ROL tells Select users: A9 takes priority

By Chris Williams. Published: 2nd Aug 2005, 22:12:44 | Permalink | Printable

Select 4? Please hold

RISCOS Ltd. has told its Select subscribers that there "isn't a definitive answer" as to how much longer punters must wait for Select 4 to be finished.

"The income from Select over the past twelve months has gone into the dual tasks of 32 bit conversion and adding new features as part of Select 4. The size of the task of the 32 bit conversion has been formidable and is still an ongoing process, but available resources have been prioritised into the A9 version," said a ROL spokesperson earlier today.

"Many new features for Select 4 have been incorporated into the A9, but the development plan means that the A9 version has to be finished first."

The A9 is a ARM9 powered range of computers from AdvantageSix, which use an 'embedded' version of 32bit RISC OS 4. The A9home is currently available in a beta state to paying developers.

The spokesperson added: "Last year we had intended to continue 26 bit and new 32 bit development work, but after a while it became obvious that working solely from the 32 bit version was the only choice. I realise that for the average Select subscriber it may appear that absolutely nothing has been done for them in the last year. The fact is however that a vast amount of work has been done, but 12 months has simply not been enough time, with the available resources, to produce a release version for 26 bit users.

"The imminent availability of the A9Home will hopefully show that the work has been done, and I hope you will be prepared to wait for the work to filter back to 26 bit users via Select 4."

Although it has been over a year since the previous release of Select, users are expected to receive numerous OS updates during a year's subscription. Various subscribers responded this evening by again demanding a version of Select for the Castle Iyonix and cancelling their subscriptions in protest.

One user said, "You have lost [my subscription] because of the lack of a definite statement with regard to Iyonix Select. And I think I'm not the only one. I will join again as soon as Iyonix Select is available or at least a definite statement will be made that it will be available. Not having a single RiscPC anymore but instead two Iyonixes, everything else would be charity."

Another added, "I do seriously miss Select with my Iyonix. Even the round buttons."

ROL say they are going to extend current subscriptions by a further three months as a "concession".

Links

RISCOS Ltd. website

Previous: The RISC OS File Repository opens
Next: 'Community' newsgroup proposed for RISC OS users

Discussion

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

While I can understand ROL's need to "target" resources (they ain't IBM or anything...) I would imagine existing 26bit customers are probably not particularly thrilled at this news (especially if they're not upgrading to the A9).

It also reinforces the skeptism some had regarding a 32bit Iyonix Select (I am sure some are even pleased they *hadn't* put down cash for Select given what now appears to have transpired). One might even suspect if *more* Iyonix owners had subscribed we'd still be seeing the above announcement - and then there'd be a predictable round of recriminations.

Perhaps this is a "push" to get A9 out before too many sales are lost to Castle's x400 - or perhaps it's all just what's said on the tin - who knows ?

AMS on 2/8/05 10:41PM
[ Reply | Permalink | Report ]

AMS:

"It also reinforces the skeptism some had regarding a 32bit Iyonix Select (I am sure some are even pleased they *hadn't* put down cash for Select given what now appears to have transpired)."

Just to be clear here, ROL's recent call for IYONIX users to show their support for Select on IYONIX did not necessarily include paying for a subscription up front. They simply wanted people to contact them pledging to buy a subscription once a suitable version was made available. They said they'd produce it once they received 100 pledges.

fylfot on 2/8/05 10:49PM
[ Reply | Permalink | Report ]

fylfot> The exact quote given on Drobe from ROL was "The plan at present is to simply continue to take pledges for Select for Iyonix and when we are in a position to proceed, we will ask to convert those pledges into real cash."

This does *not* infer that the payment was due on *completion* but rather that ROL might require payment when *they* were ready to proceed. Or else maybe I've misinterpreted what they meant. But in either event Select 4 (26bit customers) and any putative Iyonix Select subscribers would *both* have to wait in either event - the only question is would they have to pay while they waited.

