Is there any chance you'd be able to get sales figures from the companies selling RISC OS computers? No need to publish individual figures but a total figure for the past year would be great.

I've been advocating the need for bidding developers to have some information about the size of the market for a few years but since Acorn's demise and the appearance of new competing companies it may be easier to get the companies to give out some approximate figures since they're likely to want to know the size of the market as well.

These numbers could be crucial for possible developers to estimate possible sales figures instead of spending money on developing an application that might never be able to sell enough to recover the development costs.

It could also prove that the market is in somewhat better state than many people think!

Gulli

For example there are many RPC's now available reconditioned and at low price (so these would need to be included in the total otherwise an unrepresentative figure might be arrived at).

Secondly some people have been holding off committing on new hardware until Omega's release (how do you count these people ?). This vapourware release has negatively impacted the market and its hard to see how reliable estimates could be made (even if MD told you how many people had put down deposits it would not count those who hadn't but who had deferred purchasing until they knew the state of Omega and other hardware).

Gulli has a valid point, developers DO need sales information but I am not sure if this could reliably be obtained just at the moment. Remember as well people can purchase new software to use on existing hardware (so a better gauge may be to review sales of SOFTWARE).

Yes, users do buy software for their old hardware but recent sales figures would give you something to work with, at least a minimum size of the market. The report could then state the unknown factors like the Omega factor, Millipede, portable RiscStation, old computers still being used etc. It could include numbers from RISCOS Ltd. about RISC OS upgrade sales (just above 2100 in 1999 if I recall correctly).

Having numbers below the real size is better than having none. An application could be developed with that size in mind but a pleasant surprise would then lurk around the corner. Then again low numbers might actually put people off writing applications but those that would decide not to create a new application would probably not be happy with their sales based on wild guesses about the market size and leave it anyway.

I think whats really needed is a development environment that allows developers to turn round code quickly (and therefore cheaply) so that even with very modest numbers of sales they can make a decent profit.