ROL: Giving Select 4 date is tricky

Published: 23rd Jun 2006, 06:04:41 | Permalink | Printable

Developers working in spare time

Giving a fixed date for the release of Select 4 and its Iyonix port is "very hard", the developers of RISC OS 4 have admitted. In an email to Select subscribers yesterday, a RISCOS Ltd spokesperson said the company had hoped to complete the task of allowing their OS to run on 32bit ARM-compatible processors within a year - but the project over-ran.

The statement came after Select users feared their wishlist ideas for future features were being ignored.

ROL said: "There has generally been a policy of no comment on the wishlist, because we have always tried to avoid any commitment to any specific feature in any particular version of RISC OS, because it is very hard to know how long a new feature will take to develop, test and debug.

"Our hope had been that the full conversion to 32 bit could have been done in less than a year, but that proved to be widely optimistic and that is why it is very hard to give a firm date for the release of Select 4 and an even harder to give any date on Select 4 for Iyonix."

The company believes a number of the listed wishes, hosted by Smartgroups, were implemented, and that RISCOS Ltd didn't want to step on the toes of other commercial developers if there was any cross-over with existing products. The majority of the development team, a small group of programmers contracted in by RISCOS Ltd, are also working on the OS in their spare time.

According to ROL, "RISC OS development relies very strongly on people who have 'real' jobs during the day and who would not be able to justify working solely on RISC OS work to make a living.

"Many employers also take a dim view of people doing paid work for other companies. Selling their own products outside of work hours is something they will allow however. A spinoff for this is that for a lot of those people their day job gives them extra knowledge and experience to help with the RISC OS development."

Users are said to be fed up with waiting for the next release of Select, version 4. The previous release under the annual subscription scheme was in June 2004, although subscribers have been asked to cough up cash and renew their subscriptions.

Links

RISCOS Ltd website

Previous: South East 2006 show date set
Next: RISC OS to visit Glasgow, Newcastle

Discussion

Viewing threaded comments | View comments unthreaded, listed by date | Skip to the end

"According to ROL, 'RISC OS development relies very strongly on people who have 'real' jobs during the day and who would not be able to justify working solely on RISC OS work to make a living."

"Many employers also take a dim view of people doing paid work for other companies. Selling their own products outside of work hours is something they will allow however. A spinoff for this is that for a lot of those people their day job gives them extra knowledge and experience to help with the RISC OS development.'"

Comments like this show that open sourcing the RISC OS kernel makes more and more sense. ROL could keep the GUI closed so they still have something to sell.

Will! on 23/6/06 8:06AM
[ Reply | Permalink | Report ]

People are continuing to contribute to the wish list, fingers in ears screaming "la la la" at the top of their voices. There should only be one wish - for ROL to release something/anything to Select subscribers and prove they are still in the OS development business.

druck on 23/6/06 9:05AM
[ Reply | Permalink | Report ]

druck:

Doesn't the A9 prove that ROL are still in the OS development business?

Not that that helps Select subscribers though :-(

helpful on 23/6/06 10:05AM
[ Reply | Permalink | Report ]

Am I the only one that *doesn't* want select for the Iyonix? The last thing that I want is for Castle to stop developing RISC OS 5 because they think that everyone is using select.

jeffd on 23/6/06 10:44AM
[ Reply | Permalink | Report ]

1. To druck: True, they should release - as the poll on drobe suggests getting anything is more important than what it is ... perhaps rebranding Select 3 as Select 4 might do the trick :-o

2. To helpful: "Doesn't the A9 prove that ROL are still in the OS development business?". Doesn't A9home mean converstion of the code to 32 bit (26/32 neutral to be exact) since the RISC OS on A9home is without the Select 4 features... but even so it is development though.

3. To jeffd: No, your're not alone - I am interested in Select-Features for IYONIX but not a full Select.

4. To jeffd: As for "The last thing that I want is for Castle to stop developing RISC OS 5 ...": Unless I missed some upgrade I don't remember any RISC OS 5 development in the recent past apart from amending drivers for nVidia, USB or Ethernet... I'd appreciate other work being done too (e.g. a better CDFS with Joilet support).

hzn on 23/6/06 11:32AM
[ Reply | Permalink | Report ]

Hzn: re your 4th point: I've had my Iyonix a year now. It came with 5.09 and since then 5.10 and 5.11 have been released, and as I understand it development is ongoing.

