Interview with Andrew Hume

by Rob Kolstad, Editor of ;login:

Rob: Welcome aboard as incoming USENIX president. Do you have specific plans and ideas you'll be pushing to implement as soon as you take office?

Andrew: Thanks. Right now, USENIX is in good shape: it has a respectable (and increasing) number of members, very healthy finances, a good slate of conferences, and an excellent staff. Yet, as the computer world changes, as the number and types of users change, as the technical challenges in serving those users change, so does the need for USENIX to refine and adjust its technical focus; and in doing so, USENIX might want to alter its membership makeup. To be specific:

USENIX needs to reassess how big it wants to be and what its primary goal ought to be (as an organization).

USENIX needs to decide if it wants to increase the diversity of its membership. As an example, we are actively pursuing users of PC-based UNIX systems.

USENIX needs to implement fully, and perhaps expand, its new program of proactive student outreach.

Rob: USENIX's financial health is the best it's been in decades. Do you have ideas on how best to use moneys to advance USENIX's goals?

Andrew: Both the previous and new boards have had good ideas on how to invest USENIX's profits (substantially due to the competence of the USENIX staff and to Rick Adams's generosity and stewardship as treasurer). I have been and will continue to be an advocate for funding student- and standards-oriented initiatives. This includes:

student scholarships

increased fund for student stipends for attending conferences

creation of a fund for student grants for Computer Science projects

extend our standards coverage to include the IETF (which moderates most Internet related standards )

Rob: Are there any moves afoot to expand the general conference offerings back to two per year?

Andrew: I have none. While I regret the upcoming 18-month gap between main conferences, I am very pleased at the greatly improved quality of the technical papers at the general conference. I think this is largely due to the fact we have only one general conference per year.

Rob: Do you see USENIX moving its focus to that of the Internet and Web?

Andrew: It already is! Of course, it's more complicated than that. As Mike O'Dell once said, USENIX's job is moving information from where it is to where it isn't. We currently do this in a number of ways, including conferences, workshops, ;login:, Computing Systems, SAGE pamphlets, and reprintings of documentation. The key issues are focus, content, and delivery.

USENIX's focus is on three overlapping groups: CS researchers, OS and application developers, and system administrators. I have been a staunch advocate of extending our focus to include the growing PC-based UNIX community, which includes the various BSD-based systems and Linux. One direct consequence of this is the Linux-oriented activities at the upcoming Anaheim conference.

Contentwise, USENIX has been doing a pretty good job of giving its members the information they need and anticipating the information they will need. Given the ideas generated at the last board meeting, I expect USENIX will continue to do well.

USENIX has the following main vehicles for delivering information: the general conference and LISA, specialized workshops,;login: , SAGE booklets such as "Job descriptions," and the journal. Our conferences are well received, the workshops are popular, and the SAGE publications are valuable resources for the SAGE members. USENIX is working on various proposals to improve :login; (which I think is doing a fine job). There is significant concern over the journal and the small number of submissions; the board is monitoring the situation closely. Returning to the original question, USENIX has had many talks, papers, and tutorials on the Internet and Web (Java is a recent example). And we'll continue to do so as long as the membership needs and wants them. But there are a lot of other trees in the forest.

Rob: Some have observed that Microsoft's incredible market share of PC operating systems might have a deleterious effect on UNIX. Do you see the end of the road for UNIX any time soon?

Andrew: PCs are great. The speed and price/performance of modern Pentium-based PCs really drive home the advantages of using commodity hardware. And, best of all, you can get real software to run on them. While I have been using PCs for a few years, over 90% of my use has been non-Microsoft (mainly Plan 9-I play Doom on my wife's Macintosh).

My personal assessment of PC software is that nearly all of it is at best mediocre. And where are all the people who have complained about how hard UNIX is to use? Surely they would have something to say about the vortex of despair waiting for anyone wanting to reconfigure their PC. I have experienced my fair share of bugs and disasters during my prime UNIX years of 1975-1985, but none of it was as futile and frustrating as a recent attempt to change a Windows 95 PC running Exceed to use a three-button mouse. Plug and play and pray and play . . . . (The solution? Keep a supply of small mammals to offer as sacrifices, and maintain a very close relationship with a PC guru.) After living the Windows 95 experience to the full, I anticipate following my colleagues who run real software (Plan 9 or Linux or some other UNIX-like system).

UNIX, and its derivatives, has long been the dominant environment for OS and application research because of its ubiquity, familiarity, power, and accessibility. In general, I think this will continue to be true for the foreseeable future. Microsoft's dominance of the PC software world has little bearing on this; mostly, the problems stem from third party suppliers who won't tell you how to program their devices but supply a Windows driver instead. On the other hand, I am not wedded to UNIX. I can, and will use, almost any environment that helps me do my job well, and should that mean using some Microsoft operating system, then so be it.