, to the retro scene. We figure that we only get one shot at this, so we better make it count :)

Anyway, I personally see those as the extremes

I suppose we don’t need to worry too much about the sort of Chinese shovelware outfit that regularly abuses GPL (Linux inside).
They’ll take one look at the huge swathes of hand crafted assembler and back away with a flat “What.”.

I suppose we don’t need to worry too much about the sort of Chinese shovelware outfit that regularly abuses GPL (Linux inside).

Never really looked at that kind of thing. Do they publish their source or just concoct a binary and bung that out?

By the way, if anyone spots anything in the media picking up on this announcement,

Tom Holwerda of OS NEWS has always been sympathetic to RISC OS in the past.

Good news to hear,

Steve: http://linux-sunxi.org/GPL_Violations Allwinner is a serial infringer, it would seem… But, then, think how many devices you have around that probably run some form of Linux but don’t come with sources…

In this respect, I wonder if https://opensource.orange.com/fr/software/logiciels-de-la-maison/livebox/ contains everything necessary to build the Livebox firmware, or if it’s only the necessary stuff (that had to be released). Downloading what I think is my version now… [Livebox 2 Sagem]

Okay, I don’t see anything to do with either SIP or LivePhone, nor do I see anything relating to the web based management system. So I’m guessing this ~135MB (expands to around 830MB Tar) is all the open source stuff that was required to be released, but not enough to actually roll a custom firmware…

This sounds like a good move for RISC OS. Well done on making it happen!

Personally, I might have chosen a different license, but the Apache 2.0 License is not a bad choice depending on what you want to achieve, and it’s certainly better than creating a whole new license that nobody has any experience with.

I’m surprised to see some unexpected supporters of copyleft licenses in this thread! It could be argued that a copyleft license might have been a better choice if preventing closed forks of RISC OS is a key aim. However, it’s good to start from a position of asking what you want to achieve, and I suppose that’s what ROOL and ROD have done.

It’s also useful to look at other similar projects and think about possible collaboration and code reuse, and think about whether the license enables that. Another related but separate issue is the possibility of taking advantage of modern code hosting platforms to make code contribution and collaborative development more convenient.

Really good news! Will the DDE also be opened up? Having to pay to contribute is something of a barrier…

Really good news! Will the DDE also be opened up? Having to pay to contribute is something of a barrier

Paying is still a good idea!

The team at RISCOS OPEN still need funding to continue their work!

They’re doing it for free! But they still have on-costs!

We all need to support them with cash!

This is brilliant!

I have written a few more words about that topic in my blog (all German of course):

http://riscosblog.huber-net.de/2018/10/risc-os-demnaechst-unter-apache-lizenz/

Great news for RISC OS users and developers. Thanks to everyone involved in making this happen.

Paying is still a good idea!
The team at RISCOS OPEN still need funding to continue their work!
They’re doing it for free! But they still have on-costs!
We all need to support them with cash!

Not that it’s up to me, and I’m sure people are considering such things, but in keeping with the way open source development works it’s more usual to have a donation system for those who can afford it/want to throw cash at things, plus people can still by sd cards/ROMs/physical PRMs etc should they wish… You get far more engagement if there’s no financial barrier to using the development tools.

The barrier to open source involvement has always been DDE, not the OS license. It’s about time that was open sourced and made freely available.

“The barrier to open source involvement has always been DDE” – an open-source OS that you can’t compile would be an odd thing which ever way you look at it… I’m hopeful that this is actually “The OS and build environment”, otherwise it’s all a little pointless imho. I remain positive though, and will look at the actual result with interest.