We continue with Part 3 of Andrew Rawnsley's RISC OS interview.

What do you use RISC OS for in 2023 and what do you like most about it?

Pretty much all my daily work, except web browsing, and even that is beginning to be possible under Iris :)

Granted, my daily work is mostly email, programming, DTP, graphics work, tech support and so on, so I'm probably a special case, but RISC OS continues to do what I want.

I think a lot of computing comes down to what your needs are.  RISC OS comes from an era where programs did a task and did them well.  So long as your needs are met by the programs, I see no reason to change.

However, computing as a whole has changed since then.  It used to be about using your tools to create things.  Now mainstream computing is focussed around endless consumption, especially of web and media.  This means the goalposts are set by others who create content, and so computers and software have to change rapidly to meet those moving targets.

So, what do I like about RISC OS so much?  I think it is its consistency - that things work the way they always have - and that the programs meet my (somewhat odd) needs.  I like the way the programs work together.  How small programs can combine to be more than the sum of their parts.

I like how I can drag and drop things from one program to another.  I like how Acorn's forethought meant that vector graphics (Draw) are part of the OS, making that a fundamental format for passing data between apps (unlike other OSs where proprietary formats are often used as crude replacements).

I like that it is small and nimble - quick to load and swift to use with no unnecessary animations etc.

But most of all, I like the muscle memory that goes back 35 years.  It's very hard to beat that level of familiarity.  I sometimes feel that RISC OS is part of me, of my DNA.  I think maybe if you break my bones, you'll see RISC OS cog wheels where the bone marrow ought to be!

What is your favourite feature/killer program in RISC OS?

If I had to pick a single one, probably Impression.  I learned to word process, to lay out documents, to use graphics, fonts, clipart (that's a word from the past!) and so much more.  Starting with the first version on the A3000, it was a complete game changer in how I understood software to be.

We had other programs at the time, but none of them epitomised the future of computing (in the late 80s and early 90s) in the way Impression did.

Of course, Ovation Pro is a better program in most measurable ways, and I use it most days for more complex documents where Impression isn't the best fit (most notably booklets and other duplex print jobs).  But Ovation Pro came later, and the muscle-memory (that term again) had formed around Impression.  I think that plays such a big part in which we prefer - whatever we learned on - Impression vs Ovation, Schema vs Fireworkz, Artworks or Draw, Messenger or Pluto.

Of course, my desert-island "wouldn't be without" programs also include Messenger Pro / NetFetch (for email), Ovation Pro, SparkFS, Paint/Draw, ChangeFSI/DPlngScan and Fireworkz/DataPower.  Oh and Prophet!

I've also come to specialise in supporting Sibelius users, because I was/am a musician (bassoon).  However, I don't really use it a great deal myself, mainly because I rarely need to create musical scores.  I also felt it wasn't quite "RISC OS" enough for me - I actually preferred the desktop-feel of Rhapsody - although there was no questioning its capabilities.

For me, other essential programs are UniPrint (allows me to integrate RISC OS and Windows almost seemlessly), LanMan98 and the RISC OS C tools.

What would you most like to see in RISC OS in the future?

The continuing embrace and support for new technologies.  Wifi, NVME, Bluetooth to name a few.  64bit almost goes without saying, but is no small task!  Decent RISC OS web browsing, and enhanced support for video and other media.  It's a long list!  And more software.... lots more software :)

I appreciate many of these things fall on my shoulders one way or another, so I'd like to help grow our developer base to expand what is possible on RISC OS.  I want to still be using RISC OS as much as I do now, in another 10 or 20 years.  I said something similar in 2000, funnily enough, and here we are discussing RISC OS in the present tense, almost 25 years down the line!

So, I'd also like to ask the RISC OS-genie for more hours in the day to try and achieve as much as possible.  And for more programmers - goodness only knows we could use more C and assembler coders.  Preferably ones with well-paying day jobs!

Favourite (vaguely RISC OS-related) moan?

I think this has to be the "two of everything" duplication of effort that seems to have permeated RISC OS all along, and ultimately killed off so many projects.  Not to mention the effect that has on developer moral.

In an infinite market (or at least, a very large one), duplication of effort makes sense because it can drive creativity through competition. However, this only works because there are (we hope) enough customers to support the various competing products.  I know, I know, that rarely happens in reality, but it is possible.

However, in RISC OS, it is such a small market that duplication ends up being a real problem.  We don't have enough programmer hours to achieve all the goals we have, let alone for two or more coders to be battling it out effectively reproducing each others ideas.

This came to something of a head with the photo editing applications for the RiscPC - ProArt24, PhotoDesk, Studio24 (Pro), DA's Picture, Revelation, and several others.  All really nice programs that ended up on the scrap heap, except for PhotoDesk, and even then, the original developers left the scene.  There simply wasn't/isn't the market for this.

You can see it time and again in RISC OS, dating all the way back, and in almost every case, the end result hasn't been good for the platform.

I've felt this first hand with many of our applications, and it has definitely been hard to justify ongoing development.  Sadly software development for RISC OS hasn't been profitable for a long time, but it is sooooo necessary.  There's not much point buying new hardware if there's no software to run on it, and vice versa.  So, I've always seen software and hardware to be a joint thing - any profit from one funds the other.

Even tiny-selling products like PiFi saw more-or-less copies appearing, which seemed quite daft in such a tiny, specialised space.

Now, before anyone says anything, I know I'm just as guilty of this "duplication of effort" as the next man.  There's always a tendency to think that one can build something easier or more tailored, or with specific new features that justifies the duplication.  It's really hard to avoid!

However, it unquestionably harms (and has harmed) the platform.  So, yes, that's my pet peeve / moan.

But, in the interest of being constructive, I think this is where open source can help.  Historically it has been necessary to duplicate things with closed-source projects because that's the only way you can add your ideas.  Open source means that developers can add things and build on existing work with code shared for everyone.

That's why things like Pinboard 2 and the ROD TCP/IP stack come with source code, and why I tend to provide source to some of the R-Comp programs to those who ask for it (eg. Messenger).

On a grander scale, it is why RISC OS is Open Source, and why ROD championed an industry standard, open source licence.

The challenge, of course, is finding a way to balance the books in an open-source world... answers on the back of a postcard :)

To Be continued...

You can read lots of other interviews on Iconbar here and if you would like to be interviewed, just drop us an email.