Concerning ROL, it appears to be a vicious circle: they can't say when Select 4 will be available because they can't afford full-time programmers, but they can't attract subscriptions until they give a release date for S4.

Stuffed, basically.

bucksboy on 23/6/06 11:59AM
[ Reply | Permalink | Report ]

In reply to jeffd:

You are certainly not alone. I believe that having more than one operating system is damaging to our platform and we should be bringing the strands together. Select for the Iyonix would mean we had three current operating systems: 4.x, 5.x, and 5.x with select features (I'm excluding users using older versions of 4 or even 3, because they're no longer available to buy).

cables on 23/6/06 12:38PM
[ Reply | Permalink | Report ]

To clarify - Select is only the name under which RO4 is developed. There will almost certainly be no such thing as 'Select features for RO5'. There will probably be a self-contained Select 4 for Iyonix, based on Adjust32, which means RO4.4x running on Iyonix hardware with hopefully complete hardware support a la RO5. If the hardware support is missing, I'd say there'd be little reason for ROL to offer it. It would be interesting to hear how ROL plans to offer hardware support; by developing their own drivers (the hard way) or by somehow extracting the drivers from RO5 and using them in RO4.4x (the more realistic way).

Furthermore, IMO it's undesirable to have anymore forks. If 32 bit RO4.4x can replace RO5 and add the nice extra's of the Select development program, I'd say RO5 could be left solely to the embedded or other uses for Castle. Probably Castle is selling their RO IPR anyway, it's just a matter of time.

hEgelia on 23/6/06 1:08PM
[ Reply | Permalink | Report ]

hEgelia wrote>"If 32 bit RO4.4x can replace RO5..."

Given the problems ROL are having getting Select 4 running on machines it is more familiar with (like RPC and A9Home) I would not be particularly optimistic about Select appearing anytime soon. To be honest UI changes are less important to me than what the hardware does/can be made to do (in short I'd not opt for Select if I lost *any* of the current functionality of my Iyonix).

As to ROL extracting "drivers" from RO5 to use - this suggests that the drivers and Select would be compatible (given the delay so far who knows....). What if Castle reasonably says "those drivers are for use with the RO5 kernel *only*......

As to Castle selling its RO IPR I sincerely doubt that - and if they did there is no guarantee that a hypothetical buyer (with deep pockets) wouldn't *close down* ROL to avoid competition.

AMS on 23/6/06 2:02PM
[ Reply | Permalink | Report ]

The problem with ROL is that they are indeed doing stuff. In fact I heard (and saw) some interesting things at the expo regarding RISC OS. E.g. Paul told me that out o/t +65 bug they knew when Acorn turned over RISC OS about 11 still remain. Also over the years they added (a few) extra features, made it microkernel instead of macrokernel, fine-tuned performance a bit and have it allmost completely made 32-bit. This is a major effort for which i have deep respect.

Commercially however... Oh boy. Instead of increasing the market by decreasing prices they're doing it all-around. The problem is that nobody will spend 200euros just to "support a guy doing some programming work abroad".

In that respect I think Chocky is doing it much better. Chocky's support costed only me about 20 euros a year which I could miss. In fact I never downloaded or wanted anything from his website (though I could since I payd a "subscribtion fee"). I just supported Chocky his idea and had no expectation as to what I'd really wanted to get ported. And I could afford to miss a few euros (even if he'd just go to the pub with it I wouldn't have mind). Untill I got out of job again unfortunatly. But Paul is asking too much. That way ppl will undoubtly have high expectations and will regret when these expectations are not met. For 200 euros I too would one at least 1 cd with "some good stuff" at least a year.

I think if ROL would just ask 20UKP a year, open a website for downloads and send a mug or a pen (or a bloody T-shirt for all I care); it wouldn't be as painfull as it is today. Ofcourse I understand that for a few thousand UKP you can't hire full-time programmers. But perhaps Open-sourcing the kernel and driver-support might be not be bad idea after all.

epdm3be on 23/6/06 2:28PM
[ Reply | Permalink | Report ]

In reply to epdm3b: Open sourcing the Kernel I don't is in an available option for ROL. This would require changes to the license from CTL. I believe I heard Jack say some time ago thay the did not want to open source any part of RISC OS.

sa110 on 23/6/06 5:13PM
[ Reply | Permalink | Report ]

@AMS: Castel have already stated, that they are willing to sell RISC OS if an interested party turns up. But who ever owns the RISC OS head licence, they can not close down ROL or revoke ROL's licence, because ROL paid a substantial licence fee for it.

