|
|
letters to the editor
APOLOGIES
ON UNIX CERTIFICATION
[Editor's note: This note went by on a mailing list recently. It was some of the most interesting commentary on certification I've read in a long time. — RK] My UNIX certification rant: When I worked for a certain company that made Incredibly Big Machines, we were encouraged to get certified, and people often hung the test scores, not the certificates, on the wall, out of pride. It was there that I learned to value certification as a form of continuing education. Me, I learned what I know about UNIX by spending most of what I earned at my student job on every O'Reilly I could find. I even owned the Curses O'Reilly, for god's sake. And then I'd love to read those books. It was really disturbing to get into the big world and find out that only one of my coworkers read tech references on his own time. But when my teammates went for a cert, they'd hit the books, and the result was that they'd gain in knowledge they'd never have sought out for its own sake. Point being that if someone gets a certification, it's at least arguable that they're trying, and if they seek to obtain the knowledge of a more experienced admin, so much the better. Of course, it's probably wise to be wary of people who feel the need to make gratuitous references to their certifications (or college degrees, for that matter). I don't blame anyone for getting sick of paper tigers, though; I worked with a fully pedigreed MCSE who was unable to set the system clock on his servers. Regarding NT certifications: It's unfortunate, but not everyone in the extended family of systems administration is a big geek. NT does not attract big geeks. Big geeks like cool tech and nifty puzzles to solve. NT is not cool tech, and among the many, many things NT is, it is not, never has been, and will likely never be nifty. My experience is that NT admins tend to be bright folks who decided to bootstrap themselves into a computer career. For people who are looking for a career but don't yet have a Clue, certification is a well-paved eight-lane highway with a brightly lit on-ramp. This highway doesn't lead directly to a Clue, but there's an unmarked access road if you know where to look. Most of the time, the wayward novice just needs a friendly shove off the road. And of course, lots of people never get a Clue and believe that the highway is all there is. Luckily, these folks are driving around in bright orange Dodge Challengers (license: GOTMCSE) with glasspack exhausts and are easy to spot and avoid.
IN DEFENSE OF FREEBSD
In his letter in the February 2000 ;login:, David Maxwell stated that "I looked today at <http://www.freebsd.org> and <http://www.linux.org>, and I couldn't find any mention of a non-Intel installation." I don't see why he expected to find a mention at <http://www.freebsd.org>, since, as Mr. Maxwell stated earlier in his letter, "The FreeBSD folk wanted to focus on the i386 platform and put their full energy into it." The FreeBSD Web site itself says much the same thing: "FreeBSD is an advanced BSD UNIX operating system for 'PC-compatible' computers." As for Linux, Mr. Maxwell apparently didn't look too hard, since one of the prominent links at <http://www.linux.org> is "Can I run Linux on my type of computer?" The answer is at <http://www.linux.org/help/beginner/platforms.html>: Linux runs successfully on most computers, laptops, and platforms. There are several projects underway to port Linux to other hardware configurations. An overview of hardware compatibility resources are listed below:
Supported PC-based CPUs include:
Supported non—PC-based platforms include: That link has been there for several months, as I used it when searching for information about possible operating systems for a Sparc ELC. Beyond that, linux.org is not the "official" site for Linux, the way freebsd.org is for FreeBSD. The closest thing to an official site for Linux would be <http://www.kernel.org>, "the primary site for the Linux kernel source." Finding mention of non-Intel Linux is even more trivial there, as it is on the front page. Later, Mr. Maxwell states that "Linux and FreeBSD may claim that they support multiple platforms, but the code is not integrated into their source at this time." As I've already mentioned, FreeBSD makes no such claim. As for Linux, the statement is false — the base kernel source has included the Alpha, Motorola 68K, MIPS, PowerPC, and Sparc code since the 2.0 days; 2.2 added ARM and Sparc64; and the 2.3 kernels add the IA-64 and SuperH processors. Perhaps the S/390 code qualifies as "not integrated into their source" — S/390 support appears in the 2.2.14 kernel but not yet in the 2.3 series. I certainly agree with Mr. Maxwell's closing statement that NetBSD should have received more attention, but it isn't necessary to spread falsehoods and innuendo about other operating systems to make a "sales pitch" for his preferred choice. More to the point, I expect the ;login: editors to do at least some rudimentary fact checking before accepting letters for publication.
Rik Farrow, who wrote the article that inspired this exchange,
responds:
WINDOWS NT
<trost@cloud.rain.com>
Until now, I have not been particularly interested in the discussion
about whether USENIX should be paying attention to NT, but upon seeing what such attention is doing to the Association, I am now convinced we
would be much better off limiting our focus on NT to its interaction
with UNIX systems. There are plenty of organizations related to
administering NT itself; there is little point in USENIX becoming yet
another one.
|
 Need help? Use our Contacts page.
First posted: 20 Jul. 2000 moun Last changed: 20 Jul. 2000 moun |
|
Issue index