RISC World

Does RISC OS have a future

Dave Holden

Does RISC OS have a future?

RISC OS is in decline. You may not like that statement, buts it's the truth. RISC OS has been in decline for years and as a result RISC OS dealers and developers are an endangered species. If dealers and developers had fur or feathers there would probably be laws to protect them.

Understanding and accepting the decline of RISC OS is essential. If we stick our heads in the sand then this decline could become terminal, and I don't mean this could happen in five or ten years, I mean within the next 18 months.

So what can we do to reduce this decline? Can we do anything at all? Probably, but the first thing we have to do is try to work out what the causes are before we can even begin to do anything about them.

The atmosphere

One thing that I don't think is helpful is the group that I call the 'RISC OS Fundamentalists'. If you like, the 'militant wing' of the RISC OS community. These are the people who say that you can do everything using RISC OS, Windows is the invention of Satan and anyone who uses it, let alone finds anything good to say about it, should be vilified and excommunicated from the RISC OS world. Anything that can't be done using RISC OS or that can be done easier or better on another platform they don't want to do anyway, and, by extension, anyone that does want to do these things, or is even prepared to admit that they might exists, is a subject for attack.

Most ordinary users know better, and all this attitude achieves is to alienate them and help drive them away from RISC OS. In fact, the very (and vociferous) existence of these people is a symptom of the decline. As the more sensible figures give up and depart so the field is left open for these people whose rants are no longer balanced and countered by a much larger number of more moderate voices.

The basic facts

But why is RISC OS in decline? It can be summed up in one word - numbers. If the number of users falls then the sales of any product on the platform will fall. .

It doesn't cost any less to develop a project that will sell 500 units than it does if it's going to sell 5 million. In either case the development cost must be recovered, and as each item you sell has to carry a part of this development cost it naturally follows that the fewer items sold the greater the proportion of this burden must be applied to each unit. In other words, the less you sell the more they're going to cost.

If the price is higher, people are discouraged from purchasing it. Therefore you sell even fewer units and we're into a downward spiral. The financial risk is increased and the potential profits are reduced. So it makes sense to develop for a platform with lots of users. .

To put it another way - you have to be financially naive (or nuts) to develop for RISC OS.

If we are going to even slow the decline in the numbers of people using RISC OS, let alone reverse it, we need to develop new strategies because it's self evident that the existing ones aren't working.

The new machines, the Iyonix and Omega, do nothing to address this problem. That's not a criticism, it's just facing the facts. Almost everyone who has, or who will buy, one is already a RISC OS user, and not just a user but already has something like a Strong ARM RiscPC. Moving someone from a RiscPC to an Omega or Iyonix doesn't do anything to swell the user base. It may encourage purchasers to spend a few pounds on software they might not otherwise have bought, but if they have 'top end' machines already then they probably will already have most of the software they require and will simply transfer it.

RISC OS in education

Whatever you may have heard to the contrary there is almost no chance of people abandoning their Windows machines and migrating to RISC OS. It's true there's a lot of anti-Microsoft feeling, especially as the cost of licensing software such as Microsoft Office in a school or company can make quite a hole in the budget.

However there are other strategies that customers are adopting, such as moving to Open Office or Star office, or even migrating completely to Linux. They are not going to shift back to RISC OS machines.

Why? One obvious reason why this isn't going to happen is because of the high initial purchase price. There are very good reasons why these machines are comparatively expensive, and I don't want to go into that here, but the main one, as discussed previously, comes back to numbers. RISC OS computers sell in almost insignificant numbers, so they are inevitably a lot more expensive than something that sells by the million.

Moreover, although we are often told about the 'low cost of ownership' of RISC OS machines used in education this comparison does not really bear scrutiny. It only had validity when applied to a school which already possessed a large number of RISC OS machines and appropriate software. Once a school has disposed of its RISC OS computers and moved to Windows PCs then there is absolutely no valid financial argument for changing back. If a school does decide that it wants to cut its costs and abandon Windows the logical step is to move to Linux. This is no more traumatic than abandoning Windows in favour of RISC OS and the costs are comparatively minuscule since it involves the purchase of no new hardware.

Anyone who tries to get a school to abandon hundreds of thousands of pounds worth of hardware and take what will be seen as a huge leap in the dark and purchase expensive new machines with a comparatively unknown OS which has a dubious support policy, little software, no decent web browser, and which will severely restrict their future options is simply not going to be listened to.

