Select nets 1,000th subscriber | Drobe.co.uk archives

Select nets 1,000th subscriber

Published: 15th Jul 2007, 23:00:32 | Permalink | Printable

Focus now on A9home and VRPC, still not friends with Castle

RISCOS Ltd is pushing ahead with development for the VirtualRiscPC and A9home platforms after bagging its 1,000th Select subscriber. ROL admitted it is making slow progress with the Iyonix Select port, claiming a "lack of interest from Castle" - but its developers are full-steam ahead with A9home and VRPC work. The company added it has sold over 6,000 vanilla RISC OS 4 ROM sets, 1,800 Adjust ROM sets, but said it has only heard from 130 Iyonix users who want Select on their Xscale-powered machines.

A ROL spokesman told Select subscribers: "The fastest development on RISC OS Six happens with our committed partners Advantage Six and VirtualAcorn, and we have much planned to take advantage of the features available with Virtual Acorn on PC and now Mac, as well as a full release of RISC OS for the A9Home.

"The release of Select 4 has shown that there is still demand for enhanced versions of RISC OS from new and existing users, with the 1,000th RISC OS Select subscriber recently being signed up.

"We are still committed to offering Select features to Iyonix users, but the lack of interest from Castle to support the project means that we are only able to make slow progress in that respect."

Playing down a suggestion that ROL would support a 'VirtualIyonix' emulator, the spokesman added: "From our point of view the lack of new features in RISC OS 5 since its launch four years ago speaks volumes.

"RISC OS 6 has far more advanced features than RISC OS 5, as has been shown by the number of users who want RISC OS Select features on their real Iyonix."

It's understood a gap has somewhat opened up between RISC OS 4.42 on the A9home and RISC OS 6.06, released for RiscPC-class machines and emulators. For instance, issues present in 4.42 and the preview of 6.00 have been addressed in the 26bit RISC OS 6 stream, and these updates must now be brought across to the A9home OS, according to well placed sources.

Members of the close-knit development team have been hit by a number of personal set-backs, delaying efforts to synchronise the 32bit and 26bit versions of ROL's OS - effectively holding up Select for the A9home and the release of a new A9home Flash ROM.

AdvantageSix are set to announce details of a new RISC OS release for A9home users in due course. Castle were not available for comment.

Links

RISC OS Select website - keep an eye on this page for ROL's RISC OS 5 v Select 4 table

Previous: New tool to encrypt files using GPG
Next: ConvText utility now up to version 2.04

Discussion

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

I really like the sentence "From our point of view the lack of new features in RISC OS 5 since its launch four years ago speaks volumes.". While this is basically true I think there are a few things not taken into account here:

1. This kind of remark by a company collecting subscription fees from their subscribers for years but not delivering the promised results in time or with a reasonable delay but with a very significant delay but no sensible reimbursement ... and then to top things a new OS with only very little new features for their main customer base (32bitting is worth nil for the Risc PC users) ...

2. Currently there is not much as for the OS in RO6 that would make me want to part with much money to get it. New features are not enough to get me to part with money - for that the features must be worth the price. Not that there is not quite a bit in RO6 that is not in RO5 but for me to part with money the featureset has to be worth the price.

3. I guess that quite a few IYONIX users simply asked ROL the odd time to simply deliver the odd Select feature - no need for a full OS in 32 bit since we have that already. E.g. the ehanced Paint and Draw app etc. should be not problem - if they are then that suggests the question of how other even more versatile apps like Photodesk and ArtWorks run.

4. And the "committed partners" opposed to Castle. ROL, do not forget that Ad6 and VA need your OS for their hardware and thus you work together. But CTL has no need for Select - you are the one wanting to sell Select to IYONIX users, not CTL. Thus this one is completely in your ball park. CTL doesn't need your OS since they have one up and running and 32 bit too.

As for the 1,000th subscriber: Concurrent?

hzn on 16/7/07 7:18AM
[ Reply | Permalink | Report ]

"From our point of view the lack of new features in RISC OS 5 since its launch four years ago speaks volumes."

Yes, it's shocking the way that Castle have only implemented support for new hardware like USB2 and multiple graphics cards, rather than getting down to the serious business of allowing rounded corners on action buttons. More seriously, I agree with hzn that this is a bit of cheek coming from a subscription-funded company that gave their subscribers nothing for three years - especially since Castle have a well-earned reputation for keeping quiet about new developments until they're actually ready. Yes, there are some useful changes to things in Select that RO5 doesn't have - most notably some of the improvements to the bundled apps - but none of them is exactly earth-shattering.

chrisj on 16/7/07 8:49AM
[ Reply | Permalink | Report ]

I have just received Select 4 and then a resubscription notice. Suddenly I am thinking - what is the purpose of resubscribing? I have an A9 home, RiscPC and Iyonix. At the last show 'one off' copies of Select4/RiscOS6 were sold for £50. The Iyonix is already horribly non-standard over its use of the delete key which annoys me every time I use it - that's the fix I want for the Iyonix, not Select.

