Interview with Paul Middleton
Conducted by Paul Johnson
An interview with Paul Middleton (MD of RISC OS Ltd) and Paul F Johnson conducted via email just prior to the Wakefield show.
PFJ: Paul, how long have you been involved with the RISC OS market as either boss of RISCOS Ltd or any other guise?
PM: I have been involved with the RISC OS market since January 1990. i.e just over twelve years. I started Uniqueway with my brother Simon in April 1990, initially as an Acorn ISV. We first developed a Teletext subtitle preparation system, and then in 1991 we were invited to become Acorn Dealers for South Wales. We then worked with Acorn on the development of the Acorn Replay system and produced the high quality movies found on just about every CD that featured Replay movies. We were Xemplar Education agents for 2 years and then I became the leader of the team negotiating with Acorn from September 1998 onwards to take over development and marketing of RISC OS for the desktop market. Eventually this led to my becoming the MD of RISCOS Ltd in January 1999.
PFJ: In that time, what would you consider to be the most significant change to the RISC OS market?
PM: There have been a number of significant changes at various stages in Acorn's history.
Acorn had some very amazing engineers who, long before Apple used the phrase, definitely "Thought Differently." Acorn always had people with some great ideas, but more often than not they were unable to capitalise on those ideas, simply because it seemed they were restricted by Olivetti who controlled the purse strings. Acorn didn't seem to be able to focus on the commercially viable projects that would bring in revenue, rather than the Blue Sky projects that didn't.
The usual problem was that Acorn simply did not have the money to fully develop the ideas that they came up with into commercial products. That may seem a strange thing to say, but Acorn were not really a major player in the desktop market and many of their ideas were too far ahead of their time. There were many other much larger companies vying for a place in the PC market and when even Olivetti themselves couldn't make it as a major PC supplier then what hope did Acorn have? So the most significant change for me came when Acorn were finally freed of the ties with Olivetti and went public as a PLC. That gave them the funding to potentially exploit the opportunities that were available at that time. The problem was that they were then just beginning to face a major challenge by the Wintel camp on their dominance of the Education market which proved impossible to counter.
However going public was the eventual demise of the company as it enabled the corporate sharks to dismember the company in 1999 for the value of the shares it held in ARM.
PFJ: Given the advances in processor technology, how important would you say it is that we are able to make use of the likes of the X-Scale, ARM 9s and the such as a computing platform?
PM: The choice of processor is always dependent on the target market for the final product.
ARM processors have traditionally been aimed at low cost, low power devices. Acorn were the only people that used the ARM processor as a desktop computer, because RISC OS was the only Operating System that could effectively function on the 8 MHz-50 MHz processors that were common before the advent of the 200 MHz StrongARM. RISC OS still provides a very responsive system on a 233 MHz StrongARM and for most people, the only things they would probably like to change on their Risc PCs would be built-in support for more, faster hard drives, and bigger screen modes. Acorn grouped the Risc PC under their Workstations division and a Risc PC does make a very good general purpose workstation. However RISC OS is very much tied to the ARM processor and should not try to be something that it is not. i.e if you want to manipulate large maths intensive spreadsheets then a PC processor will be much better at the job. If you want a low power computer then you use an ARM chip.
Any new faster processor such as the XScale would be nice but the price and availability of those chips have to be appropriate as well. With the pace of current processor developments there is every possibility that a 64 bit ARM processor may become available in the next couple of years, thus making the current discussions on 32 bit versions of RISC OS, and which processor to use, irrelevant.
In the meantime the StrongARM SA1110 processor still offers 26 bit processor modes and significant speed advantages over the current SA110, without needing the change to a 32 bit version of RISC OS.
By the way, it is very interesting to note that Microsoft have decided that Pocket PC 2002 (a.k.a Windows CE) will now be available solely on the ARM platform. There is no doubt that on identical hardware that RISC OS will be much more responsive than Windows, so all we need now is the appropriate hardware and a manufacturer who wants to use RISC OS to justify future development of RISC OS for new processors. .
PFJ: If you could make one major change to the current operating system, what would it be?
PM: The one change I would like to make would be to implement Universal Cut and Paste / Drag and Drop between any writable icon.
PFJ: Would you say that the long term survival of the platform lies in the embedded systems rather than it's current desktop market?
PM: RISC OS has always relied on products outside the desktop market to justify its existence. Currently many more RISC OS desktop machines are sold that end up "embedded" than are actually used as desktop machines.