JGZimmerle on 23/6/06 5:17PM
[ Reply | Permalink | Report ]

I'd be suprised (and a little disappointed at the insanity of those who did it) if more than a small handful of people forked out for another year of non-delivery. It would have to be one seriously impressive update and be available on CD or ROMs to buy and use right now before I personally considered buying a copy of RISC OS from RISCOS Ltd.

Unfortunately for RISCOS Ltd's shareholders, most computer users just want a functional computer and to be able to buy things that exist and that work on that computer. Not many people, not even RISC OS fans (who have proved themselves slightly more dedicated than the average Windows user), would happily keep a company afloat by effectively making annual donations out of the good of their hearts. This saga will be directly contributing to the contraction of the market. Since ROL is playing catch-up to every other desktop OS on the planet, it doesn't help to also create a perception of incompetence and lack of direction by failing to provide product for already-received cash.

arenaman on 23/6/06 5:18PM
[ Reply | Permalink | Report ]

AMS: "UI changes are less important to me than what the hardware does/can be made to do"

For me, the UI is the *only* thing keeping me using RISC OS. Cheaper, faster machines with multitudes of free software are out there now, and I find myself using them more and more. I still come back to RISC OS and use my Iyonix as my main machine but only because it is more pleasant to use, not because it provides functionality that other platforms don't provide.

"...there is no guarantee that a hypothetical buyer (with deep pockets) wouldn't close down ROL to avoid competition"

What competitor would seriously be scared by ROL and their part-time programmers enough to warrant spending money to close it down? This is a ludicrous suggestion surely only now held by conspiracy theorists!

IMO developing RISC OS commercially for desktops simply isn't working; that much is obvious from the troubles plaguing ROL to even release a few updates after 200EUR/subscriber and 18 months of nothing released. Open sourcing the kernel appears to be impossible without a massive change in licensing. We have two competing OS forks held by, for all intents and purposes, rival companies, and the split is getting wider all the time as development diverges each fork. There are huge gaps in required software, only partially filled by expensive commercial software (Oregano3, HTML Studio, PhotoDesk etc) and some excellent work by Peter Naulls and co. on smaller ported applications.

Where now for RISC OS? Something will have to change. The biggest threat to RISC OS isn't Microsoft any more, it's the impossible situation imposed on it by the two companies in charge, and the free and increasingly superior alternatives in the GNU/Linux world.

tamias on 23/6/06 6:57PM
[ Reply | Permalink | Report ]

JGZimmerle>"@AMS: Castel have already stated, that they are willing to sell RISC OS if an interested party turns up. But who ever owns the RISC OS head licence, they can not close down ROL or revoke ROL's licence, because ROL paid a substantial licence fee for it."

What reliable source have you got for that ? Has Castle issued a public statement/press release to that effect?

Even if Castle were to have stated that they would sell the head-license and that they *did* your second point doesn't follow.

Just having "paid a substantial license fee" *does not* guarantee that the licensee [ROL] couldn't lose their license. This *has* happened before to other companies. For example one scenario might be the buyer simply *pays* ROL back their original money - points to some termination clause in the contract and says "bye bye best of luck". Or if they're less *nice* about it they might just tell ROL you breached clause X, Y or Z now get lost....

In any event the *uncertainty* about the intent of a purchaser of the head license might be enough to encourage people to leave (I'd probably be one of them).

Change isn't always good - and at the moment RO is unstable enough without wishing such a change I would have thought!

Besides as ROL seem to be slipping further and further behind schedule it may be more likely that it is they that sell their added value IP back to Castle - now wouldn't that be ironic

AMS on 23/6/06 9:14PM
[ Reply | Permalink | Report ]

In reply to AMS:

Oh wouldn't that be nice: RO Ltd sell their enhancements back to Castle and the forks thereby merge. Can't see it happening though.

TonyStill on 23/6/06 11:36PM
[ Reply | Permalink | Report ]

Why doesn't everyone who would have subscribed put their cash into a fund to buy RISC OS from Castle? Then they could make it available under the GPL, potentially realising any value the software has - indeed, potentially generating value as people start to do interesting things with it.