CKH2 on 16/7/07 11:04AM
[ Reply | Permalink | Report ]

CKH2: I presume you're aware of the existence of Steve Fryatt's "PC Keyboard" module? [link]

Also, fwiw, Acorn specified (in the Style Guide iirc -- I don't have it to hand to check) that, when using a PC keyboard, the delete key should have PC behaviour. The Iyonix has a PC keyboard.

jmb on 16/7/07 11:59AM
[ Reply | Permalink | Report ]

Can we confirm that "Castle were not available for comment" is a genuine statement?

If it is, then I'm sorry but that's just not good enough.

People are spending over a THOUSAND pounds at a time with this company. Others have spent this money with you in the past and are waiting on your next move for a major purchase. You cannot treat customers like this and expect to be regarded as an fair and decent company.

If there is a not future for RISC OS then *please* Castle can you have the decency and basic respect for people to say so. Witholding like this has caused immense damage already.

AW on 16/7/07 8:09PM
[ Reply | Permalink | Report ]

Cannot ROL simply get the information from ROOL, or does ROOL not have the source code to hardware specific areas?

sa110 on 16/7/07 8:17PM
[ Reply | Permalink | Report ]

AW:

I'm not quite sure what Castle have to do with it? The product being discussed is ROL's, and as far as I can see, Castle have delivered everything that they have promised to Iyonix owners (ie. a machine, a 32-bit OS and subsequent updates). The issue of RISC OS 6 is between us and ROL, if we want it.

stevef on 16/7/07 9:53PM
[ Reply | Permalink | Report ]

CKH2:

As noted above, there's a version of PC Keyboard (version 2.09) specifically updated to allow the Iyonix keyboard to be mapped back to the Acorn-style Delete, for those who want it (the opposite of its more usual use for making rogue applications behave in the correct RISC OS 5 style). I've been told it works fine like this...

stevef on 16/7/07 9:54PM
[ Reply | Permalink | Report ]

Although "lack of interest from Castle" is blamed, these two statements seem entirely different in tone:

"[...] it has *only* heard from 130 Iyonix users who want Select on their Xscale-powered machines." (emphasis added)

"RISC OS 6 has far more advanced features than RISC OS 5, as has been shown by the number of users who want RISC OS Select features on their real Iyonix."

Hmm.

Jaffa on 16/7/07 10:06PM
[ Reply | Permalink | Report ]

Well RISC OS developer may be being a bit disingenuous by not mentioning the license fee paid by Castle for RISC OS but "no comment" as if the even bigger investment in Castle by thousands of RISC OS users over the past 10 years is not their concern and beneath them is downright ignorant and arrogant if true.

This issue reveals the ghastly situation that an OS with world-beating software is in: the main hope for hardware development aloof and distant and the OS developer hedging its bets on the A9 - a worthy but inferior computer to the Iyonix - or facing certain oblivion with the sweetened pill of a lumped retirement sum if the only hardware RISC OS runs on is PC/Mac based.

AW on 16/7/07 10:54PM
[ Reply | Permalink | Report ]

AW:

I still don't see why Castle need to comment. RISC OS 6 is nothing to do with them, and I'm not aware that they ever promised it to those buying Iyonixes. Given that, where does the ignorance and arrogance come in?

stevef on 16/7/07 11:20PM
[ Reply | Permalink | Report ]

Doesn't "Castle were not available to comment" simply mean that when Drobe phoned them they weren't there to answer the phone? It doesn't say "Castle said, "No comment"", which is what people seem to be reading it as.

OwlArt on 17/7/07 9:05AM
[ Reply | Permalink | Report ]

OwlArt: Or Drobe phoned and asked to speak to somebody to provide some comment, and they declined

rjek on 17/7/07 9:36AM
[ Reply | Permalink | Report ]

Wouldn't that be "Castle declined to comment" rather than being "unavailable to comment" though? I'd have interpreted "unavailable" as not being able to speak to someone.