The desktop market is the visible face of RISC OS, but without the other markets then the RISC OS market would not be viable.
As far as the UK is concerned we have to accept that we do not have the political clout to force the Government to change its relationship with Microsoft and hence the dominance of the PC platform. I had the chance to talk with one of the UK e-envoys recently and asked him about the Government's attitude towards different operating systems. He stated that the Government had a policy to ensure that their services could be accessed by any operating system and that their web sites should not be tied to any specific OS. As regards any other support for specific Operating Systems it seems that the Government are officially agnostic.
However in the Education market the promotion of Interactive Learning Systems, which are only available on PC or Mac systems, and the discounts that Microsoft give to schools make their position almost unassailable whilst schools have dedicated computer rooms. If the Government plan of one computer per child was ever to reach reality then RISC OS might once again have a chance to compete for the provision of basic computing services. The Ron project (RISC OS on netBook) for example would enable the provision of a portable computer with a vast back catalogue of education software, at cost much lower than a normal PC portable could match. £500 seems to be the magic price that schools are looking for in order to make the "One per child" goal achievable. .
PFJ: Would say that one of the main strengths of RISC OS hardware has also been it's biggest weakness in that the amount of backwards compatibility has meant that for the large part of the time, people have not required to upgrade their machines as most software will run fine (if not a tad slowly) on an A5000 that will also work on a RPC (assuming they've not made any use of the likes of >OS 3.5 specific parts of the OS such as dynamic areas).
PM: For the Education market backwards compatibility was considered essential as schools were used to expecting that Acorn computers would last at least 5 - 7 years and that as new models came out that they would be compatible with existing software. Whilst the growth potential of schools and the home market was still enormous Acorn were happy to still maintain that compatibility as they didn't want to alienate their major customers. However the net result was that schools didn't need to upgrade software and the long life of the hardware also meant that schools didn't need to upgrade their hardware. If schools had been forced to upgrade to Risc PCs in 1995 then we wouldn't have been left with a legacy of schools still using RISC OS 3 computers here in 2002 and Acorn might still be around.
PFJ: Is it fair to say that from various corners, the Select scheme has come in for a lot of criticism.
PM: It is interesting to note that virtually all the people who have criticised the Select Scheme have done so without being subscribers, so their complaints are rather hollow. You yourself Paul fall into that category. The vast majority of the subscribers to the scheme have been very happy with it, and they are the ones that count.
PFJ: Two of the large criticisms has been the delay over a version of the Microdigital and RiscStation ranges. Roy has stated that there is a version for the RiscStation in development, can you say if the same applies for the Mico and what the main delay has been over the CDs?
PM: Again the criticism seems to be rather one sided. I believe that there are only three people who have contacted us wanting to know when there will be a RiscStation compatible version of Select, and only one enquiring about a Mico version.
The RiscStation version exists but needs more work before it can really be used as a publicly released product. We are awaiting information from Simtec to enable that to happen. The Mico version may follow at some stage.
The CD was held back until the Installer was fully developed to ensure that the upgrade from RISC OS 4.0x to Select was a smooth as possible. .
PFJ: Can you counter the comment made that vital information for developers such as changes to the !Boot structure should have been made available to all developers, not just those subscribed to Select? There have been instances of software breaking due to the disappearance of PreDesk.
PM: Unfortunately we are running a business not a charity. Preparing any form of documentation is an expensive process. Acorn didn't used to give away PRMs. If you wanted to learn about RISC OS you had to buy the Manuals. They didn't give them away for free, unless you were one of the small group of "key" developers.
We offer a slightly better deal in that you will get the OS and the manuals by joining Select. If you had joined Select yourself Paul, you would have known that PreDesk has not disappeared. It has been relocated to allow for multiple user setups. Any application which has obeyed the long established guidelines for writing their choices to <Choices$Write> does not have any problems. Only those that mistakenly write explicitly to Boot.Choices have problems with the new Select Boot sequence.
PFJ: How damaging has the announcements of the likes of the Evolution and Omega been on the sales of current desktop machines and confidence as a whole?
PM: Back in 1998 Senior Acorn Managers vowed that they would never again pre-announce a product because of the bad publicity over the delays to Phoebe. Before then, Acorn had always been very secretive about new products. Apart from "key" developers, the first time that most dealers and developers got to hear about a new product was on the day it was put on sale. That way there were no false hopes raised and users could not be disappointed by delays between a product being announced and the time when it was actually available on sale.