AMS on 2/8/05 10:58PM
[ Reply | Permalink | Report ]

It does seem rather suspicious that ROL are pushing to release a 32-bit RISC OS for a computer which hasn't even been released yet, but can't be bothered (at the moment) to support a computer that's been available for 3 years now.

Smiler on 2/8/05 11:12PM
[ Reply | Permalink | Report ]

AMS:

Yes, you're right. But I suppose the important bit of what I was /trying/ to say was that ROL was not asking for subscriptions from IYONIX users just in case they *might* go ahead with a suitable version. Once they knew the support was there they would go ahead. Though we might now take their plans less seriously in light of this latest annoucement.

The whole point of Select, I thought, was that users could get their hands on the latest features from ROLs development team before anyone else. This worked in the case of Select vs. Adjust. Select users benefited from the beginning as the operating system was developed under the scheme. Users who stuck with RISC OS 4 had to wait until Adjust before they got their hands on all the updates.

But now the reverse seems to be happening. Select subscribers have waited for months and months to hear of a new version. And now they learn that the A9home users will benefit before they do. It simply does not make sense.

Select customers should be ROL's first priority.

fylfot on 2/8/05 11:24PM
[ Reply | Permalink | Report ]

Can anyone explain the business sense in focusing an OS release for a machine with currently very few users (the A9) as opposed to a machine with a large established user base (the Iyonix)? I subscribed to Select for two years after I got my Iyonix in the hope that a 32 bit Iyonixified version would be released, but decided my money would be better spent supporting other projects given that the 32 bit OS supplied with the Iyonix suited me perfectly well.

The A9 is a great machine for certain purposes, but I hardly see it dwarfing Iyonix sales given its limited scope for upgrades (I am talking about fixed graphics capabilities and lack of USB 2.0, here).

I would see more sense in focusing the initial 32 bit release for RiscPC users, in an attempt to get a single unified OS ready for all platforms (not that a 32 bit version for the RiscPC would bring many benefits in itself).

ksattic on 2/8/05 11:30PM
[ Reply | Permalink | Report ]

ksattic: I believe you've answered your own question regarding the business sense involved here: ROL clearly disagree that the Iyonix will sell more copies of Select for the A9 than they will for the Iyonix. They also make it clear that their order of priority is: A9, Select 4, possibly Iyonix.

To work out why they think that, it's pretty easy to do the maths. Make the assumption (just for a moment) that each Adjust-32 license is the same price as a regular copy of Adjust-32.

Every A9 machine ('Home' or otherwise) will have a copy of Adjust-32 with it. Iyonix users are certainly not a guaranteed market; many say they're already happy with 5.0X so take-up may not be great.

ROL are asking for 100 pledges for the Iyonix before they would consider progress - ie 100 more customers than they have now. If Ad6 were to sell only 100 machines, they would no doubt be extremely disappointed. Therefore, ROL are likely to sell more A9 licenses in the first couple of months of A9 production than ROL would expect to sell Iyonix copies, even if the 100 pledges target is made. That's not including not having to worry about major compatibility issues like 'how do you load Select onto an Iyonix when you don't have PCI support in Select'? OK - now I'm happy for you to make the license cost calculation less for a bulk purchase - if you halve it, you're guessing that Ad6 would sell 200 machines in total. I think you can see why ROL are making the assumption that they are. Of course, a die-hard Select owner then has an easy choice in any case - Iyonix with 5.0x, or A9 with Adjust.

As for the RISC PC - code will be 26/32 neutral, so the same code will apply easily to both machines; given the injection of life into the market the A9 would bring, I guess the priority will be making sure that Adjust-32 will meet the hardware requirements of the A9 before releasing general OS changes in Select 4 as features for both A9 and RISC PC users (given the common codebase).

md0u80c9 on 3/8/05 12:17AM
[ Reply | Permalink | Report ]

md0u80c9: "ROL clearly disagree that the Iyonix will sell more copies of Select for the A9 than they will for the Iyonix".