OwlArt on 17/7/07 9:46AM
[ Reply | Permalink | Report ]

OwlArt: Well, being able to talk to a telephonist who says nobody's in at the moment strikes me as more "unavailable" than "declined", although one is easy emulated if you want to avoid reporters.

rjek on 17/7/07 10:20AM
[ Reply | Permalink | Report ]

stevef:

"Castle have delivered everything that they have promised to Iyonix owners (ie. a machine, a 32-bit OS and subsequent updates)."

Sort of. They raised expectations about future development though with the Merlin project, which quietly died a death.

[link]

[link]

To their credit is the formation of ROOL, which is a fantastic development, and the release of the sources to RISC OS will I'm sure bear fruit.

I recently purchased two Iyonix computers, and would dearly love to update them with the Select features I value on my Virtual RPC Adjust (e.g. enhanced Paint, Draw, Image filer). A combination of updates from ROOL and the release of Select features from ROL would be perfect for me.

Stewy on 17/7/07 10:40AM
[ Reply | Permalink | Report ]

In reply to AW & Stewy & others:

RO6 for IYONIX is indeed purely a ROL issue - period. That Ad6/VA work together with ROL is no surprise, since they need ROL's OS, whereas CTL does not. If ROL (or whatever other company) wants to supply software for any computer the work is in their ball park. Question arising is why ROL doesn't seem to want to supply the intersting Select features like the enhanced Apps etc. as normal apps for RO5 - that is surely possible, other suppliers manage to supply apps for RO5 too. OK, ROL keeps on claiming that that is not possible due to their kernel being so different; this might be important for a few things but most Select features can be done on RO5 for sure. Furhtermore they simply state that they don't know the hardware and thus can't supply a full OS - who wants a full OS anyhow?

A while ago ROL claimed they need 100 IYONIX user wanting Select - well they stated to have 130 and I assume that more would buy something ROL has to offer ... if they do really offer something. My personal impression I get from whatever information about IYONIX Select is that ROL doesn't really want to do IYONIX Select but instead of clearly stating so they keep blaming the lack of progress on others, be it that 130 IYONIX Select wanting users is too little, that CTL doesn't help them as much as they want to, lack of resoruces, lack of paying IYONIX Select subscribers ....

Furthermore if they would really want to get the IYONIX users to part with money for Select they should offer the new Paint, Draw, Trashcan, Image Filer, Image Convert etc. as a software bundle for RO5 at a resonable price. I know that quite a few IYONIX users would be interested in just that - as a matter of fact I'm not sure I even want a ROL OS for my IYONIX pc - why pay for a complete OS and perhaps tackle with the odd issue and problem just to get the odd enhancement when those enhancements could be supplied as add-ons to the current one?

As for Merlin my guess is that CTL was more or less forced to ditch it since it kind of died after the legal dispute ROL-CTL where one result was the impression that desktop RO is ROL's game and furhtermore AFAIR ROL started to promise to supply Select to IYONIX so that both companies doing the same, i.e. supplying the odd RO desktop enhancement would have been a waste of resources.

hzn on 17/7/07 11:53AM
[ Reply | Permalink | Report ]

Nicely said Herbert. If ROL offered a software bundle as you suggested I would buy it - but I refuse to subscibe to Select with no evidence that they are serious about making an Iyonix version.

Eddie on 17/7/07 12:36PM
[ Reply | Permalink | Report ]

Shouldn't the objectives of ROOL facilitate ROL to produce select for the Iyonix?

Surely the problem must be that it hasn't progressed as far as would be needed?

(Presumably with the disagreements that Castle have had with ROL in the past, they see as intransigence on Castle's part.)

The ROOL initiative seems like it could be the answer to a unified OS model.

As long as ROL and ROOL keep/make the structure of the two RISC OSes compatible, so a working system can be constructed from a mix of components as needed, having two systems needn't be the disaster it could be if aprroached wrongly. (It might even have adavantages).

jess on 17/7/07 12:48PM
[ Reply | Permalink | Report ]

The reasons why the ROL applications can't be ported to RISC OS 5 have been stated time and time again: they rely on new features of the OS to provide their enhancements that CTL's RISC OS does not have.

In terms of porting an OS, nobody other than Castle has managed to write an on-the-metal OS for the Iyonix. Both Linux and NetBSD depend on RISC OS having already set stuff up. Castle need to provide documentation for their hardware to let other OS authors port their operating systems to the Iyonix. I don't spot any such details on their web site.

rjek on 17/7/07 1:20PM
[ Reply | Permalink | Report ]

Drobe (or the Icon Bar) should really have some kind of betting service where we people can put their money on everyone having more or less the same conversation in two, five, ten years time. The profits could fund development on a Free Software version of RISC OS, perhaps...

guestx on 17/7/07 1:48PM
[ Reply | Permalink | Report ]

Rob: since Select is softloading anyway, there is no need for it to go "on the metal" at all - just like the Linux port available, RISC OS 5 will have set up everything nicely. Apart from that, the IYONIX does not use "magic hardware" anywhere, just bog-standard stuff.

