Sun Posts Increasing Loss

Chromodromic writes "Sun Microsystems posted an increasing loss at a time when many tech firms are beginning to report stable or increasing earnings and stocks are looking up. According to the Wall Street Journal, it looks like Sun, the formidable peddlers of Solaris, Java, and UltraSPARC Fire servers are facing competition from measly ol' Dell and Intel. Even Scott McNealy has been reported to concede in a May 2002 meeting with top execs that Sun has to change, including building up trust with customers that have been put off by McNealy's sometimes controversial personality and Sun's reputed internal disarray which according to Merrill Lynch is indicating that Sun requires a makeover. The Merrill Lynch report was, in fact, particularly scathing and has raised a few Wall Street eyebrows."

Comments

Not surprising

[pagz]

My university's Laboratory for Computer Science did a test between a Sun machine and a IBM compatable running linux in order to see if they could justify the cost of buying new Sun machines like they always have. IIRC the Sun machine cost five times more and performed three times worse than the IBM.

This was on running code from the profs (so research code), which is mainly what the machines would be used for.

Well they deserve it

[arvindn]

As many have noted, Sun have never formed a coherent strategy about linux. Their statements re. linux seem to be a mix of hostility, skepticism and euphoria. Also, they have a finger in every pie without a clear vision of where they want to be in a market of ups and downs. And lately they have shown that they are not above cheap marketing gimmicks either -- witness the branding of Mad Hatter as the "Java desktop system" (its actually just another linux distro.)

I personally don't trust Merrill Lynch

[jsse]

but I do agree SUN is doomed.

I've been talking to a senior financial trader early this year, he said SUN's stock price is sky-rocketed to a point that they have to produce at $0 cost and sells for ten years to make up for the hyped value. Which is, of course, almost impossible.

Until recently I do believe SUN has already stuck one foot into its doom. As I speak we've already ruled out Solaris in several enterprise projects in favour of Linux. The cost of ownership is one factor, and the full-range maintenance supports from IBM, HP and Oracle is indeed a killer.

It's true that(don't flame) Linux has much to catch up with Solaris in enterprise deployment, but the market demand for Linux will only cause Linux to catch up fast and thus SUN's products will soon lose their market competitiveness very soon.

SUN's required fix

[Corpus_Callosum]

Sun Micrososystem is a company that built it's success through UNIX eliteism. Much like Apple, Sun was a company that you were proud to do business with. They had some of the greatest minds in the industry working for them (Bill Joy, James Gosling, ...), they sold the coolest hardware which often was even the fastest hardware (but not neccessarily - it was mostly fast enough). They had great support, etc... It was a COOL company to work for, with, or be a customer of.

Today, they are the same company they were 6 years ago. With the same operating system, the same hardware, but without the cool people and in fact without much at all that is still cool. The fact that they haven't changed with the times is exactly the problem.

In order for Sun to fix itself, it needs:

• A super cool, fast and cheap workstation. We are talking a cheap 4-way (or 8-way) Opteron with a 3D display or something similar. It has to be the best bang-for-the-buck on the market with features and "cool factor" that no-one else has. McNeally should walk across the street from the Cupertino campus and ask Jobs how to make this happen.
• To re-build their reputation as the price/performance leader. This is what kept their financial engines running strong through the 90s and they need to do it again. Even if they have to sell at cost in order to build the economy of scale, they MUST do this and do it NOW. They should shift to AMD processors in a huge way until their multi-core ultrasparcs hit, they should do whatever is neccessary. Period.
• They need to kiss and make-up with IBM. IBM can make a good partner for Sun. But Sun has alienated IBM and now IBM sees them as a pesky competitor instead of a competitive partner as Sun needs them to.
• They need a new center of gravity. Java was a perfect center-of-gravity for a long time. But Java is boring now. Nobody cares anymore... Sun has hundreds, if not thousands, of beautiful research projects that are sexy and cool... These generally stay research, which is unfortunate. They need to go harvest a couple of these and revv up their PR engines..

The greatest mistake that Sun can make right now is to assume that they will "pull out" of their death-spiral by making Java Desktops and waiting for the next generation of ultra-sparcs to hit. That is exactly how they can guarantee their own death. To live, they must kill their own business and allow the new, innovate stuff that they have in their labs to rise like a pheonix from the ashes of what was killed.