Slashdot Mirror
The Faded Sun
jlowery writes "Robert X. Cringely seems to
think so. Forget the hardware side: what does this mean to the future of Java? Will there be enough incentive to continue to develop the language for whoever acquires Sun? Or will Java developers have to swallow hard and submit to the whims of the dark overlord? Maybe I'll switch to Mac development, after all."
You are right about your comments about Solaris, but are mistaken about Sun's stance regarding your other comments. Regarding the executives thinking that the future is in selling hardware alone, you're incorrect - Sun is trying to sell itself as a solutions vendor - not just hardware, but software and services as well. See http://www.sun.com/solutions/ for more info. Regarding your comment saying that Sun is ignoring high-volume low-end side of the market, again this is not correct - sun has introduced the LX50, and 1U rack mount system that runs linux or solaris (x86). See more about that at http://wwws.sun.com/servers/entry/lx50/
Finally, a word about Robert Cringley - how many times does this guy need to be wrong before the industry starts ignoring him?
To get a 100 disk fiber Disk array, connected to a 8 way box. One that doesn't throw a drive or a controller or a cpu every other week. Sun Solaris sucks ass, Debian GNU/Linux is the One True OS, but the hardware works. We have tons of linux boxes, but they all suffer from running on x86 hardware. Whether we buy it part by part and build it ourselves, buy it from Dell or HP/compaq they just fall apart.
We've got several Dells with RAID drives, 4 CPUs, never a linux crash (ala BSOD), but we get on average 1 major hardware crash every year on EACH machine due to hardware going south.
We also have similarly equipt sun servers (that suffer from an even higher load). We're talking about 17 SUN 4500's, the ONLY failure over the last four years was due to a fiber controller failing, it was a dual controller, but a firmware mismatch caused the 2nd controller to not come on line properly. 1 outage and it was our fault anyway, if we had upgraded the firmware like we're supposed to it would have never happened.
Sun might need to get out of the cheapo 1U, $2000 "server" market and clean house Concentrate on selling expensive, quality hardware to people who can afford them. If whatever your sever is doing generates real money buy sun you won't regret it.
What a load of nonescense! On paper yes Sun did come out $2 billion down last year. Want to know why? Four(?) companies it had previously bought which it had to write down its valuations of. Take a look at its accounts, it took a charge of $2.125 billion for "Impairment of goodwill and other intangable assets". Read this to mean "some accountancy stuff that doesn't mean diddly to the companies operations". Thats it, period. It was a brave, forthright financial disclosure it could have put off for a couple of years, or dripped out, but it wanted to do a clear-out, get all the bad news out in a one-er and be able to post figures uncontaminated by that stuff from now on. The actual operations made money, I think it was around $10 million. Granted for a company of its size this isn't much, but this is the figure to look at. They increased cash reserves, its only the companies paper valuation that dropped $2 billion, they didn't actually loose any money. According to the basis of Cringely prediction, Sun continuing on exactly the same path, the market doesn't get any better, etc, etc, in 5 years time it will only have $50 million more in the bank than today. Does someone want to explain to me how this means its going to fold? IANAFA (I am not a financial analyst) but that sounds like bullshit.
Sun has a huge cash reserve, $1.5 billion, another $1 billion in stocks and short term securities, and other bits and pieces. Add all the assets together, excluding plant, 'intangables' and the like and its got $8.3 billion it could pull together if necessary. Oh, and it has no debt at all, period.
Cringley strikes me as a very poor journalist, he didn't even take the time to look into the basic details of the recent accounts, or if he did he was incapable of understanding them. Why does anyone bother reading this cretins opinions, he does seem to have a track record of being unnecessarily sensationalist and outstandingly wrong.
Disclaimer - I work for Sun as an engineer. Whilst I can't say too much on this topic I would say this year is looking pretty good thank you very much. The views expressed here are my personal opinions.
Stealing a rhinoceros should not be attempted lightly.
As a current employee of Sun Microsystems, I can at least clear up one little factoid in the article that every Sun pundit likes to mention for dramatic impact without either understanding or wanting the reader to understand.
The so-called $ 2 billion loss was a one-time writeoff that had to do with the revaluation of various companies that Sun acquired. People who bother to research their facts rather than simply spit them back verbatim for shock value would see that this is something that many companies do, and is more a sign of the bad economy than necessarily bad management at Sun. Without that write-off, Sun would have made a small profit.
I can't really comment on the other points in the article, since a lot of it is subjective, and anything I might say on it would be inherently biased by the fact that I work for Sun.
Karma: Frotzed (mostly due to the Frobozz Magic Karma Company)
let's start with the facts.
He says the battle will be lost to cheap stations that use AMD and Intel hardware.
No sane company at this point is going to put mission critical applications on Intel hardware unless it scales horizontally amazingly well. weblayers - yes. application servers - yes. big databases that are read-write? NO.
And who is going to provide the 1-hour onsite response time that comes with Sun's Platinum service for those flocking to cheap hardware?
Sun's legacy servers (4500, 6500, e10k) were pretty amazing, but had some faults (Ecache failures, lack of true power redundancy, etc). But Sun's new line of servers is truly amazing. The 4800, 6800, and e15k all support true partitioning, including FULL separation of power circuitry between partitions! as with the last line, they are very interchangeable with each other. now add solaris, an OS that is stable, and scales extremely well up to 106 processors and 512 GB of RAM in one machine (read again, that is 512 GIGS of RAM). did I mention hot-swappable CPUs? did I mention that Sun's partnership with Hitachi lead to Sun's offering of 75 Terabyte SAN-attached arrays?
So, Mr. Cringely, who exactly is going to fill this gap for Enterprise servers for mission critical apps if Sun tanks?
But yet he claims that Sun has "no real technical leadership". how about that. so they dont. most companies with "real technical leadership" sit on the sidelines and daydream about marketing products with this kind of quality.
I guess if Sun tanked, people could still buy IBM or HP hardware and run (gasp) AIX or HPUX. I've been responsible for AIX in my life, and it's not really pretty. And IBM's linux offering on mainframes seems as absurd to me as spending the money for a twin-turbo porsche and then asking for vinyl seats because you don't like the feel of real leather.
In a sense, I'm biased because I have built my career around being an expert in Sun hardware, Solaris, Veritas tools, and Perl. But then, this is exactly why I am able to know how big corporations think. CTO's aren't wandering from big UNIX machines for awhile when it comes to anything important...
A year spent in artificial intelligence is enough to make one believe in God.