It is of course not sensible to replace the whole RISC OS 5 stuff with the typical "Select softloading mechanism" which just replaces the whole OS. You'd have to reinvent a lot of stuff that works just fine (USB drivers, nVidia drivers, IDE/ATAPI stuff). I guess the way much of the Select extensions were done is pretty much incompatible with the idea of softloading only a few parts of it over another RISC OS version. Which is just too bad for RISCOS Ltd. in the current situation.

Anyway, there is still a lot of Select stuff which should be portable in a modular way to extend RISC OS 5 with minimal changes - the whole ImageFile* architecture and with it !Paint, the thumbnailing filer, CDFS, networking stack, alpha channel and CMYK sprites... problematic things would surely include the window manager changes (round buttons) and perhaps the system-wide cut&paste&drag of textfield content.

I am quite sure that the money from 130 additional Select subscribers would be more than enough for the work needed to provide such an "IYONIX Select". So we end up with the same conclusion as always: RO Ltd. could do it, but they don't want to do it. For whatever reasons.

hubersn on 17/7/07 2:59PM
[ Reply | Permalink | Report ]

hubersn: tbh, I'd have thought the ImageFile stack and the new sprite stuff would be much more complex than the WIMP upgrades. Quite a few things depend on new features in the kernel, where all that's happened for rounded buttons is the abstraction of the WIMP's icon drawing routines to an external module.

rjek on 17/7/07 6:25PM
[ Reply | Permalink | Report ]

rjek: I thought that ROL were supposed to be *

stevef on 17/7/07 6:48PM
[ Reply | Permalink | Report ]

rjek: Good to see that the Show Help and Preview buttons still submit the form, even in NetSurf 1.0...

Anyway... I thought that ROL were supposed to be removing stuff from the kernel and abstracting it to external modules, not adding yet more functionality and then tying down the "ROM applications" to that new code?

stevef on 17/7/07 6:50PM
[ Reply | Permalink | Report ]

Yes, there is a sense that ROL's answers are excuses rather than truth. Like someone else said, there seems to be an inconsistency in the two reasons given for not producing an IYONIX Select. If a lack of support from Castle is the show-stopping reason, then why was it ever suggested that at least 100 people would need to register their support in order for the project to be pursued? And, of course, 100 people did get together on a pledge bank, but it was quite ruthlessly ignored.

In response to Steve Fryatt's question ("I'm not quite sure what Castle have to do with it?"): I think the point is that it is felt that Castle *should* have something to do with it. If there's a market for a Select on IYONIX - which there quite obviously is - would Castle not benefit from perhaps a few extra sales of IYONIX pcs and a generally healthier and happier market place, clearly something that would be good for everyone? In other words, ROL and CTL need to cooperate.

Ultimately this is a political issue. Anyone who's asked Castle what they think of Select will know that they haven't got much good to say about it. Select offers nothing that would offer them a commercial advantage when selling to industry over RISC OS 5. So perhaps Castle just don't give a sh*t about the enthusiast market any more? I think this is why I probably have a soft spot for ROL and Advantage6. They're doing stuff that doesn't, on the surface, make much commercial sense. But the difference is RISC OS excites them in a way that has little to do with a profit margin. Most people will know that ROL pay their developers appallingly, if at all. Stuff gets done because of cooperation between geeks, essentially. So, faced with an uncooperative, hard-nosed "show us your money" approach from Castle, there are obvious reasons why the two factions don't get on.

The RISC OS community is a bunch of very decent and generous individuals. I think we forget this because of the bickering caused by the CTL vs ROL politics. Offer an average RISC OS users a slight sniff of the possibility of development and they'll get their cheque books out, even when there are poor guarantees of a return. Without meaning to place myself within the category of "decent and generous" (but go on, I will anyway), I renewed my Select subscription recently when I was not well placed to do so financially. I think there are many like me. Loyal people who love using RISC OS and want to secure its future. I think, therefore, we deserve a bit more respect from ROL and CTL. ROL need to be completely honest with us. We're tired of the lame excuses and the changing goal posts. Is it, as I suspect, because CTL just isn't interested in the enthusiasts' market any more? And if this is true, then I hope CTL might realise the damage of their actions. Perhaps CTL should have a little more respect for those who handed over good money for an IYONIX (the same people who have probably supported you for years, buying Kinetic upgrades, etc) and who do value Select and, therefore, wish to have Select on their computers. I mean, just how disrespectful is it to say to someone who has bought an IX and a Select subscription that Select is crap and we won't make big bucks out of it? This, I submit, is essentially Castle's view. They've said as much at shows, I'm sure others will agree. The "decent and generous" are packing up and leaving because of the fragmentation and because they're treated like sh*t.

