Canadian Information Processing Society (CIPS)
 
 

CIPS CONNECTIONS

INTERVIEWS by STEPHEN IBARAKI, I.S.P.

International authority in Linux

This week, Stephen Ibaraki, I.S.P., has an exclusive interview with Hoyt Duff, a widely regarded international expert in the Linux operating system.

Hoyt has a varied background including as a senior vice president for a bank, college instructor, restaurateur, special education teacher, playing cocktail piano on weekends, and tending to his family business—the longest sport fishing pier on the East Coast.

Hoyt also spends his time as a columnist and regular contributor for Linux Format magazine. He has helped edit English versions of the Mandrake 7.x and 8.x documentation, and he has helped edit the early English content for the webzine, www.linuxnetmag.com.

Discussion:

Q: Hoyt, thank you for agreeing to this interview.

A: It's nice to be on this side of the table, Stephen. I'm usually the one asking the questions and here I'm not restricted to a single topic by an editor, so it's nice change of pace.

Q: You have programmed in FORTRAN on the IBM 360/65, Pascal on the IBM PC, and 6502 assembler on the Commodore VIC20. Please share both humorous and lessons-learned stories from your noteworthy history.

A: My programming experience is from a time when the dinosaurs still roamed the Earth. I learned FORTRAN as an undergraduate student at Cornell University as part of a hotel-related class. We used it to generate reports and I quickly found out that FORTRAN was a poor choice for printing business reports. I did find a few obscure commands that caused the prof to ask me to re-code the applications so he could understand them and give me a grade. I learned the value of commenting code and I learned to enjoy debugging it. When I transferred to Old Dominion University (in Norfolk, VA) to major in business, I picked up a few extra dollars for lunch money helping the computer science students debug their FORTRAN code.

A bank I worked for purchased an IBM PC (one of the first units with a cassette data port on the back) and installed some interesting apps on it. The expensive goal-seeking spreadsheet (this was back when Visi-Calc would not do such a thing) required that the instructions for the spreadsheets be written in Pascal and then compiled. Yecch. The entire point of the spreadsheet was to maximize the bonuses for the President and Chairman. Again, the most valuable lesson learned was documentation, but I also learned about being the computer geek in the compute-illiterate world; it was as horrid an experience as Dilbert ever had.

The VIC20 was fun. I performed hardware hacks on it like adding 4MB of memory on board to get a whopping 8MB total and integrating a slow-can TV decoder board. I began programming for it in the tokenized BASIC it used, but fell into writing assembler subroutines because they were faster. I hacked game cartridges to change the game play. I also dabbled a little with a FORTH compiler on the VIC 20 as well, but I never could find anything useful to do with it. I never got a disk drive for the little beast, but I did have a 300 baud VICmodem.

As times and computer languages progressed, I never kept up. I don't consider myself a programmer and I didn't need programming skills for my work. From my Linux experience, I can read source code and follow the program logic, but I have no desire to learn any kind of programming unless I'm forced to with a gun at my head.

Q: You discovered Linux from an article in Boot magazine and wrote your first article for MaximumLinux magazine. What intrigued you about Linux from those early times?

A: I had explored MS Windows as far as it would allow and then some, but I was intrigued with the unlimited potential for tweaking everything about the Linux OS. The fact that the source code was open and available for experimentation was fascinating, and all the development tools came with it. I explored several alternative operating systems at the time and was most intrigued by BeOS and was sorry to see it go.

Q: Can you share with us your take on the history of Linux?

A: The story itself is well-known. I believe that the philosophy of GNU and the way it has been used by Linux will have far-reaching effects on the future. It is re-defining the ways in which we go about doing things. One of my favorite authors is journalist James Burke who succeeded in making history become exciting and relevant for me. His TV series and book “The Day the Universe Changed” would most certainly include a chapter about this phenomenon were the stories being written today.

Linux and GNU are all about thinking about things differently. One of the hallmarks of intelligence is the ability to think about things in new and different ways. Linux certainly embraces that; perhaps that's why it has such great appeal.

Q: What lessons, experiences, and stories can you share about your work with Linux?

A: The most profound affect my work with Linux has had is my introduction into the worldwide Linux community. Americans are rightly criticized for being parochial and I have enjoyed being exposed to other cultures, specifically their way of thinking about things. As well, when there exists a community coalesced around a singular task that transcends cultural boundaries, it is easier to get to know people as individuals rather than by geographic or ethnic labels.

Writing about Linux has helped me learn more about it.  Although there are many areas about which I don't care to become enlightened (DNS, Apache, and programming to name a few) that leaves about ten million other related things I may explore. It seems that there are always new things to experience, new ideas to visit and new people to meet.

