Posted by: Anonymous Coward
on February 15, 2005 02:23 AM
I really have to question the wisdom of a review that approaches the subject with the attitude that more features equals better product. From a security perspective, such an approach is fundamentally mistaken, regardless of platform. A balanced review would take a close look at how well the system has tracked advances in interoperability, stability, standards compliance, and security practices.
Let's talk about Linux distributions in particular. The 2.6 kernel is clearly the place to be in terms of functionality, robustness, security, and future development. However, a lot has changed behind the scenes, the device driver interface in particular. So it's a serious amount of work for a distribution to bring itself up to the new kernel. If RHEL came out with a new release with no other changes whatsoever, it would still be a significant step forward.
The reviewer evidently doesn't understand this basic consequence of kernel design, so what follows is really no surprise. Never mind that the "standard" test environment used in the review doesn't seem to be particulary standard at all -- no justification is offered for its particular choice of hardware being representative, let alone standard, within the industry. The point is that hardware vendors have to release drivers compatible with Linux 2.6, or they have to be developed independently, and then these drivers have to find their way into various distributions.
A review which conducted an objective study of hardware compatibility among Linux distributions would be welcome. This review doesn't even touch the subject.
Likewise, a review which compared security benchmarks across Linux distributions would be of great benefit to everyone concerned with the principle of security by default.
In fact, any review which objectively measures some characteristic of interest across distributions creates a useful point of reference for both the industry and end users. Let's see more of those, and less of this opinionated fluff!
Learn about Linux
Download Linux
Get Linux help
Featurism
Posted by: Anonymous Coward on February 15, 2005 02:23 AMLet's talk about Linux distributions in particular. The 2.6 kernel is clearly the place to be in terms of functionality, robustness, security, and future development. However, a lot has changed behind the scenes, the device driver interface in particular. So it's a serious amount of work for a distribution to bring itself up to the new kernel. If RHEL came out with a new release with no other changes whatsoever, it would still be a significant step forward.
The reviewer evidently doesn't understand this basic consequence of kernel design, so what follows is really no surprise. Never mind that the "standard" test environment used in the review doesn't seem to be particulary standard at all -- no justification is offered for its particular choice of hardware being representative, let alone standard, within the industry. The point is that hardware vendors have to release drivers compatible with Linux 2.6, or they have to be developed independently, and then these drivers have to find their way into various distributions.
A review which conducted an objective study of hardware compatibility among Linux distributions would be welcome. This review doesn't even touch the subject.
Likewise, a review which compared security benchmarks across Linux distributions would be of great benefit to everyone concerned with the principle of security by default.
In fact, any review which objectively measures some characteristic of interest across distributions creates a useful point of reference for both the industry and end users. Let's see more of those, and less of this opinionated fluff!
#