ROL and CTL owe it the community to cooperate on this one. If they can't, then we might as well leave them to it.

(Apologies for such an emotional comment. I've had very little sleep recently. But don't worry, it's not the ROL vs. CTL politics that's keeping awake).

fylfot on 17/7/07 6:59PM
[ Reply | Permalink | Report ]

As you you say they're both in it together and "no comment" when the whole situation with RISC OS is unresolved and is everything to do with Castle as much as it is ROL is the pits.

If only there was an intermediary experience and respected enough and big enough in character to bring the two parties together.

AW on 17/7/07 7:21PM
[ Reply | Permalink | Report ]

AW wrote>"but *no comment* as if the even bigger investment in Castle by thousands of RISC OS users over the past 10 years is not their concern and beneath them is downright ignorant and arrogant if true."

Well that's how you're choosing to read it, I would disagree though. Castle paid for and developed the Iyonix and the OS to go with it. Their primary responsibilities are to *their* customers. At no point was it ever mooted by Castle that Select *would ever* run on Iyonix. When I bought my machine (in 2003) - I knew full well that Select would not be available at that time (or possibly ever). I was happy enough to content myself that I was buying and paying for a more capable and expandible platform than anything Select natively runs on (and funnily enough that still hasn't changed).

Iyonix is a fine machine, the OS it uses is being opened - and that offers a much greater scope for development than Select can offer. Yes work is needed on the User Interface front (where Select has an edge) but there's nothing insurmountable here in my humble opinion.

AMS on 17/7/07 7:30PM
[ Reply | Permalink | Report ]

stevef: Just because something is abstracted from the kernel does not mean it can be run on anything. Things in the most part have been abstracted by creating new interfaces, which RO5's kernel does not have.

rjek on 17/7/07 8:31PM
[ Reply | Permalink | Report ]

I don't see why you all let yourself being provoked into these discussions by these kind of articles. I think the Drobe editors are getting pain in their bellies of thier hilarious laughters when they read what effect they sort.

Just wait and see what comes and (maybe just like I am more and more) when it's not worth it anymore you can completely turn to the dark side one day.

Cheers boys and girls

Ike on 17/7/07 9:14PM
[ Reply | Permalink | Report ]

rjek: "The reasons why the ROL applications can't be ported to RISC OS 5 have been stated time and time again: they rely on new features of the OS to provide their enhancements that CTL's RISC OS does not have."

And independent developers have said time and time again, RISC OS was designed to allow every API to be extended and enhanced via the software vector mechanism, service calls, upcalls. etc, etc. The pseudo technical excuses hold no water, ROL could provide a useful and desirable set of enhanced features to run on top of RO5 and make good money out of it, if they wanted to.

druck on 18/07/07 09:05AM
[ Reply | Permalink | Report ]

Why would they need to make Iyonix select, they make a fair amount of the same money not making it. Time and again PM has told punters if they want Iyonix select they should 'subscribe', of course nothing has come of it. If you keep paying the man to not deliver, why would he bother making it? </aplicablerantaboutthewholeselectscheme>

flibble on 18/07/07 10:48AM
[ Reply | Permalink | Report ]

Peter: The actual wording on the Select Subscription form is: "I am now solely an Iyonix User, I want to subscribe to Select 32 for Iyonix as soon as it is announced".

Does that seem fair to you?

jc on 18/07/07 11:17AM
[ Reply | Permalink | Report ]

Yes, that is fair ... if you are not expected to subscribe to (and pay for) select26 in the mean-time.

Paying for select 26 was the suggestion PM gave in public comments dating back to 2005 (that I was referring too). Though if that has changed, then good, cos it was crap.

flibble on 18/07/07 11:37AM
[ Reply | Permalink | Report ]

I think what ROL need to do, is have some kind of web registration form were Iyonix users can register there interest in having Select for Iyonix. This could also give them options for a full OS replacement Select or just a soft-loadable Select Feature Pack.

sa110 on 18/07/07 11:49AM
[ Reply | Permalink | Report ]

sa110: With respect, that's probably the last thing anyone needs. Frankly, it's about time it was accepted that it's highly unlikely there'll be any version of Select for the Iyonix this side of hell freezing over. No amount of posturing here will prevent that.