Meanwhile, RISC OS Ltd. could continue with their proprietary branch of the operating system, just as the various prior agreements permit. Castle could also have some kind of special licensing arrangement.

But I suppose with the major players still living in the world of 1980s microcomputing, we can't expect something as sensible as the above to ever materialise until they're all in liquidation.

guestx on 24/6/06 2:55PM
[ Reply | Permalink | Report ]

epdm3be:

"made it microkernel instead of macrokernel"

Not true.

Will! on 24/6/06 4:58PM
[ Reply | Permalink | Report ]

sa110:

"Open sourcing the Kernel I don't is in an available option for ROL. This would require changes to the license from CTL."

No part of Select 4 is licensed from Castle.

Will! on 24/6/06 4:59PM
[ Reply | Permalink | Report ]

guestx:

"Why doesn't everyone who would have subscribed put their cash into a fund to buy RISC OS from Castle? Then they could make it available under the GPL"

Because the under the terms of Pace's license from Castle, open sourcing RO5 would require Pace's permission.

Will! on 24/6/06 5:03PM
[ Reply | Permalink | Report ]

Will>With respect.

ROL licensed *RISC OS* from *Pace*. Castle bought Pace's head license. Therefore whatever obligations ROL had to Pace - ROL now have to Castle.

It is *my* understanding from what was published at the time that you are correct that Castle can't (as a condition of purchasing the head license from Pace) GPL the RO5 source - however as *much* of Select is *still* just plain old RISC OS - yes the one Castle own the license for - then the same obligation extends to ROL. That's precisely why I suspect guestx suggested that people buy out Castle's rather than ROL's license as Castle's license is the *main* one.

As to Select 4 yes any parts of it that don't exist in RISC OS are indeed ROL's intellectual property - but given that these run on good old RISC OS (which ROL still have to license) the license terms still exist. I'd point to ROL's partial open sourcing of the Printing system - but they were unable to release *all* of the Printing source for that very reason. They could only release the bit's they made - not the bits that already existed and which they need to license off Pace (now Castle).

AMS on 24/6/06 5:19PM
[ Reply | Permalink | Report ]

One of the reasons why Select 4 is so late is that ROL has been re-writing much of RISC OS. The primary reason was to make it 32 bit compatible, but a secondary reason was to replace all code originally licensed from Pace with code owned entirely by ROL. ROL's intention is that Select 4 will be completely free of any third party licensing issues.

Will! on 24/6/06 5:37PM
[ Reply | Permalink | Report ]

Will!: "Because the under the terms of Pace's license from Castle, open sourcing RO5 would require Pace's permission."

AMS: "Castle can't (as a condition of purchasing the head license from Pace) GPL the RO5 source"

So, even if Castle actually are the owners of the RISC OS code, they are tied to some kind of exclusive licensing arrangement with Pace? Insert obligatory comment about business sense in the RISC OS scene.

guestx on 24/6/06 6:12PM
[ Reply | Permalink | Report ]

In reply to Will: ROL are only 32 bitting the existing RISC OS code. The very code that is licensed from CTL. Your reasoning behind ROL rewriting the whole of ROL, with the resources they have is wide speculation to the extreme. Besides, even if they had totaly rewritten RISC OS, they could no longer call it RISC OS and as it would be the same as RO5, I'm sure would not be able to sell it without a license from CTL. So either way, ROL still requries their license from CTL.

sa110 on 24/6/06 6:18PM
[ Reply | Permalink | Report ]

sa110:

"Your reasoning behind ROL rewriting the whole of ROL, with the resources they have is wide speculation to the extreme."

It's a little bit more than speculation.

"even if they had totaly rewritten RISC OS, they could no longer call it RISC OS"

Yes, they could. RISC OS is not a registered trademark.

"and as it would be the same as RO5, I'm sure would not be able to sell it without a license from CTL."

I've no idea how you figure a totally re-written RISC OS 4 would be the same as RO5. And anyway, Pace made some major changes to RISC OS between the version they licensed to ROL and the version they sold to Castle.

Will! on 24/6/06 6:46PM
[ Reply | Permalink | Report ]

guestx wrote>"So, even if Castle actually are the owners of the RISC OS code, they are tied to some kind of exclusive licensing arrangement with Pace?"