"Clearly" ?

Let's read that through just one more time.

"ROL clearly disagree that the Iyonix will sell more copies of Select for the A9 than they will for the Iyonix."

Right.

I'm not even going to start on the rest of what you wrote.

What ROL believe is obviously clear to you (hence the huge chunk of indecipherable text above), but not for most of the rest of us.

You do some work for ROL, right?

dgs

dgs on 3/8/05 1:22AM
[ Reply | Permalink | Report ]

dgs: OK - I may well have had a little expressive dyslexia there for a moment: I believe the remainder of the point was clear, which is why you won't comment

md0u80c9 on 3/8/05 2:18AM
[ Reply | Permalink | Report ]

I must say that the work ROL does does seem strange.

1. First of all they convert their RISC OS to 32 bit despite of one for 32 bit being there (RISC OS 5) and with that more or less just collect money from their paying subscribers to 26 bit Select and give them null in return for their money. 2. Then I remember them asking IYONIX users to subscribe offering a statement of intent to then work on Select for IYONIX - and this offer is pretty old and by now no results for these users either. By the way, ROL claimed a long time that they can't offer Select features for IYONIX users due to kernel changes needed - but on RISC OS Expo 2005 in a chat with users the statement of ROL was more like that they want to sell IYONIX users a full RISC OS instead of just some Select features. 3. And now ROL even ignores their paying subscribers more by focussing on the A9home. 4. But on the other hand their work is needed for the A9home since the other systems have a working RISC OS available.

Well let's look at the business side :-)

1. Looking into the A9home is probably a financially intersting path since it is a nice machine (I had one to try out and on some tasks it is close to IYONIX speed and on others never less than half of it and it was pretty stable so that it is indeed a RISC OS hardware with quite a bit power) and since it can be expected to sell in some figures. 2. As for the IYONIX pc it might be hard to get the 100 extra commitments since lots of IYONIX users are pretty happy with RISC OS 5 and till now RISCOS Ltd failed to deliver a concise list of new features they want to bring to the IYONIX pc with their Select so that the users can see if it is worth it - perhaps they don't write up that list since it would be too short since I don't want technical details like "module made 32 bit" or "reference to Acorn removed" but functionality like "DHCP client", "rounded buttons" or "image convert". 3. And when I think about the huge 26 bit Select user base ... well quite a few of them probably keep their subscription even when a new Select is still off quite some time ...

Assuming that ROL does step by here:

Please ROL, offer Select features for the IYONIX pc - at a fair price! That would probably make the IYONIX pc users happy and you'd probably get income - perhaps less per user but on the other hand probably from more users and that with much less work than finishing off a full operating system for the IYONIX pc including the hardware drivers needed.

hzn on 3/8/05 5:55AM
[ Reply | Permalink | Report ]

I'm pretty sure that, at Guildford last year, I paid an extra sub as both a RISC PC & Iyonix user, so that I'd get Select/Adjust for the Iyonix.

charles on 3/8/05 8:06AM
[ Reply | Permalink | Report ]

So what's a "fair price" for a Select upgrade for Iyonix users ? The price a user is willing to pay or the price that covers costs for the manufacturer ? Surely the whole point of the 100 commitments is to at least determine what a fair price could be and I'm assuming 100 people didn't come forward. If it costs 10,000 to work on Select for Iyonix that's 100 quid each which I'm sure a lot of people would say is too high. But are you asking me to believe that 10,000 is an extortionate amount of money to be spent developing code? That 10,000 would have to cover wages and production costs or is the expectation of Iyonix users that working on Select for their machine will only take a week or so ! I think its fair to say it will be man-months of work in which case that 10,000 is clealry not enough which in turn pushes up the unit cost of an upgrade to the end user.

A realistic assement of costs and the need to pay wages is missing here !