Q: What are your most important “best practices,” tips, and shortcuts regarding installing, configuring and working with Linux?

A: Four words: Document. Document. Document. Backup.

Did I mention document? You should always record what you do and what you know. The body of Linux knowledge is too vast to be easily remembered by anyone. As well, the current state of Linux documentation is abysmal (though improving). The man[ual] pages are woefully inadequate, for example, and filling them with some examples would make them better. It's good to see the Linux Documentation Project encourage and support better docs, but we must all do a better job.

Programmers in particular need to document their code in a more useful manner. I say that because programmers are notoriously poor documenters and good writers are usually not very good coders. With less cryptic comments in the source, I could better figure out what the programmer wants the world to know. Again, I would be happier if they just included a few examples in the manual pages.

The extant poor documentation is what got me started writing about Linux: I needed to understand what I was doing. There were disparate bodies of knowledge out there, some outdated, some plain wrong. That stuff still exists on the Internet and any web search will turn up a load of info-drek with every query.

Backing up your information is important as well. Smart people learn to do that after their first mistake no matter what their attitudes may have been. Experience can be a cruel teacher.

Q: What are the major components in the Linux environment, what are their roles and how do you see the roles expanding into the future?

A: The only true component of the Linux environment is the kernel. I like what I see in the new 2.6 kernel and I'm glad there is a healthy and dynamic development community out there. I am concerned to see Digital Rights Management become embedded into the hardware because that means the kernel must deal with it. In a proprietary world, I believe that means trouble for Linux.

The developments in the desktop have been impressive, but it's discouraging to see such an emphasis on eye-candy at the expense of compact, bug-free code. Desktops for Linux systems also seem to follow the lead of MS Windows and Mac OS, especially when it comes to flashy and dazzling behavior. I saw a preview of the next Windows desktop the other day and I wonder how long it will take for all those “features” to find their way into Linux? Perhaps we could lead the way in productivity instead?

Q: What do you see as the strengths and weaknesses in Linux? What are the major challenges to working with Linux?

A:  The strengths of high reconfigurability and the UNIX heritage are also the big weaknesses. Linux is still too difficult to configure and manage for the average MS Windows or Mac user even though they would have no trouble using a Linux desktop.

Linux is still also a work in progress. How can you get comfortable using a moving target, especially when the “next version” fixes so many annoying bugs and provides functionality that other operating systems have had for a while? If you take the version number of any Linux application and divide it by ten, you'll get a better idea of where it really is in development.

From its UNIX heritage, Linux gets a big heap of dense jargon that totally confuses the newcomer. Archaic UNIX customs permeate the operation of the OS as well. While these words and traditions help define our community, they also are an incredibly huge hurdle keeping people away making that perhaps the biggest challenge.

Q: Which “version” of Linux do you feel offers the best opportunities for businesses and how would you support your views?

A: Once installed, there is no significant difference between the different flavors of Linux except their homegrown management tool and application packaging methods. One seems as good as another in actually getting work done, which should be the main criteria for any business. Therefore, the “best” “version” is the one offered by a vendor that will provide the support and applications that the business needs.

Q: What are your views about the Open Source Movement, where is it today; how will it evolve in two, five, and ten years? What do you see as the major counterparts to the widely used desktop environment and applications? Where do you see them situated today and describe how they will evolve in the future?

A: By its very nature, the Open Source movement is open to constant re-invention. History has taught us that things that don't adapt will ultimately perish; I see a successful future for whatever the movement will become.

Q: Do you foresee a “killer app” coming out of the Linux environment? How about making some predictions on what technologies will survive in the long term?

A: I looked at some articles I wrote a few years ago about what the killer apps of the day might be or needed to be. I was wrong then (they came and nothing monumental occurred) and I'll probably be wrong now: I don't see a “killer app” at all; that's an over-hyped media term akin to the “Holy Grail”. The only app missing is a drop-in replacement for MS Exchange Server. That's just re-inventing the wheel, not advancing the art. The SCO lawsuit (and I predict their eventual loss) will do more to advance Linux than any “killer app” as it makes more people aware of what Linux really is.

Q: Describe your current book and share some tips from the book?

A: Red Hat Linux 10 Unleashed will cover the next release of Red Hat, now in beta testing. The distro itself looks pretty much like Red Hat 9 with bug fixes, but they are making changes to accommodate the 2.6 kernel which, I assume will make its formal appearance in Red Hat 11 in the spring. An updated 2.6 kernel for Red Hat 10 will likely be available prior to then. It also looks as if they will be including YUM, a package management application similar to Debian's popular app.