To be honest, with the undignified political nonsense between ROL and CTL (not to mention the plethora of broken promises and moved goalposts), and the antics of certain people on usenet, it's a wonder there's anyone left using RISC OS, let alone developing software for it.

jmb on 18/07/07 12:09AM
[ Reply | Permalink | Report ]

druck: The issue is that lots of these new features that Iyonix users want depend either directly or indirectly on changes to the kernel - you can't provide them via the usual APIs.

It's still not clear to me why anybody would *want* the ROL userland "enhancements" - some of the kernel work is worthwhile, but most of the other stuff is ugly and of limited actual use.

rjek on 18/07/07 12:34AM
[ Reply | Permalink | Report ]

Without undue respect to PM, is it time for someone else to head ROL? A new, charismatic face to undertake radical changes and bring some fresh excitement to a company whose product is beginning to look pretty stale tbh. There's been a lot of changes under the hood, yes, but that's not terribly easy to sell to Joe Bloggs.

But I think the platform has got to the stage where using traditional business models isn't going to cut it. The shared-source idea is great, but I don't think it goes far enough.

timephoenix on 18/07/07 12:44AM
[ Reply | Permalink | Report ]

Josh Eiken — I agree. May I make a suggestion?

Paul Beverley is a great ambasssador for RISC OS. I have always found him approachable, non-confrontational and reasonable. He did a great job with Archive magazine, and if he could be persuaded to head ROL now that he has handed over Archive, it would be great. I might even be persuaded to subscribe to select on a strict understanding that I would be receiving applications for my Iyonix.

This is probably fruitless speculation, because I cannot see PM standing down.

AFT on 18/07/07 1:41PM
[ Reply | Permalink | Report ]

Don't propose it here, propose it to ROL's shareholders (a list is available from companies house website), as they are the ones who get to choose.

flibble on 18/07/07 2:25PM
[ Reply | Permalink | Report ]

rjek: the kernel business is nonsense, what can be described as the kernel in pre-RISC OS 5 is a tiny amount of non vectorised processor and machine specific code, absolutely none of which is necessary to the features of Select which would be desirable by users and programmers.

The only thing you can't do in a limited softload is some of reorganisation of code in to separate module which has occurred in Select, mainly on the whim of its chief programmer rather than any sound technical requirements. RISC OS 5 had its own reorganisation to form the HAL abstractions and would not gain anything from such changes.

I'll say it again, there are no valid technical reasons why ROL cannot offer the vast majority of Select features over RISC OS 5 with very little development effort. They would have a potential market of Iyonix users of far greater than the existing 1000 Select subscribers, and they would be earning a considerable amount of cash to feed back in to developing their OS for RPC and A9 owners. Everyone wins, no one looses.

druck on 18/07/07 2:32PM
[ Reply | Permalink | Report ]

I've been reading this thread with a great deal of interest. There seems to an impressive quantity of nonsense on here from people who should know better.

Druck - you may recall we spoke about select for Iyonix at Wakefield and I explained some of the problems- which you agreed with (no I am not going to repeat them here, you ought to be able to remember).

The basic problem is this - with co-operation it can be done. Without co-operation it can still be done, but it will take a lt lot longer. If you lot really want Select for Iyonix (and I don't think some of you do - it's just an excuse to stir things up) then go to Castle and tell them to stop playing silly beggers. I have years worth of e-mails going back and forward between ROL and Castle about this. It's simple. ROL want to do it. Castle don't want ROL to do it.

VirtualAcorn on 18/07/07 3:02PM
[ Reply | Permalink | Report ]

In reply to rjek (a few comments back):

"tbh, I'd have thought the ImageFile stack and the new sprite stuff would be much more complex than the WIMP upgrades. Quite a few things depend on new features in the kernel, where all that's happened for rounded buttons is the abstraction of the WIMP's icon drawing routines to an external module."

I guess the ImageFile stuff is more complex, but just think of ImageFS which does run on IYONIX with no kernel changes - nobody said that ROL doesn't have to amend their code to run on RO5, but quite a few of the Select enhancements like ImageFile* and the apps can be made to run on RO5 since similar or even better features and apps manage.

hzn on 18/07/07 4:50PM
[ Reply | Permalink | Report ]

In reply to flyfot:

"... And, of course, 100 people did get together on a pledge bank, but it was quite ruthlessly ignored. " Well, I must admit that I am not surprised that ROL now claims 130 to be not enough... As I already stated they could simply offer quite a few select features to IYONIX but refuse that despite that being possible with much less effort and hassle opposed to a full OS and thus probably quiker money and a better cost/income ratio.