I have read many many times that Iyonix owners are happy with RO5 and see no need to upgrade to Select/Adjust for "nothing more than rounded buttons". With statements of intent like that based on misunderstandings of the content it should come as no suprise that ROLtd are abandoning Select for Iyonix. It seems to me that the 100 commitments was a last check by ROL and Iyonix owners failed.

mripley on 3/8/05 8:31AM
[ Reply | Permalink | Report ]

In reply to mripley:

As for a "fair price": This is certainly not easy since the potential buyer looks at what the product offers him or her and then has to decide if the price is worth it. From ROL's side of the deal they have to live off it as you correctly outlined.

As for your last sencence starting "I have read many times...":

The problem on this one is that until now I don't know what ROL will offer as for new functionality (and perhaps as for removed functionailty) with their IYONIX Select compared to the current RISC OS 5 - and as far as I know this lack of knowledge holds true for quite a few other IYONIX users.

So unless ROL puts forward a list of the features their IYONIX Select will offer in addition to what RISC OS 5 offers how are the IYNOX user to know what ROL intends to offer ... or do they expect customers to say, yes I'll pay for something of what I don't know what I will really get.

hzn on 3/8/05 9:21AM
[ Reply | Permalink | Report ]

"It seems to me that the 100 commitments was a last check by ROL and Iyonix owners failed"

As an Iyonix owner that has never subscribed to Select, I don't consider that the failure is down to me or others like me. To get my money, ROL need to convince me that I will get something useful for that money. They have so far failed to do so. In which case I prefer to use my money support projects that WILL add useful new functionality to our platform (like Firefox and Gimp-Print, to name two that I have supported). I'm afraid nebulous promises that perhaps there may be an Iyonix Select at some unspecified time in the fairly distant future, and perhaps I might find is useful will never compete with properly targetted developments with a known and useful end product.

mrtd on 3/8/05 9:27AM
[ Reply | Permalink | Report ]

The solution for ROL is very simple. I appreciate that such products may have long lead times, but they've had that and more. Offer a concrete product, and people will buy it, even if it only initially has a subset of features that you might like to include. If there's a specific Iyonix product available, then I'll buy it and I imagine lots of others will too. Please no more beating about the bush and wishy washy language.

mrchocky on 3/8/05 9:46AM
[ Reply | Permalink | Report ]

If there aren't already more features in 5 than in Select there will be at some point the way the development is going.

This list by Castle [link] suggests it won't take months to put Select features in 5. But the plan seems to be to make a complete new OS replacing 5, that will take a lot of (useless) time and the result probably will keep lagging behind in features.

ROL keep on doing nothing for for their old customers and failing to convince new customers. Hopefully for them the A9 will be a success, as you need Select to make it run the A9s success wil be ROLs success.

Jaco on 3/8/05 10:01AM
[ Reply | Permalink | Report ]

The trouble with producing a concrete product and then selling it is that it's a bit more risky than getting the money in advance, which is rather obvious, and ROL seem to be very, very reluctant to take any risks whatsoever. Time and time again they appear to have stuck to what they've got (OK, the A9 is a little different, so presumably something must've happened there), rather than grasp at opportunities. I can't help thinking that that's pretty bad for the long term future of RISC OS.

SimonC on 3/8/05 10:04AM
[ Reply | Permalink | Report ]

Sorry if I repeat some of ideas already expressed, but this thread is becoming very long! I don't understand the ROL point of view. Select subscribers are supposed to pay for updates to the OS, so, given that A9 AFAIK is still a development machine, I doubt that A9 32bit Select fall in this category. Also, why use money from generic subscribers to the Select scheme? I think is fair to open a pledge/subscription scheme for developers wanting to push the machine. If I own a RiscPC (or IYONIX) and want Select for that machine, why I'm paying for a product that probably will never use? What's the point? Ignoring IYONIX in favour of A9 is still suspect IMHO. To clarify: I own a RiscPC with RO 3.7. No intention of buying Select, nor IYONIX, nor A9. Effectively, I don't want a new computer (of all sort) at all!!

