Alan Hargreaves' Blog

The ramblings of a former Australian SaND TSC* Principal Field Technologist

One of my journalistic pet peeves

Journalists who don’t do their homework, relying instead on worn out clichés.

Todays peeve comes from an interview with Jonathan by Gavin Clarke.

The paragraph in question is

As such, Schwartz indicated spending precious R&D on commoditized or expensive hardware is on the way out for Sun – in at least some areas (NAS and storage apparently excepted.)
The company has, of course, come in for heaps of criticism for channeling dollars into its proprietary Unix hardware, and software, architecture with UltraSPARC and Solaris while the world has moved to “commodity” Intel and AMD hardware running Linux and Windows.

Proprietary is clearly not the right word to use here.

He makes reference to UltraSPARC and Solaris as being “Proprietary”

Come on Gavin, you can do better than that.

SPARC is one of the very few cpu architectures out there that is actually not proprietary. Go ask AMD how open Intel’s architecture is.

For more than a year now, Solaris has been licensed under the OSI approved CDDL, making it also open.

In no way, shape or form can you truthfully refer to either as proprietary.

Gavin, as a journalist your job is with words. One would expect you to be able to choose them a little more appropriately.

Technorati Tags:


Written by Alan

June 23, 2006 at 5:18 pm

Posted in OpenSolaris

4 Responses

Subscribe to comments with RSS.

  1. Understood.
    But it is also desirable to suggest a better alternative, after having levelled a criticism. From your perspective, how can this be more correctly worded?
    A reader scanning over this would likely get the off-the-bat impression that the writer merely meant:
    <If I want to run SPARC Sol, I can’t run on a SPARC box of other than Sun’s manufacture>, and he didn’t think that all the way through first.
    By adding positive input to a negative comment, the Criticism Usefulness Flag toggles from 0 to 1.

    Jim Cotillier

    June 24, 2006 at 10:13 am

  2. You do make a valid point Jim. The sentance actually stands on it’s own if you simply remove the word. Interestingly reading on, I note that he does not try to say that the Intel and AMD chips are not proprietry, in which case one must wonder why he made the comparison.
    Perhaps what he was trying to do (for the hardware architecture) was to contrast the commodity nature of the Intel/AMD architecture as against SPARC. That would have been a valid comparison.
    I can only speculate that as far as the Linux vs Solaris comparison, the point was being that Linux is currently popular and used by lots of folks (as is Solaris actually, but you don’t see the press), and Solaris is not Linux. It’s a very subjective comparison and I personally don’t believe that it’s inclusion adds any meat to the article.

    Alan Hargreaves

    June 24, 2006 at 4:22 pm

  3. Jim, one thing that I will correct is your statement about if you want to run Solaris on SPARC, you must run Sun. That’s actualy not correct. There are a number of other vendors who build machines around SPARC, most notably Fujitsu.

    Alan Hargreaves

    June 24, 2006 at 4:26 pm

  4. Alan,

    Yes, indeed! What I meant by “[he] didn’t think that all the way through first” is just that — that the writer didn’t think of SPARC outside of Sun, and probably should have, because it’s certainly there. Then, as long as the well-known SPARC standard is allowed to hold sway, such incompatibility worries remain groundless. Sorry for the mislead there.

    As far as comparing Solaris to Linux, it’s hard to comment on that in a word, other than to say I strongly encourage such commenters to go look hard at the respective histories of the systems, because to know where you are, you need to know where you were.

    Neophytes to a discipline react to what they first see and know, and that tends to be the outer layers of the facade. Learning the history gives depth.

    From the BSD roots through to the massive engineering infrastructure that has been built over the years, Solaris is of course the major OS in the greater UNIX arena, capable of dealing with the power and extensibility requirements of the largest of enterprises.

    Linux has indeed evolved explosively as well, since I first began experimenting with Linus’ 0.99 kernel back in ’91 or so. But I invite commenters on the subject to read, before posting, the corpus of what Andrew Tanenbaum and Linus have both said about the evolution of Linux from the world of MINIX.

    The “ur-DNA” of Linux comes in part from MINIX, a system which, even today, Tanenbaum is keeping small enough to be functional, but from a CS perspective still provably correct algorithmically. Thus, though Linux has in its own right evolved well toward the comprehensive, we see the two systems are just different animals, and the core DNA is never truly repudiated by the respective organisms.

    So, the truth is that generally, whenever I run in to a “Why doesn’t Solaris do -X- the way Linux does?” kind of posting, I just go into sparse file mode and skip ahead over the NULLs.

    Jim Cotiller

    June 25, 2006 at 12:47 am

Comments are closed.

%d bloggers like this: