RISC World

Letters Page

They are made from words, and letters too, and some of them, really come from you...

Once again the RISCWorld, whoops, sorry, that's Foundation RISCWorld editorial address has been very busy with a number of topics under discussion. If you have written in and your letter hasn't been published this time then don't fret it will probably be appearing in the next issue. Now with that our of the way lets move on to this...

Dear Sir,
I hope this will reach you in time before the 14th May deadline.
Having bought the RISC World DVD at Wakefield this year I've only just noticed Matthew Thompson's and Paul Brett's mention of the BBC/RISC OS text adventure "Dominion". Paul makes the comment that he would like to have contacted me about some aspects of the game and I'm sorry I didn't read this earlier. He's got a good point in suggesting authors should put contact details on their website and so there is now an email up on the web-page at and I would welcome any questions or suggestions about the game from anybody.
Also, I'm happy to offer a map of the game and unlimited email help from anybody who's stuck or who would like to read the climactic ending!
Yours Sincerely,
Andrew Weston

Thanks for the e-mail. Sorry but your letter didn't arrive in time to make it into the last issue. I am please that you've added a contact link to the website. It's one of those things that's easily overlooked. I know because I have done exactly the same in the past. Given the small nature of the RISC OS market I think it's essential that everyone can get in touch with software authors, either to suggest improvements or to ask for help. Whilst on the subject of help thanks very much for the offer of assistance to any Foundation RISCWorld readers who get stuck in the game. It's this level of support that marks out RISC OS as a friendly community.

Dear Aaron,
I've been reading all the comments about Select but there really does seem to be so much confusion and conflicting comments being made, one minute I read that someone has had a subscription letter saying "Iyonix features hopefully in 2008" then read on the newsgroup that Paul said "it won't ever happen" , so what was all the we need 100 people to sign up for Select all about? and what about people who have continued to put money into Select in the hope of an Iyonix port only to read the above?
I understand that ROL need some info from Castle about the Iyonix which is understandable being that it has different things in it, but what exactly are these problems? cannot ROOL assist with it if ROL ask them ? maybe the source code being released can resolve these problems?
I can't see any reason why Select couldn't work on the Iyonix, obviously it needs time and money and work to be done, but it surely can be achieved with a bit of teamwork etc? As someone pointed out if ROL do not intend (or can't) do a RO 5 Select then maybe the developers who use OS5 and are interested in enhancing RO 5 could then work on it instead, no point them doing the same thing as ROL could be about to release!
Hopefully that all makes sense, I think someone from ROL needs to come out and state the facts of what is really happening and why hasn't and so on. I know what is posted on the newsgroups is often drivel, so thats why I thought i'd ask you, someone who knows the real situation rather than thinking they do!
Matt Thompson

The situation is very simple. ROL tried to get a version of Select running on the Iyonix. This "prototype" was shown at various venues (eg. Wakefield a few years ago). However it was very unstable and not suitable for a release.

ROL asked Castle for assistance with this to get it working properly. Castle then asked for many thousands of pounds to "help". As an aside it's worth pointing out that all the other manufacturers have assisted ROL with getting RISCOS to run on their hardware. Both VirtualAcorn and Advantage6, as well as RISCStation have contributed (both financially and with technical help) to make sure Select runs on the relevant hardware.

This situation has remained the same for some time. In 2004 Castle and ROL signed a new agreement that would have got Select onto the Iyonix. Castle broke this new agreement within a very short space of time. ROL then tried to go through the dispute procedure that was in the new agreement. This required both parties to submit information to an independent third party who would make a binding decision. Castle refused to comply with their own terms and as such the agreement was held to be null and void and ROL reverted to running under the original agreement that had been negotiated with Acorn/Element 14.

So it's been stalemate for some time. However last year ROL and ROOL had a long meeting (I know as I was there) where we discussed how we could co-operate and avoid duplicated effort. A lot of this discussion is covered by an NDA so I can't say a great deal. Now ROOL might be able to assist ROL with the problems with Select on the Iyonix, but at the time the relevant sections of RISC OS 5 were not available and ROOL had no idea when they might have them. I believe that they have now been subsequently released.

At the Wakefield show 2008 ROOL released some more code and this *might* have the solution to the problems with Select on the Iyonix. In my view Castle still don't want ROL to get RISC OS 6 running on their hardware, but, they now have little (or no) control over the situation and the relationship between ROOL and Castle doesn't seem to be especially close anymore.

As a final point it's worth me pointing out that the facts have been repeated ad-nausuem by myself, Dave Holden and others, but nobody seems to want to listen. As for the "confusion" over Select the only "information" that matters is what ROL officially say. Secondhand "information" on the newsgroups isn't to be relied upon. Only official statements from ROL have any weight.