bernie on 3/8/05 10:18AM
[ Reply | Permalink | Report ]

bernie: then one could argue that it's up to ROL/Castle/whoever to convince you that you _do_ want a new product. That's really what this is about. Of course, if there isn't a product, then your reluatance is entirely understandable. Again, I appreciate the long lead times sometimes required (look how long Firefox took, and certainly isn't finished), but to put it into perspective, my Iyonix will be 3 years old next month - that's 3 years of income ROL could have had. There's risk with every product, but 3 years is a long time to delay a product.

mrchocky on 3/8/05 10:50AM
[ Reply | Permalink | Report ]

Personally, I think ROL have made the right choice here. It has been perfectly clear for sometime that in order for ROL to survive in the long term, they needed a 32bit OS and modern computer to run it on. AD6 have supplied the modern hardware. ROL are supplying the OS. Whilst producing the 32bit version of Adjust may have taken them longer then they anticipated, which has had a knock on effect of delaying Select 4. I think when one looks at the bigger picture, ROL have done the right thing.

As for Select 4, ROL have already stated that they are going to extend subs by 3 months, and Also stated that the new renewals will cover a 15 month period or a year after the release of Select 4, whichever comes sooner.

ROL seem to get critised no matter what they do. This time however, I think what they are doing is right.

sa110 on 3/8/05 10:55AM
[ Reply | Permalink | Report ]

Not producing a product is the right choice? Sometimes, perhaps, but not in general. As for the the A9, there were over 2 years before that in which ROL had a chance to produce something for the Iyonix. As for extending subs, that means my subscription expired in December rather than October last year (after paying for 3 years). Taking time out to pursue long term plans is fine - Look at what Oregan did with Oregano (even though RO users might have suffered), but you can't pursue software development in a vacuum. You need to be making sales, or you won't be able to do anything.

mrchocky on 3/8/05 11:00AM
[ Reply | Permalink | Report ]

Despite all the kissing and making up between ROL and Castle, it seems pretty clear that we have a forked OS still, and little real prospect of that changing in the near future. ROL's stance seems to be to support any plausible competitor of Castle's: first VRPC, now the A9. If ROL really wanted to embrace Iyonix owners there are plenty of good commercial reasons to justify that: owner base, modern hardware, and scope for development being 3 of them. The A9 is splendid and I wish it well, but it has 3 years' of sales to make to catch up with the Iyonix and that is not going to happen quickly, if ever. What is needed is a situation where future Iyonix buyers can opt either for RO5 or Select-for-Iyonix: if the latter is superior, as it may well be, most purchasers will choose it, and existing owners will upgrade. A unified OS will then have been achieved, and much future time and energy duplication saved.

bucksboy on 3/8/05 11:06AM
[ Reply | Permalink | Report ]

In reply to mrchocky:

But ROL are producing a product. They are busy finishing off Adjust32 for the A9, then will bring out Select 4 for the older hardware, as for the Iyonix - how long is a piece of string?

sa110 on 3/8/05 11:11AM
[ Reply | Permalink | Report ]

3 years long.

mrchocky on 3/8/05 11:16AM
[ Reply | Permalink | Report ]

But let us not forget, ROL and CTL have not exactly been on rosy terms these last 3 years.

sa110 on 3/8/05 11:27AM
[ Reply | Permalink | Report ]

The whole subscription "paradigm" always seemed pretty strange to me and I'm not at all surprised that it gets people/subscribers riled. (I stayed clear of Select, but have bought Adjust - since it's actually something real

)

I wonder if RO Ltd had/have problems getting credit which prevented them pursuing a more traditional business model.