If I recall correctly (and I am quite happy to accept correction if I am wrong) the license was sold to Castle with "strings attached", I believe that Castle *still* have obligations to Pace (e.g., supporting the version of RO Castle provide Pace and not open sourcing it). Whether this makes sense or not (at one level) is beside the point - in that if the only way Castle had to obtain the license was to do that - I am sure they had little choice but to agree.

Will>To my knowledge RISC OS source is in the hundreds of megabytes of source (and a fair proportion of that ARM assembler). Other than making it 32bit compatible (basically changing the way flags/PSR is used) ROL would have left most of the code untouched - to do otherwise would have made the task *impossible* given the small number of people they had working on it.

I am not denying that ROL *have* added functionality, have "tidied things" and improved the UI - but that is a *long way* from a claim that they re-wrote the whole OS. Also I'd point out it would be relatively easy to check by analysing code from say (3.7) and comparing it to Select 4 and I'd be very surprised if more than a few percent of the code has changed.

Bear in mind Pace had the full source in 1999 and it took them (and subsequently Tematic/Castle) both with large teams until 2002 to produce a 32bit clean OS - on the other hand you expect us to believe that a very small team in ROL managed to re-write the whole OS in *one year*.

Part of RISC OS's problem is that because much of it is assembler it is *difficult* to maintain (on the other hand it makes it relatively fast so that's a two edged sword). It is not within the whim of *anyone*, RISC OS Ltd included, to simply junk the whole thing and start over - if the OS ROL offer contains the original RISC OS code (trust me it *must*) in addition to new code contributed by ROL then ROL still have an obligation under their RO license to the head license owner.

AMS on 25/6/06 3:18PM
[ Reply | Permalink | Report ]

AMS: "If I recall correctly (and I am quite happy to accept correction if I am wrong) the license was sold to Castle with "strings attached""

The "license"? Who actually owns the code? If it's Pace then Castle is in only a marginally better position than RISC OS. If it's Castle then they must have agreed to some pretty stupid exclusive licensing agreements with Pace to not be able to exercise other licensing rights as the owner of the work.

guestx on 25/6/06 6:22PM
[ Reply | Permalink | Report ]

"If I recall correctly (and I am quite happy to accept correction if I am wrong) the license was sold to Castle with 'strings attached', I believe that Castle still have obligations to Pace (e.g., supporting the version of RO Castle provide Pace and not open sourcing it)."

Pace didn't sell a licence to Castle. Pace sold the copyright of the code itself to Castle and then licensed it back from them, with a condition that Castle would not do anything that would diminish the value of Pace's licence. Pace use very little of the code licensed from Castle these days anyway.

"I am not denying that ROL have added functionality, have "tidied things" and improved the UI - but that is a long way from a claim that they re-wrote the whole OS. Also I'd point out it would be relatively easy to check by analysing code from say (3.7) and comparing it to Select 4 and I'd be very surprised if more than a few percent of the code has changed."

Do it. I think you'll find the results interesting.

"Bear in mind Pace had the full source in 1999 and it took them (and subsequently Tematic/Castle) both with large teams until 2002 to produce a 32bit clean OS - on the other hand you expect us to believe that a very small team in ROL managed to re-write the whole OS in *one year*."

ROL have been planning to replace the licensed code ever since their initial falling out with Castle and they haven't finished yet. So quite a bit more time than *one year*.

Will! on 25/6/06 10:49PM
[ Reply | Permalink | Report ]

Will!: Hang on a minute! If I was a Select subscriber (thank heavens I'm not), I would be even more angry if your speculation is indeed true. Rather than wasting their time on recoding stuff which is already done, should they not just get a new Select release out as soon as possible? OK, it's a great idea for ROL to recode RO and therefore releasing their contractual obligations, as that should make further development cheaper and easier. But why the hell would they allow their customer base to diminish so rapidly while they finish work which is really all a bit irrelevant to their PAYING customers? Surely the recoding stuff could have been done later?

Or is there some sort of clause in the license that a Select-style release is not allowed using Castle-owned code?

