News

Features

Guest Essay

Technicalia

Security

Community

Commentary

Home

To the Editor:

Distributors should specialize

I basically agree with Joel Hammer's response. It really and truly is not a question of Linux users' refusal to purchase actual products with real cash. Many of them do this frequently. The biggest problem, at this point is that Linux has not sufficiently matured to provide a predictible experience for developers, users, or companies marketing products.

As a Mandrake Linux fan, I have been faced with the concience wrenching question of whether to become a subscriber. Yet I have learned that Mandrake is supporting developers of the K-Office program. K-Office does have some potential and might be a great suite of productivity applications whenever it gets further along. Also, there is a strong belief that Linux will "take off" after the appearance of "killer apps" convinces consumers that they must have this product. However, I bought Mandrake-Linux (several times) because it appeared to be the most advanced Linux distribution, not as a K-Office platform. So, in this situation Mandrake appears as an operating system packager, but is actually working on productivity applications. Now they are supposedly having a financial crisis, and it is becomming apparent to me that their operating system product has not received the attention it deserves. IMHO, Mandrake-Linux will have attained a higher level of maturity when the staff of that company can properly focus on preparing a well integrated set of kernel, GNU, and installation/maintenance programs. In fact, I do not agree with Mandrake subsidising K-Office development, or any other productivity software development. It is too far removed from the kernel. If packagers must subsidize the desktop, let them work on window managers and widgets. Programmers may find it more "cool" to work on K-Office, etc. but, as a paying consumer, I want the distribution packager centered on the kernel. Mandrake should recognize that they must specialize and become the best at what they do, which is not making kitchen sinks.

Linux, or is it GNU/Linux (but never forget XFree86, TeX, Apache, Bind, Sendmail, Mozilla, Samba, StarOffice, PostgreSQL, etc.), is a very good, ok - the best, operating system available for the PC. As the market and user base grows, the opportunities for distribution packagers will become more abundant and eventually, when the necessary components have been written and passed beta stage, a puzzle will appear that can be assembled for the desktop. Perhaps some Linux distribution will achieve this holy grail soon, and it will be the one that puts the package together - probably not the one that writes the package.

I found the long and interesting history of OS/2 on this site, and the guest commentary by Shawn Gordon of Kompany (whose products I have purchased, but can't recommend for reasons listed by Joel Hammer). Both articles are very informative and I appreciate this website posting them. Both articles point out (indirectly) the pervasive lack of focus in the computer industry, which prevents anyone from formulating a coherent plan and then having the discipline, perception, and resources to stay with the plan and adapt it to changing circumstances. I used to believe that a really great office suite (or pick your favorite app) would be a great boon to Linux. Since Linux is growing anyway, it will be the distribution packagers who will take the Linux thing to the next level and the "desktop" will follow. They will do this by offering a superior platform that is well documented and not too difficult to manage. All of the "killer apps" are out there now, they just don't yet have the place to land.

Thanks for an interesting website.

Kevin (Last name withheld by request)