I can understand the desire of manufacturers to want to announce a new product as soon as possible in order to steal a march on their competitors. However there comes a point when a very early announcement of a new product can have a negative effect on potential customers, who feel let down by excessive delays and lack of news about product availability. Ultimately delays will have a negative effect on the company concerned and will inevitably make its customers much more wary about wanting to wait for future products.
In a niche market such as ours it is very important to keep user confidence by releasing products on schedule as we have always done at RISCOS Ltd. It is very unfortunate for everyone that some other companies have been unable to deliver products in a timely manner and I hope that those delays do not have a long term effect on the market.
PFJ: What is the state of play over updates to the Acorn C package? The last that was publically told (at MAUG last year) was there was to be a large update around November.
PM: For reasons outside our control there has not been any progress on that. We have had discussions with ARM over the work that they are doing with their new development tools to see if they can be used for future RISC OS projects, but I can't give any more details at present.
PFJ: Will the PRM for RISC OS 4 be made a publically available document (for purchase) and if it is, can you give us an indication on when it will be available?
PM: Yes it will be. We do not have a timescale for that at present though as there is still a lot of work in order to fully document the code for public release.
PFJ: Would you care to comment over the reports which have been made (on usenet, iconbar and foundation newsletters) that VirtualAcorn is shipping OS 3.1 illegally? Surely a product such as VA will help bring people back to the fold (there have been many instances of this happening and a rise in the price of second hand RISC OS machines since VA hit the streets)?
PM: VirtualAcorn claim to have an Agreement with Pace which as far as we are concerned is contrary to our exclusive Agreement for the sale and development of RISC OS in the desktop market. However since Virtual Acorn claim to have an Agreement with Pace it is ultimately up to Pace to confirm or deny that claim. The fact that they have made no comment as yet, should not be taken to imply that the VirtualAcorn claim to have an Agreement is correct.
I do not believe that Virtual Acorn is helping the RISC OS hardware developers sell more products. It may bring more people into using RISC OS 3.11, but at present I very much doubt that anyone who used to have an A5000, and has then moved over to using a PC will be tempted back to buy a RISC OS computer because of Virtual Acorn. I very much doubt that an existing PC user who discovers RISC OS for the first time will then go out and buy a RISC OS computer. I would also point out that any profit that Virtual Acorn is making is solely going into the pockets of those running VA it is not furthering future development of RISC OS, or encouraging people to buy a new RISC OS computer.
(I cannot let this one through without a few comments - firstly as a former Development Director of RISC OS Ltd I should point out that I am fully aware of the RISC OS Ltd licence, and that VirtualAcorn in no way breaks any parts of RISC OS Ltd's rather limited non-exclusive licence. Secondly we have a number of examples of customers who have used VirtualAcorn and then gone and bought a real computer. Thirdly we do indeed profit from VirtualAcorn, but then to quote Mr Middleton from earlier in his interview "Unfortunately we are running a business not a charity". - ED)
PFJ: In your opinion, what would be the most productive way end users can support and maintain the platform?
PM: The obvious form of support is to buy our products and those of the other companies that are supporting RISC OS development, rather than those that are encouraging software piracy.
PFJ: Is RON still an ongoing project (as opposed to a suspended project) and if it is, then when will we be able to see it?
PM: Yes, Ron is still ongoing, but the requirement from the original funder has changed.
We are currently waiting to see what Psion are going to do with their next product release before a decision is taken as to whether a full desktop release of Ron is viable. At present I would have to say that it is unlikely that a full version of RISC OS will become available for the current Psion netBook. That is not to say that future Psion products could not be supported.
PFJ: Finally, do you believe that the key to the future of the market is for all developers (both hardware and software) to get along and work together rather than the clash of personalities we currently have?
PM: In any market, clashes of personalities are inevitably destructive. However the RISC OS market as it currently stands is essentially an enthusiasts market, which means that nearly all the companies involved are very small and consequently there is much rivalry between certain companies. I would very much like to see more co-operation between developers, but it really needs a much larger organisation than current exists within any of the current companies in order to make greater all round co-operation between everyone a reality. It also needs an end to the destructive attitude of certain people on the newsgroups and web sites who have personal grudges against specific companies and/or individuals.
Large organisations don't want to deal with a sector where specific companies have followings of people who waste everyone's time by constantly complaining and arguing over everything. They want to deal with a market where they see a co-operative supporting environment which enables them to get their products to market in the shortest time and with the least amount of hassle.
Conducted by Paul Johnson