Has there been any triple-handling of code (i.e. recoded prior to Adjust32/S4, then 32-bitted again)?

timephoenix on 26/6/06 9:01AM
[ Reply | Permalink | Report ]

timephoenix:

"Rather than wasting their time on recoding stuff which is already done, should they not just get a new Select release out as soon as possible? OK, it's a great idea for ROL to recode RO and therefore releasing their contractual obligations, as that should make further development cheaper and easier. But why the hell would they allow their customer base to diminish so rapidly while they finish work which is really all a bit irrelevant to their PAYING customers? Surely the recoding stuff could have been done later?"

RISC OS Ltd's business sense is very much in line with Acorn's.

"Or is there some sort of clause in the license that a Select-style release is not allowed using Castle-owned code?"

There's no clause like that. Obviously previous Select releases have contained licensed code.

"Has there been any triple-handling of code (i.e. recoded prior to Adjust32/S4, then 32-bitted again)?"

Not that I know of.

Will! on 26/6/06 9:16AM
[ Reply | Permalink | Report ]

>> Giving a fixed date for the release of Select 4 and its Iyonix port is "very hard"

Absolutely pathetic.

nikgare on 26/06/06 11:58AM
[ Reply | Permalink | Report ]

Will wrote>"RISC OS Ltd's business sense is very much in line with Acorn's. "

A rather worrying turn of phrase.... it was after all Acorn who pulled the plug on the platform on black Thursday and many believe left their customers in the lurch.

But maybe you meant something else ?

will wrote>"Obviously previous Select releases have contained licensed code."

So your read on things seems to be that there were *some* licensed code in Select's *before* 4 but not Select 4 itself.....

So in effect you're saying that ROL have *completely* re-written RISC OS and that *none* of the code from versions licensed to them at all? And they did this between Select 3 (which from your comment *did* contain licensed code) and Select 4 (not released yet - but alledgely a subset of which runs on the A9Home). So in effect ROL with one or two programmers in a year did what Acorn/Pace/Tematic and Castle failed to do since 1999 with dozens of people?

You do realise *even* if that were true *any* ancilliary data and programs used (e.g., BBC BASIC, and all the icons, helpfiles,) would *also* be subject to license. Additionally if what you said were true *anyone* could implement RISC OS without having to pay a license to *anyone* - this would also apply to Castle who could simply implement all of the Select features without paying ROL a penny.

If you ask me it's what your saying is all *totally absurd* - and I've checked it *isn't* April 1st

AMS on 26/06/06 6:59PM
[ Reply | Permalink | Report ]

"it was after all Acorn who pulled the plug on the platform on black Thursday and many believe left their customers in the lurch."

You weren't there.

"So your read on things seems to be that there were some licensed code in Select's before 4 but not Select 4 itself....."

Yes.

"So in effect you're saying that ROL have completely re-written RISC OS and that none of the code from versions licensed to them at all?"

Do you want to try that again in English?

"And they did this between Select 3 (which from your comment did contain licensed code) and Select 4 (not released yet - but alledgely a subset of which runs on the A9Home). So in effect ROL with one or two programmers in a year did what Acorn/Pace/Tematic and Castle failed to do since 1999 with dozens of people?"

No. ROL started this project shortly after the beginning of their dispute with Castle. It's been in progress (albeit very slowly) for almost three years now. The replacement of the licensed code is a gradual process. Select 3 contained licensed code because at the time of its release, much of the licensed code had not been replaced and most of what had been replaced had not been sufficiently debugged to be released.

"If you ask me it's what your saying is all totally absurd"

Fortunately, no-one did.

Will! on 26/06/06 7:49PM
[ Reply | Permalink | Report ]

"Additionally if what you said were true anyone could implement RISC OS without having to pay a license to anyone - this would also apply to Castle who could simply implement all of the Select features without paying ROL a penny"

What on earth are you talking about?

Will! on 26/06/06 7:54PM
[ Reply | Permalink | Report ]

I guessed some time ago that that is what ROL have been busy doing with punters' 'subscriptions'. Doesn't surprise me, and yes it is absurd. 'RISC OS Jihad' fanaticism is the same the world over.

Roll on migrating more things to the macs learning to use x86 Linux and doing less with RISC OS.

blahsnr on 26/06/06 7:59PM
[ Reply | Permalink | Report ]

AMS: "Additionally if what you said were true anyone could implement RISC OS without having to pay a license to anyone - this would also apply to Castle who could simply implement all of the Select features without paying ROL a penny"

Will: "What on earth are you talking about?"

