1) Is Linux heading towards a major rather than minor computing role, or is it still too early to tell how things will unfold?
I think it is still hard to predict. The market is changing so fast.
Not only do we have the shift from the desktop to servers and thin client machines happening - and it is happening at least in consumer spaces with things like NTL's TV set internet stuff and the ongoing projects from other big vendors. We also have the exploding mobile market, the phone/PC fusion and more. So it's hard to guess.
Linux does well as a server OS, it does extremely well on the thin client and embedded environments too. The configurability and the fact that it isnt owned by a competitor to the companies using it make it a big win for them.
The desktop will be the hardest battle. The Microsoft lawsuit might help there if there is enough will in the US Government to split Microsoft and force the units to act alone. We now have most of the GUI environment we need for end users.

2) What do you think has been most integral to Linux's growing success and visibility in the past 18 months?
a) In terms of the OS itself (eg, why is Linux such a contender?) and
b) In terms of marketing success (eg commercial deals, product announcements, backing of big name players, etc).
The biggest thing of all has been the rest of the world seeing free software going from a crazy concept to a marketed advantage and marketed in a way that the business community understands. The stock market flotations also gave it a definite air of reality.
You can get news out of hype, and there has been a lot of Linux hype from companies either because they sell Linux products or because they see it as a stick to bash Bill Gates with (or both). Hype doesn't actually create a large user base and sustained deployment which Linux unlike Java is getting.
In terms of Linux itself I think the Gnome and KDE desktops have been the big shift in thinking. Linux has picked up people who believe in OpenSource and who want Linux to be part of a bigger community than just the computer wizards.

3) Where would you like to see Linux go today?
a) what could be done to most enhance the OS itself and why would that element be important (user interface, killer app, or whatever) and
b) in terms of commercial success.
Is total world domination (in the commercial space) important?
The big step is the desktop. There are vendors with clear interests in this sort of area - Helixcode and Eazel for example who are doing real money work on the GNOME desktop. The other big area to deal with is high availability clustering. Wensong Zhang did the first free Linux clustering for web server failover and the like but that is only the baby steps. To do full clustering and to be able to position Linux to replace things like VMS as a highly available (and here we are talking minutes of downtime a year) clustering solution is no small job.
I personally think the desktop or at least the thin client end of the desktop is the most important. Linux has good security features which makes it ideal for things like call centre environments, although perhaps less so for their staff. Building a thin client Linux environment with hotdesking, sensible shared file store and good network efficiency is a golden opportunity.
Total domination is bad. The Microsoft dominance already badly misled people about how to choose systems. Instead of 'what tool do I use for the job' it's 'well it was shipped with the box'. Linux is a tool, Windows is a tool and so are numerous other systems. It's really important people go back to looking for the right tool for the job. That will never always be Linux. No single tool can do everything well.

4) If it's important, or if you think that Linux is going to take a larger role in the commercial world anyway, where do you think it will have its greatest success, and how might that change the computing industry?
For example: at first, Linux seemed a desktop OS for enthusiasts, then began to sneak onto servers and has staked out some serious web server territory, of course. But IBM is eyeing it, for example, as being as important to applications as TCP/IP is to the internet. Software companies will port to Linux, they say, because it's easier than porting to Unix flavours or NT. That benefits hardware companies and big app companies like SAP, according to that view. I'm interested how you think things might develop; if you'd agree with the IBM scenario, for example.
Linux is the sum of contributions so it will go where the contributors take it. Right now that is everything from IBM mainframes to pocket computers. Some of it is through investment and funding and a lot of it is because someone just thought it would be neat if Linux did XYZ and had fun working on it.
I can see Linux becoming the generic OS for embedded applications and servers. Where people will pick Linux unless they need some given feature that favours another system. I don't know what will happen on the desktop.
Only time will tell.

5) There's much talk now of Linux fragmenting into flavours. Do you think this could happen?
The ISVs don't and won't stand for it. The customers won't stand for it and the community definitely wouldn't. There is a lot of work right now to get definitive standards via the Linux standards base (linuxbase.org). That has real big vendor backing because the vendors want to ship 'Foo for Linux' not 'Foo for Red Hat', 'Foo for Mandrake' etc. Also non-Linux people are quite interested in this too. We may end up creating a common interface for applications on the PC in any Unixlike environment. That is definitely good for customers.
Ultimately nobody who wants to build the Linux space either commercially or for fun has any business or technical reason for creating an incompatible variant. In fact the pressure is entirely to improve compatibility.

6) Can Linux and the open source attitudes of the developer community surrounding it co-exist with Linux's commercial success?
Do you see strains arising out of these developments?
The developer community on the whole seems not to care too much.
Certainly the developers I work with both in and outside Red Hat are more loyal to Linux than to their companies. There are certainly marketing people who see other vendors as the enemy. Programmers never cared for marketing peoples' views anyway ;)

7) Linux companies hit celestial IPO heights recently then have crashed since then. Is this a 'correction'?
A failure of Linux to meet expectations in the commercial world?
A pointer that an open source OS won't make anybody money?
I think there is plenty of scope to make money, as it seems to do a lot of people. Its certainly not going to make Microsoft like profits and that's probably a good thing for everyone long term. It is about time people got fair prices.
The stock valuations just remind me that the US technology stock market is basically a gambling den. It seems to be based on riding hype to the highest value you dare then trying not to be the last one to sell. In time I imagine the Linux valuations will settle down.

8) Finally, what motivated your own involvement with Linux and why does it remain so central to your life?
Well I got into it by accident. I wanted an OS to debug my multi-user game on and Linux hacking became a sort of hobby.
Now I'm paid to give away code and watch it empower people all over the world - especially developing countries. I don't know many jobs to compete with that one.
Alan