"... Is it, as I suspect, because CTL just isn't interested in the enthusiasts' market any more? And if this is true, then I hope CTL might realise the damage of their actions." Well what about the damage ROL did with their behaviour: As time passed by the initial milestones and offerings were reduced, select releases delayed more and more but desipite all that the expectation that users keep paying year by year and even blackmail be claiming that if subscribers refuse to re-subscribe they would have to stop. Simply put in the long run you didn't even get 30% of what you initially signed up and paid for (assuming you kept subscribing). And then that "we need 100 IYONIX pledges" which came by but to what end?

From my point of view this is not a serious business behaviour... and thus unless ROL does really offer something of use to me which is there to take away I won't part with money - their track record makes it impossible for me to believe the offerings.

hzn on 18/07/07 5:00PM
[ Reply | Permalink | Report ]

Aaron Timbrell:

Whilst I can understand that it might not have been easy, and that there may still be testing to do, I also understand that ROL have already shown Select features running on an Iyonix at previous shows: [link]

This (amongst other things) is why it's difficult to believe that there are still major technical difficulties, given that some of what people are asking for appears to already have been achieved. My feeling is that it would be great if ROL can make this available.

To be totally clear, I personally don't doubt that the differences between CTL and ROL make things difficult, and I can totally believe that it's not economically sensible to finish Select for Iyonix.

This honestly isn't intended to stir things up. I would personally really like to have the alpha-mask sprites, proper cut and paste in icons, filer-thumbnails, drawfile icons etc. that Select can provide, and would of course be happy to pay for them. But what can be frustrating for an end user like myself is that -- apparently unlike others such as yourself -- we don't have the benefit of knowing the inside dealings that go on between CTL and ROL. The only 'facts' we have to go on are the announcements made by the two companies.

The only thing that seems to be clear is that the situation for Iyonix users remains very unclear. Personally I think it's unfair to blame ROL or CTL for this. The real culprit is the size of the market. But nonetheless it can be frustrating.

flypig on 18/07/07 5:04PM
[ Reply | Permalink | Report ]

In reply to fibble:

"Why would they need to make Iyonix select, they make a fair amount of the same money not making it. Time and again PM has told punters if they want Iyonix select they should 'subscribe', of course nothing has come of it. If you keep paying the man to not deliver, why would he bother making it? "

True - simply put they don't have to. But then PM should simply say so clearly instead of beating around the bush requesting e.g. 100 IYONIX users to say "yes" and then stating "100 is not enough" etc.

hzn on 18/07/07 5:05PM
[ Reply | Permalink | Report ]

In reply to VirtualAcorn:

"If you lot really want Select for Iyonix..."

Well, quite a few do *not* want Select for IYONIX but just some of the Select features. But that is something ROL doesn't seem to want to do.

hzn on 18/07/07 5:07PM
[ Reply | Permalink | Report ]

hzn>The point is with RO5.xx open sourced Iyonix users are *not* strictly dependant on ROL (or Castle - other than their releasing the source to ROOL) in order to improve the OS and add bits. Some of those bits *could* implement User Interface functionality present in Select but absent in RO5. The fact it might be acchieved by grabbing software vectors (on RO5.xx) or using the abstracted kernel features (on Select) matters not a jot. At the end of the day if RO5 has it's current advantages and has an "appearance" similar to Select that would be good enough for most Iyonix users.

AMS on 18/07/07 6:43PM
[ Reply | Permalink | Report ]

Both companies need to converge on something. Select on Iyonix I don't think is realistic as Castle paid heavily for RISC OS. RISC OS 5 shouldn't be made obsolete like that any more than the work on RISC OS 4 (i.e. RISC OS 6). Interoperability is the best in the interim and even that requires communication.Ultimately there as to be convergence in some form.

AW on 18/07/07 8:36PM
[ Reply | Permalink | Report ]

Aaron: we talked about getting RISC OS 6 to run as the exclusive OS on the Iyonix, something which requires information on the Iyonix hardware which Castle haven't supplied, so obviously leads to major difficulties for that route. The discussion now is about providing a range of Select features as a value added product to be used in conjunction with the base OS of RISC OS 5.

Unfortunately ROL seemingly refuse to acknowledge the second option exists and prefer to sit back and prevaricate about why they can't proceed on the first option. An exclusive core OS on Iyonix hardware would be the cleaner and better long term solution, but it also has considerable short term disadvantages given the huge amount of hardware specific driver code which would have to be needlessly duplicated, then support issues for both ROL and Castle while it reaches the same level of maturity as RO5 has over the last 4.5 years. Castle in particular would be unwilling to have to invest resources to investigate problems reported with hardware which turned out to be due to running a 3rd party OS.