Quite! Why can't anyone implement "Select features" without paying ROL, as long as they don't lift the code straight out of Select? We're not living in an EU commissioner's "intellectual property" fantasy just yet.

guestx on 26/06/06 8:04PM
[ Reply | Permalink | Report ]

Will>Yes sometimes my typing doesn't go where I want....

When I wrote>"So in effect you're saying that ROL have completely re-written RISC OS and that none of the code from versions licensed to them at all?"

What I *meant* to write was "So you (will) in effect are saying that ROL have re-written *all* of RISC OS and *none* of the code from previously licensed versions are used in it."

I therefore should be able to apply Armulator to any part of it and find *nothing* from the old ROM there.

GuestX>Yes again appologies - so long as code is written without reference to the licensed source then there is no reason for anyone to pay the IP owner any license fee. But I don't think that that can be shown for Select 4 can it ?

AMS on 26/06/06 8:22PM
[ Reply | Permalink | Report ]

Will: "RISC OS Ltd's business sense is very much in line with Acorn's." So, throwing wads of cash at projects which are not immediately profitable? Are you suggesting that the embedded market is also beckoning to ROL, hence their rush to get an OS which doesn't need to be non-competitive with Pace's STB-version of RO? If they think the embedded market is going to become their primary one, then I guess the recoding makes sense.

But that will not reassure Select subscribers. Unless this is going to help an Iyonix-compatible version of Select avoid any legal obstacles, I can't see how users paying for this unnecessary process is fair. Customers want a product, and that means usable desktop features, not recoded stuff that already exists.

However, if this process does get a healthy influx of cashflow running through ROL again, then desktop users should reap the benefits eventually. But who's gonna stay loyal when their money has been used so poorly?

timephoenix on 27/06/06 09:07AM
[ Reply | Permalink | Report ]

"Are you suggesting that the embedded market is also beckoning to ROL, hence their rush to get an OS which doesn't need to be non-competitive with Pace's STB-version of RO? If they think the embedded market is going to become their primary one, then I guess the recoding makes sense."

No. RISC OS has failed in the embedded market. In 1999 its greatest asset was the ability to render fonts well at TV resolution. In 2006 that counts for a lot less and in every other regard it has been overtaken by other embedded OSes. Today even Pace use very little of the code they owned in 2003.

Will! on 27/06/06 09:24AM
[ Reply | Permalink | Report ]

Please login before posting a comment. Use the form on the right to do so or create a free account.

Search the archives

Today's featured article

RISC OS artist wows public with digital artwork
A RISC OS-using artist has described exhibiting his digitally-created work in a public gallery as a "rewarding experience". Richard Ashbery, who used ArtWorks and Photodesk to create his images, showed off patterns and colourful illustrations to punters, who told him his work made a change from the oils and watercolour masterpieces usually exhibited.

1 comment, latest by socris on 18/11/08 4:23PM. Published: 17 Nov 2008

Random article

Archive usage survey: VRPC edges past Iyonix
RiscPC king still rules the land

84 comments, latest by datawave on 17/11/05 01:27AM. Published: 10 Nov 2005

Useful links

News and media:
Iconbar • MyRISCOS • ArcSite • RISCOScode • ANS • C.S.A.Announce • Archive • Qercus • RiscWorld • Drag'n'Drop • GAG-News

Top developers:
RISCOS Ltd • RISC OS Open • MW Software • R-Comp • Advantage Six • VirtualAcorn

Dealers:
CJE Micros • APDL • Castle • a4 • X-Ample • Liquid Silicon • Webmonster

Usergroups:
WROCC • RONE • NKACC • IRUG • SASAUG • ROUGOL • RONWUG • MUG • WAUG • GAG • RISCOS.be

Useful:
RISCOS.org.uk • RISCOS.org • RISCOS.info • Filebase • Chris Why's Acorn/RISC OS collection • NetSurf

Non-RISC OS:
The Register • The Inquirer • Apple Insider • BBC News • Sky News • Google News • xkcd • diodesign

© 1999-2009 The Drobe Team. Some rights reserved, click here for more information
Powered by MiniDrobeCMS, based on J4U | Statistics
"Please give the Castle bashing a break... please!"
Page generated in 0.138 seconds.