The implementation of Red Hat's “Red Hat Linux Project” will profoundly affect the book since many useful features (like multimedia) are being removed form the stock Red Hat distro to accommodate Red Hat's corporate plans. I and lead author Bill Ball plan on covering these areas in more detail so our readers can truly “unleash” Red Hat for their own use.

One of the most useful tips in the book is to ALWAYS create a backup of any configuration file before you modify it. This is a recurring “gotcha” to many people, new and experienced alike.

Q: What books are you planning for the future?

A: A friend and former police homicide detective with an Army Intelligence background wants to collaborate on a fact-based fictional anthology of detective stories. That will be fun. I'm also compiling a cookbook to celebrate the upcoming 50th anniversary of my family business, Lynnhaven Fishing Pier in Virginia Beach, Virginia. The cookbook will include anecdotes, remembrances, pictures and recipes form the last half-century of experiences there.

As for Linux books, I have no plans for any others since the Unleashed book takes so much time with revisions and updates. For myself, I prefer to write magazine articles. If I could come up with a good concept for a business-focused Linux book, I might be pursuing that, however.

Q: Can your share your views on the major competing technologies today, the nature of these technologies, similarities and differences, their strengths and weaknesses, market penetration, and where you see them in the two year and five year time frame?

A: While I am as geeky as the next guy and adore new gadgets and new technology, I am more pragmatic in my use of them. Remember that kitchen and household appliances were considered liberating technologies when new, but women in particular haven't seen too much liberation as a result. Likewise with new information technologies: what about the paperless office we were promised?  I need new things to do useful work for me and to liberate me by allowing me to do things in different but useful ways. Most new gadgets don’t do that, but keep those toys coming anyway.

Q: What do you see on the horizon that businesses and IT professionals “must” be aware of to be competitive?

A: The same thing they have always needed to be aware of: know your business, know your customer, and keep an open mind. Don't abandon what works for a new fad, but don't fail to learn about what is new and useful.

The political and legal aspects of IT are rising on the radar thanks to Microsoft anti-trust experiences and SCO’s lawsuits.  The best tool any professional has is information, so they should keep abreast of the industry and learn enough to sort through the marketing and legal hucksterism and discover the truth. The free and open discussion of IT issues makes that happen.

Q: What do you feel are the five hottest topics of interest to both businesses and IT professionals today and what will be the topics in two years and in five years?

A:  The cost, stability, support, continuity, and utility of the IT infrastructure and the information itself have been important and will continue to be important. The challenge, of course, is how each business defines those things for itself. I believe that Linux offers more to help deliver on the needs of business than other operating systems.  The culture of Linux provides a free and open discussion of IT and a better opportunity for business to learn just what these things mean to them.

Q:  What would be your recommended top references for IT professionals?

A: The best references they have are the IT people who work for them. Listen to them. Keeping up with IT in general by reading trade publications or perusing IT-focused web sites is actually a valuable use of time if those sources are succinct and current.

Q: What are the top ten challenges facing businesses and IT departments in the next five years and what are your recommendations to meet/overcome these challenges?

A: The entire field is one of challenges. The biggest is using IT to meet the organizational goals. Pointy-haired bosses are a bane to effective IT management, so subtle education is a must in order to maintain your “place at the table”. Keeping up with technology allows the IT manager to make intelligent choices and effective plans. As IT becomes more important to an organization, IT managers must have good management skills, good interpersonal skills, and good corporate political skills.

Let me make a pitch for writers as well. Besides the obvious benefit that you will learn more as you develop the knowledge to write about a subject, being a published writer in the technology field is a relatively easy task to accomplish (as opposed to successfully writing a New York Times' bestseller). Practiced and enhanced communication skills are always a plus. Writing provides the added benefit of helping the community and it's good for your resume.

Q: If you were doing this interview, what five questions would you ask of someone in your position and what would be your answers?

A: Only one question: What's the biggest problem to the wider adoption of Linux?

A comedian said that he was a vegetarian not because he loved animals, but because he hated vegetables. I see too many people using Linux not because they love it, but because they hate Microsoft. These same people also glom on to some other aspect of the OS (editors, desktops, widget sets) and treat it like a religious matter based not on what they like about their choice, but what they hate about the alternatives. They also denigrate those who make different choices. This negativeness does nothing to foment the wider adoption of Linux, but has quite the opposite effect. It poisons the well for us all and makes decision-makers hesitant to consider Linux. These bigots need to grow up, but I suspect they never will. Solve that problem and Linux will spread.

Q: Hoyt, thank you for sharing your knowledge and experiences with our audience.

A: Thanks for the opportunity for some introspection and rambling. Now, back to work.