Now for an on-going discussion. You may remember from the previous issue that Michael Poole we having problems with reading the latest Foundation RISCWorld CDs. I asked him how old his CD drive was and what speed it was...

Dear Aaron,
Sorry for being slow replying - flat tack at work. The old IBM drive is 6x.
Michael Poole

I suspect that's the problem. Old drives sometimes have problems with modern CDR discs - so the new disc might not work (hopefully it will) but the type of CDRs that can be obtained vary and we are at the mercy of the market. We did purchase 10,000 CDRs of a particular make that did work in older drives, but we have now used all of these. Given the price of CD-ROM drives it's not worth carrying on with this old 6x unit as it's only going to cause problems again. APDL can sell a replacement drive with a new version of CDFS. Overseas shipping shouldn't be too expensive.

Now lets return to Matt Thompson's earlier letter as, having read my reply, he had this to add...

Dear Aaron,
From what you've told me I think it's worth putting things in black and white and in big capitals. From my point of view all the fingers are pointing at ROL for not doing a version of Select for Iyonix, when Castle are clearly to blame for not helping but they never seem to get criticised.
Personally my view is that it's all politics and people being petty with each other, the market is too small to all be in-fighting, they should be helping each other out. The OS fork has caused unnecessary problems and something we didn't really need either, new machine yes, but not new variation of the OS. So where does Castles future lie? Surely they can't be making that much money, where as the appearances of RO6 and Select will generate income, plus also the Foundation if they bother to reply to people's emails!
I do believe that should Select appear on the Iyonix people will buy it and support ROL, after all most of us coughed up when the scheme was first started and only dropped out due to machine change, I would still be in there if I hadn't no question. Time will tell on this matter but the appearance of ROOL , and if they work with ROL then that hopefully will produce results as RO5 does need developing and Castle aren't doing anything about it.
Matt Thompson

I think you a right, it's all bloody silly, I'm not sure if you could call it 'politics', just daft. The Iyonix is the best piece of RISC OS hardware ever released. RISC OS 6 is the best version of RISC OS. If we are to try and take on the big boys then it makes total sense to do so with the best possible product, which would be a combination of the two. What's interesting is that if you read the Peter Bondar interview in this issue this situation is exactly what he was worried about 10 years ago. The market is indeed too small for in-fighting.

Luckily pretty much all of the nonsense is in the past. RISCOS Ltd and RISCOS Open Ltd have a good working relationship with none of the "history" that exists between ROL and Castle. So that's good news. The remaining problem is one of resources. ROL is a small company with limited resources. A version of RISC OS 6 for the Iyonix would need to re-coup its development cost. If we go back to the 'pledge' that was set up a couple of years ago there were just over 100 people prepared to pay for Select for the Iyonix. If each one of those were still prepared to pay £100, then that would give a budget of just over £10,000, which, lets face it, is a drop in the ocean. Or, to put it in perspective, considerably less that Peter Naulls charged the community for his FireFox port.

I still firmly believe that it can be done, but it won't be easy and, as we have seen, it certainly won't be quick. Select for the Iyonix is going to have to be fitted in between other jobs and done as a set of targeted releases. As for Castle, well who knows that they are up to, but I haven't seen any press releases, announcements, updates or indeed any activity for a year. The company should be filing its accounts at Companies House by the end of the month and that should give us all a better idea of what its been up to.

Now we have the latest production from Martin Carrudus...

Hi Aaron,
Please find attached my latest offering: !Oyez .It grew out of waking up early, sometimes six in the morning, thinking it was about time to get up, when my alarm was in fact set for eight. I wanted regular time announcements during the night, rather like the old crier that called out the time. I actual fact, I wouldn't hear them, as I am dead to the World in sleep!
Test it out. You can choose an alarm announcement, and there is a 'Snooze' feature as well. It might also be useful if you're plonking away on your computer and have some appointment to meet. It will give you regular time checks, then prompts you when you have to tear yourself away from the computer. I have also supplied my source C code. Spot the redundant bits of code, and don't be offended by the list of rude words in it!
BTW Thanks for including !Persian on the latest subscription disc. It's not very sophisticated, but it is fun!
Martin Carradus

Many thanks. I have included a copy of !Oyez in the Software directory. It's quite an interesting application and demonstrates the use of the Speech module very well, although I think that speaking the contents of the info window ought to be an option, rather than by default. Having said that I can see a number of uses for this already. As an idea it might be worth considering adding some 'automated' functions so that !Oyez could run a particular file when an alarm went off, for example to start a backup process. Or how about interfacing it with the Iyonix sound sampler so that you could control it with speech? So you could shout at it in the morning to turn the alarm to snooze? Just a couple of silly ideas.

To contact the RISCWorld letters page please e-mail us using the following e-mail address. The deadline for letters being published in the next issue is the 10th of September.

Aaron Timbrell