Providing some/most/all Select features as softload over RISC OS 5 avoids those issues. If could be done as a single or phased series of releases, whichever suits development resources and revenue potential. This would get Iyonix owners onside and part of the Select scheme and contributing to the development effort, and show Castle that there is benefit to having Select run on the Iyonix. Maybe then the work to get RISC OS 6 can proceed, and without the huge pressure of a big bang release, leading to more stable product.

druck on 19/07/07 09:27AM
[ Reply | Permalink | Report ]

In the meantime, whilst we all squabble about Select, I shall look forward to receiving the new Pixelmator app for my Mac (www.pixelmator.com). Only £60 too. Why bother with Paint?

Eddie on 19/07/07 1:56PM
[ Reply | Permalink | Report ]

In reply to Druck: ROL does have a choice. Either to try and produce a full version of RISC OS 6 for the Iyonix hardware, or to produce a kind of "Select Plus Pack" for Iyonix with some Select features. The first route would probably be a better one to take, but as you have said Castle haven't supplied the information that's been requested.

So looking at a "Plus Pack". Despite what some people maintain a lot of the changes to the RISC OS 6 applications *do* rely on other changes made to RISC OS. If you have a Select4 RISC OS 6 then why not try running the apps on the Iyonix and see what happens. All the disc based components in Select4 are 32bit neutral. However lets look at Paint. This now uses the Image File Renderer, so this needs to be present. This in turn relies on other things being present. It's not just a case of a couple of days work to port things across.

This brings me to another point. Currently everyone is busy working on RISC OS 6 for the A9. This is the next project that needs to be completed. ROL has taken note of people's comments. Select4 has (finally) been released. The OS has been documented. A PRM has been produced that gives developers the information they need. In the past ROL has (in my opinion) been working on to many projects at once. The result being that although a lot of work was being done nothing got up to release standard. This led to various people claiming incorrectly that nothing was being done.

ROL, like all RISC OS companies has limited resources. In my view these resources are best being currently targetted at the A9, a full 32bit platform. Once RISC OS 6 is released on one 32bit platform that is the time to look again at the Iyonix. It is not good time management o divert people of the A9 at the moment.

In reply to flypig:

ROL has shown an Alpha version of some Select features running on the Iyonix. However what's been shown is Alpha quality and is nowhere near good enough to be released. What ROL have shown is a test build, which was produced in order to evaluate the project, both commercially and technically. I would like nothing more that to see RISC OS 6 running across all machines on the platform, but it can't be done immediately. There is no magic button that can be clicked. People have to do the work and at the moment they are busy on RISC OS 6 for the A9. Once that is completed then what has been done can then be applied to the Iyonix.

VirtualAcorn on 20/07/07 11:37AM
[ Reply | Permalink | Report ]

How predictable: I got modded down. It's symptomatic for the state of the RISC OS market.

Look how vulnerable we are; when a developer runs into a train we have a serious problem. Hope this will not happen to the few that keep the little software we have alive.

But then again; they get older too and will eventually die, hopefully of old age.

Cheers.

PS related to this thread a song of good old doctor Hook & the Medicine show springs to mind: the freaker's ball which has some lines that say " the greatest of the sadists and the masochists too are screaming please hit me and I'll hit you". That's the state of RISC OS today!

Ike on 21/07/07 10:55PM
[ Reply | Permalink | Report ]

in reply top Ediie Lord

An even batter app is Pixel32. Been in development for years and your license covers YOU to run it on any machine you want. I have it on WinXP, Linux and MacOSX dual boot on all.

see [link]

If only we had this for RiScOs :-)

nijinsky on 24/07/07 10:10AM
[ Reply | Permalink | Report ]

Eddie: > Why bother with Paint?

Maybe because Pixelmator, like Iyonix Select, hasn't actually been released yet?

richcheng on 30/07/07 2:14PM
[ 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

Select preview
Every feature in Select 1 covered, what's in, what's out, what's it look like and what do we think of it?

9 comments, latest by diomus on 15/3/03 1:10AM. Published: 31 Oct 2001

Random article

New mailing list for Omega users
Wanna be in my gang?

16 comments, latest by ian.l on 24/8/04 5:14PM. Published: 23 Aug 2004

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
"D underheads R adicals O r B loody E xcentrics"
Page generated in 0.4065 seconds.