The computer is not RISC OS

However, what we are, or should be, primarily interested in is not the numbers of dedicated RISC OS computers but the number of RISC OS users. We need either to get new users, or to get back users who have now left the platform. The ideal is that people will be using RISC OS on dedicated hardware, the most important thing is that they are using it. Whether they have an A7000, RiscPC, Omega, Iyonix, Windows PC or crystal ball doesn't matter, what matters is that they are using RISC OS.

People are not going to abandon their PCs to return to RISC OS. Neither will they, at least initially, buy a RISC OS computer that may or may not be suitable for their needs but which they can plainly see is going to cost them more than double what a good Windows PC will cost. We need some lateral thinking.

Breaking the im0passe

So how do we persuade these potential customers that they should even look at RISC OS? Make the computers cheaper? As we have already discussed the computers can't be made any cheaper as the development budget is split among such a small number of units. However we are interested in RISC OS users, so can we sell PC owners a copy of RISC OS?

The only way to do this is to supply emulation software on their PC. It's the only way they are going to be able to keep what they already have and get something extra. We are not asking them to abandon what they already have and are familiar with, whatever it's faults, neither are we requiring them to spend a huge amount of money. We are not promising a magic solution to all their computing problems. RISC OS emulation on a Windows PC just gives them more options without requiring them to give up anything.

RISC OS under emulation is also an ideal way back into education. It requires minimal expenditure for the school, just another 'software Licence', and once installed allows RISC OS dealers to sell RISC OS educational software to the school. In many cases where the school previously used RISC OS computers they may already have software available, which is obviously an incentive for them to run RISC OS under emulation, so there's no immediate or obvious gain the dealers. However, the school then becomes a potential customer, and that's better than if they were just running Windows because that would mean there is no possibility of selling them RISC OS software.

RISC OS under emulation is the only hope we have of increasing the RISC OS userbase. Without it I would anticipate that within a year or so the number of users will have declined to the point where they will no longer be able to support a viable RISC OS industry. Even now you can count the number of dedicated full time developers on the fingers of one hand (without needing to use thumbs), and if things go on as they are then it won't be long before even they have vanished.

Future development

Lets apply this directly to my own company, APDL. At present we own more software titles than anyone else. In fact, we probably produce more commercial software titles than everyone else put together. We actually have a lot more stuff than appears on the web site because there are many programs that require work before they can be re-launched. We have a major development program to buy more rights, and, just as important, to update and improve much of our software. This program has been effectively put on hold or drastically slowed down for the past several months because of my worries over the future of the RISC OS market. I'm not prepared to commit tens of thousands of pounds to this development program unless I can see some realistic likelihood that there's going to be a market there. The only way I can see that there will be a market which will make my investment viable is if we can continue to expand the RISC OS user base by way of emulation.

Virtual A5000 has proven that the product is viable and shown what can be done, even with an obsolete OS. We need better emulators, using later versions of RISC OS, so that the package is more attractive and modern and a better representation of RISC OS as it is today (that is, a better 'advert' for RISC OS) and able to run the latest software. If we want to develop the market, and I mean the RISC OS market in general, we now need to move on.

Without a true RISC OS marketplace the OS has no future. A few enthusiasts writing freeware or porting various odds and ends does not make a 'market'. If there is to be any hope of selling the new hardware products the prospective purchasers need to be able to see dealers and developers who are available to supply software and support them. No amount of rhetoric or enthusiasts on newsgoups will make them spend twice the price of a good PC on a RISC OS machine unless they can see a thriving marketplace.

It's the old 'chicken and egg' situation that has prevailed since the earliest days of computers. Before people would buy a particular computer they wanted to see that there was plenty of software; before anyone would write software they wanted to see a sufficient number of computers sold. We're now looking at the 'other end' of this equation. Before developers will commit to work on existing software, much less write new stuff, we need to have some confidence that there's going to be a market there to sell to; before any potential new user can be persuaded to move to RISC OS or existing users encouraged to stay they need some assurance that the developers will be doing this work. The only way to break this cycle is to get more users, and the only way I can see to get more users is emulation software.

The criticisms

One of the things that the 'experts' have said is that people don't buy software to run under emulation. Rubbish! None of these people are commercial developers or have any idea of the true picture. There are few enough developers remaining for RISC OS, but if you talk to any of those that are left, for example, Icon Technology, ExpLAN, R-Comp, or, of course, APDL, we will tell you how important we believe this development is. Even Gerald Fitton, who distributes the Colton Software range (PipeDream, FireWorkz, etc.) and who was initially highly sceptical has now produced a software CD especially for Virtual Acorn users.