Adam

adamr on 3/8/05 11:48AM
[ Reply | Permalink | Report ]

mrchocky: regarding my post, the bottom line was intended only to clarify my neutrality. My point is, I don't understand why spend money to make a product used by a future machine, when the Select agreement was to provide updates to an OS already existing. To be (almost) rude: if the maker of A9 wants a particular OS, fund the development of it NOW and regain the costs LATER (as I imagine Castle do). Money raised from Select subscribers is for funding current (normal) Select. Said that, I agree with you that IYONIX deserves the same consideration from ROL.

bernie on 3/8/05 12:01PM
[ Reply | Permalink | Report ]

ROL have changed tack a couple of times, this is normal in business as you have to adapt to changing situations, priorities and resources some of which may not be publicly known about. I am surprised Peter Naulls on the select mailing list implied (I can't check as I seem to have deleted the posting) that ROL could relatively easily and quickly release Select for the Iyonix, as I would have thought he would have known that combining the select features into the Divergent OS without access to the hardware specific code would be significant work.

p.s. Peter Bondar estimated that to 32bit the OS would take 25 man years, So Pace and ROL are to be congratulated!

CJE on 3/8/05 12:53PM
[ Reply | Permalink | Report ]

CJE:

Didn't Castle do this aleady?

25 Man years is a long time......

I have Adjust, and tbh, have enjoyed it...it's got all the stuff I needed...an up-to-date TCP/IP Stack ;@)

I'm going to hold off buying any more computers until this has all settled down and the final bugs removed ;@)

Would be nice to have an A9Home, and an iyonix ;@)

em2ac on 3/8/05 1:15PM
[ Reply | Permalink | Report ]

In reply em2ac:

A9home is a nice little machine. So quiet it stays on 24/7.

sa110 on 3/8/05 1:21PM
[ Reply | Permalink | Report ]

In reply to em2ac. No not Castle, Pace did it on thier divergent version of RIS OS which then Tematic took over, then Castle took charge, in the meantime ROL produced Select which was based on a different version of the sources, it had had no 32bit conversion work done on it (newer bits were written 32/26bit neutral). It was the other parts that it was estimated would take 25 man years todo.

CJE on 03/08/05 1:30PM
[ Reply | Permalink | Report ]

In reply to CJE:

It may have been estimated at 25 years, but ROL appear to have almost completed the conversion to 32bit/26bit neutral in substanially less time than that. All we can do is applaud their efforts and hope that Select 4 follows not too long after the completed Adjust 32.

sa110 on 03/08/05 1:34PM
[ Reply | Permalink | Report ]

Correct me if I'm wrong, but 25 man years is 1 man work - 25 years time. So, if the number of involved persons change, time changes too. Who knows the number of developers involved in 32bitting?

bernie on 03/08/05 1:55PM
[ Reply | Permalink | Report ]

I haven't implied that at all. In fact, I didn't make any mention or implication of difficulty. But regardless of any difficulties, it's still a very long time to overcome them.

mrchocky on 03/08/05 2:32PM
[ Reply | Permalink | Report ]

I'm afraid I don't have a great deal of faith in what PB might have said. Nor does it take into account automated tools like ARMalyser and the various things that Pace employees used and similar things likely employed by ROL and Castle. Apart from that, ROL _already_ have something that is close to saleable for the Iyonix - it was demonstrated at the last two shows, so I can't see further deliberation as anything but pointless stalling.

mrchocky on 03/08/05 3:03PM
[ Reply | Permalink | Report ]

"Pointless stalling" You are assuming you have all the information, neither you nor I have all the information, if ROL could make an imminenent release then why woudn't they? Many factors may be involved and some/all of them may not be appropriate to make public. And for you conspiracy theorists out there, I am not thinking of legal disputes.

Re PB he was reknown for being over optomistic, in this case fortunatly he was wrong in over estimating.

