Columns

Letters

Linux Experience Not the Norm

I read with amusement John Williams' troubles in installing a version of RedHat Linux ["Is Linux ready for the enterprise?," October 1999, p. 92]. While I am very familiar with the troubles of installing operating systems on new machines, I did have a few problems with his article.

The gist of the article is that Linux is not ready for the "enterprise" because it's hard to install. Like I've said, I've installed a lot of operating systems (different ones — not just Windows) and the easiest I've found was the BeOS. But then again, it did not recognize all of my hardware either. My point, and Mr. Williams mentioned this as well, is that no operating system installs "easily," without a hitch, every single time and on every machine. (As an aside, I've had the most problems with Windows NT.)

However, the fact that he installed it on a machine with bad memory in the first place and continued on with this article harms his credibility.

Saying the RedHat phone support person had a "nasty tone" was uncalled-for. (This is purely subjective anyway. What was his tone like?) In addition, he didn't need to recompile his kernel to add modules; that's why they are modules ... so you don't have to recompile the kernel!

Mr. Williams says Linux found "address conflicts" between his network and sound cards. So, he re-formatted the hard drive (because this was a Linux-only machine), and installed "Windows and Windows tools" to "reset the Ethernet card." Are you serious here?

By the way, what does this have to do with the "enterprise" anyway?

He deduces that because this is what he went through to get Linux installed, it must be the norm. Therefore, this "means more IT labor and thus more money to deploy Linux." I'm sorry, but do you really believe that Mr. Williams' experience was the norm? I know I didn't have an experience remotely like that.

I thought the article did point out some deficiencies in RedHat's support that they might want to work on. But I got the impression that Mr. Williams wasn't really interested in seeing what was involved in getting things set up and working as much as he was in reporting the number of days it took RedHat to respond to his problems (which were odd in the first place). I can tell you from experience that hopping on IRC or Usenet would have returned much more satisfactory results (granted, if he had another machine).

I also found his comments about Linux on the desktop odd as well. He reports that "most corporations use a wide variety of tools on the desktop" and that Linux lacked them. It would have been very helpful and informative to let your audience know what applications Mr. Williams was talking about, but you didn't. It could be that the applications are there and he didn't know it — but it becomes an unsupported claim/ opinion, and thus worthless.

For the typical employee who uses Word/Excel every day with browsing and E-mail capabilities, Linux can be a perfect and cheap match. How many times have Mr. Williams' users deleted SYSTEM.DAT in Windows 95 because they were "cleaning up their hard drive"? Because of userspace, and Linux's Unix heritage, this, of course, is impossible [with Linux].

All in all, I found the article interesting, but misrepresented to the audience and not very objective. Naturally, I'm looking a little more closely at Application Development Trends to see if it's the kind of magazine that provides independent, educated material. If not, well, we won't get it for very much longer.

Michael J. McCormac