It's not just users that Virtual Acorn is bringing back., who produced the well known programs Personal Accounts and Prophet and who abandoned RISC OS for Windows some years ago recently began to produce a special version of Prophet for Virtual Acorn. They attended the Wakefield show earlier this year to promote this, and so successful was it that they have now asked APDL to hand back Personal Accounts, which we have been distributing for the past two years. So a valuable ex-RISC OS developer who had previously left the scene because of falling sales has now been attracted back.

So far there have been two commercial products that have used emulation technology to run RISC OS on PC based hardware. One is the free standing VirtualA5000 emulator, the other is the MicroDigital Alpha. One of the criticisms of the Alpha is that it takes money 'out of the market'. Again, it's only people who have no idea of how business actually works who could possibly say this. A RISC OS dealer actually makes similar money selling an Alpha or an Iyonix, and in both cases the majority of the cost of the hardware goes to purchasing from 'outsiders'. Remember that Acorn RiscPC motherboards weren't made in the UK. The significant point here is the money that goes to dealers. If the dealers are able to supply a product that their customers want and that makes them some money then they can stay in business. Without dealers we don't have a viable market.

(I have been reading some right old rubbish on this theme recently. To set the record straight RISCOS Ltd makes more money on each Alpha than MicroSoft does - ED).

A further criticism is that sales of emulation software will affect sales of 'real' hardware. As I've already said, I believe that what is important is the number of RISC OS users, not the number of people who are running RISC OS on ARM powered hardware. If a software product costing a few tens of pounds can out-perform a computer costing over £1,300 then don't blame the low cost software. Logically, which product would you think has the better chance of expanding the RISC OS user base?

In reality, of course, there are still many advantages in using the 'real' hardware, and any true enthusiast will almost certainly not be satisfied with emulation. However, there are many people who need to use Windows and who, for a variety of reason, don't want two separate machines. For these people dedicated hardware is not an option, so should we just write them off?

Another argument, from the rabid anti-Windows protagonists, is that Windows is so terrible that no-one should be exposed to it, so RISC OS emulation under Windows is a 'very bad thing'. Windows crashes all the time, nothing works properly, you'll be plagued by viruses etc., etc. If Windows (and particularly Windows XP, which seems to be a favourite target) is so bad then inevitably exposing Windows users to a viable alternative must be a good thing? And surely similarly allowing RISC OS users to get some experience of the (alleged) horrors of Windows will encourage them to stay with RISC OS? .

Of course, the argument disintegrates completely if Windows isn't quite so bad as they say. Perhaps this is one of the reasons why so many people seem to have abandoned RISC OS in its favour, whatever it's faults, and hurling insults will not make them change their minds. Giving them decent emulation software might help them remember some of the benefits of RISC OS and perhaps keep in touch.

The Solution

The biggest problem at the moment is that emulation has reached a crisis point. If it's going to help the RISC OS market to expand and move forward then the software itself, or, more accurately, the machines it emulates, have to move up a gear. This means Virtual RiscPC. The software used on the Alpha portable shows the way, but (for various reasons) that is tied to specific hardware. To reach the necessary wider audience we need a software product at a sensible price. .

If it is going to be priced so as to encourage people to come back to RISC OS or, if they are new users, to give it a try, then RISC OS 4, as used on the Alpha, is probably not sensible. The logical way to proceed is to use RISC OS 3.7, which would enable the product to be sold at an attractive price. If there is a demand for later versions of the OS then this type of upgrade could be supplied by the people best placed to do so, RISCOS Ltd, which would bring in further money to go towards development of RISC OS itself.

VirtualAcorn are in discussions about the licensing of RISC OS 3.7 with RISC OS Ltd. The recent changes to the ownership of RISC OS present an excellent opportunity for those involved. VirtualAcorn can be used to halt and reverse the decline in numbers of RISC OS users. This will be good for the users, good for the software developers and good for the dealers. Both MicroDigital and RISCOS Ltd have embraced the opportunities presented by VirtualAcorn, but in my view time is running out. If the decline isn't halted very soon then I predict that the future for RISC OS will be very black, although perhaps mercifully short.

Dave Holden