CJE on 03/08/05 5:03PM
[ Reply | Permalink | Report ]

Which information are you referring to? I don't think ROL are in a position to do an imminent release, but certainly soon, as demonstrated by what they had on show. If there really are other factors involved,then they should be referring to them (even if details aren't mentioned), not the stuff that's been said and this article quotes.

mrchocky on 03/08/05 6:40PM
[ Reply | Permalink | Report ]

Thing is ROL seems to *persistently* position Iyonix at the bottom of their pile of things to do. They have (for example) openly supported the Omega, when MD needed it they even supported RISC OS under Windows (MD's Alpha PC Laptop being the first to receive ROL's RISC OS version for use under emulation), and now that the A9 is in the wings it's the A9 that gets support.

Is it just me or does this amount to a *pattern* ? It may or may not be pointless stalling - but it *is* consistenty done whatever way you look at it.

AMS on 03/08/05 7:02PM
[ Reply | Permalink | Report ]

Reluctantly, I'm coming round to the same point of view as AMS. There is little evidence of any enthusiasm by ROL to produce an Iyonix Select. For example, I'm an Iyonix owner and a FRU subscriber (so I'm on ROL's mailing list even though not a Select subscriber) and I only know of this Iyonix pledge scheme by rumour - I've never been contacted. In fact, I thought it was 100 *subscriptions* they were after.

I'm not a marketeer but I can't understand a sales technique that relies on rumour and people seeking out your product when you have access to much of the potential customer base (where is the ROL announcement anyway - serious question?).

I also believe, and I've said this before, that the Select product for Iyonix should be something that runs on top of the Castle RO5 core. Why should the graphics and GUI facilities added in Select (and, since I'm a potential customer, my ignorance of what else is in Select is surely understandable) impact the core OS? Looked at this way, the product would not be complex to provide now (ROL are on record from a long time ago as saying that all their new development is 26/32-bit neutral).

Finally, I really think the duplication of Iyonix hardware specifics would be pointless; the only likely market is Iyonix machines that already have it.

TonyStill on 03/08/05 8:40PM
[ Reply | Permalink | Report ]

In reply to bernie: Yes I absolutely agree that the correct behaviour or ROL would have been to look into servicing their PAYING subscribers with a new Select instead of diverting that money off to adapting Select to the A9home. And now they seriously expect IYONIX pc owners to put forward money (I know that they won't collect money unless they start) when their recent history does show that they use the money for what they think is worth it instead of for what I pay it for - not me; I'll wait for the real product and then decide if it is worth the price.

OK, in the initial setup ROL planned to go for a 32bit operating system which was one reason for me to subscribe, but that was dropped so I'd say the subscribers of Select are already used to not believe what ROL states as what they intend to do ;-)

I can accept that the odd feature mentioned to be included in some release doesn't make it in time - but that has to be limited to the odd feature. But as a subscriber to 26bit Select I'd absolutely not accept that my money is used for something that is of no use to me or that what I paid for (upgrades offering a reasonable amount of new functionality) do not come by or are deleayed too much...

In reply to mrchocky: Wow, then my IYONIX pc is reaching the three year mark soon too. And we all know of ROLs committment to support all "RISC OS systems" offering Select for them. OK, everybody knew that that was not really true, since nobody expected them to offer Select for hardware like my A5000 :-)

Perhaps RISC OS 5 was not considered to be RISC OS by them for quite some time so that they thus pretty much ignored the IYONIX pc - erm. but didn't they state that they dropped their 32bit plans due to lack of hardware for that - that hardware is there since some 36 months and even has a 32 bit RISC OS which should save them lots of work... So until recently for ROL "all RISC OS systems" obviously meant most 26bit systems like the Risc PC, A7000(+), Omega and VirtualRPC and now the A9home is included (and the others on a back-burner).

To ROL: I dare repeat myself: Finish off the A9home and service your PAYING select subscribers. Then start off to wrap up packs with Select FEATURES to run on top of RISC OS 5. That should even be financially viable since why duplicate all the work Castle did spending lots of hours (wasting them really) expecting the users to pay for that waste. Just do the smaller task to make the intersting Select modules/apps/code run on RISC OS 5 and sell that at a accordingly lower price - you might even get more users wanting that and thus make more money.

hzn on 04/08/05 08:15AM
[ Reply | Permalink | Report ]

Isn't the problem with "just adding the Select stuff to RISC OS 5" is that a lot of the changes have been to the core of the OS, so without those changes being made in 5, the Select features won't work. While it's not an issue in this case, the I believe the DHCP client in RISC OS Select won't work on a plain 4.02 machine - you need the other "behind the scenes" changes to support it.

cmj on 04/08/05 4:20PM
[ Reply | Permalink | Report ]

There's always complexities like this in any project. It's true that some features might work, and that there might have to be work elsewhere to make them work fully or sensibly on RISC OS 5. But it's a little bit academic unless we can name specific features that might or might not work. Nevertheless, at least some of these issues have been overcome, as evidenced by what was showing on ROL's Iyonix. I don't expect a full set of Select features, and perhaps not even a large proportion straight away, but they have to start somewhere, and IMO, there's enough for a product.

mrchocky on 04/08/05 8:19PM
[ Reply | Permalink | Report ]

I believe AMS is right in that there is a pattern in their 'Iyonix evading' behaviour of ROL, which might be attributed to the still existing conflict between Castle and ROL. I say -still existing-, because I am hardly convinced their 'differences' are resolved by the apparent lack of (public) unified and common progress on RISC OS. If it has been resolved, why is the public still unclear about it then?

Surely certain 'deep rooted' Select enhancements rely on ROL's sourcecode tree, but other more cosmetic features may well be delivered as a series of patches and enhancements rolled into a neat Select for RO5 upgrade. But then, IMO, we would miss the refinements, 'cleaning up' and changes made to the kernel and associated modules of RO4, which also was a focus of the Select initiative. In this light, I can understand why ROL would choose to bring Select for the Iyonix in a self-contained form. If that is the case, ROL would have to know exactly what there is to know about the Iyonix, in order to fully support its architecture. Jack Lillingston of Castle made clear at the recent Expo in Holland, that ROL is welcome to get the tech-spec... but then ROL would have to pay for it, I guess and develop all the necessary code and integrate it into their own sourcecodes which means a lot of work=time=money.

Anyway, personally I find it disturbing how ROL chooses to treat its customers of 26 bit Select. Perhaps, as a form of compensation, affected subscribers of Select can be offered a discount on the A9home when released, instead of a free 3 months extension.

hEgelia on 07/08/05 02:01AM
[ Reply | Permalink | Report ]

I suspect I'm missing something obvious here, but aren't there two discrete possibilities under discussion here? Namely, Select for RISC OS 5 and Select for Iyonix. From what I can tell, what was spotted at Wakefield was probably Select for Iyonix - that is, RISC OS 4.5 ported to Iyonix hardware.

I'm saying that because without agreement, ROL engineers wouldn't have access to RISC OS 5 source, and I'm assuming such an agreement hasn't been reached otherwise that news would have been made public.

ninja on 07/08/05 03:02AM
[ Reply | Permalink | Report ]

IIRC what was on show at the RISC OS Expo was an Iyonix running RISC OS Select 32 wich in turn used some of the hardware specific modules from the RISC OS 5 ROMs.

JGZimmerle on 13/08/05 11:08PM
[ Reply | Permalink | Report ]

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

+++ Message board +++

Read messages or start a new thread

Search the archives

Today's featured article

LCD monitors and RISC OS
Shopping for TFT?

25 comments, latest by skock on 9/4/04 1:23PM. Published: 13 Mar 2004

Random article

Star Fighter 3000 Review
RISC OS gaming classic better than ever

25 comments, latest by The Doctor on 9/1/04 8:54PM. Published: 5 Jan 2004

Create a new account
Forgot your password?

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
"This poll is not indicative, and doesn't reflect public opinion"
Page generated in 0.3018 seconds.