What Sun can do now
The new problems Sun's executive faces now include:
- Southeastern's attempt to sell the company out from under management
has given at least IBM and possibly others unprecedented access to Sun confidential
information - information a competitor should not have and which must now
- The publicity accorded the leaked information about the discussions has caused
significant damage to Sun's long term credibility as a key infrastructure supplier - creating
considerable FUD, delaying customer purchase decisions, and causing a lot of "market analysts" and
"journalists" to prove their commitments to the more powerful advertisers by presenting
vitriolic and uninformed commentaries
attacking Sun, its executives, and its customers -
this one, by Computerworld's Thibodeau , for example, ingratiates itself with just the right
people to look objective while managing all three forms of attack in only about 360 words.
Since the marketing organization fundamentally focuses on making large discounted sales to a
small group of big customers and, unfortunately, these are the buyers most likely
to make long term changes in response to precisely the kind of FUD loosed by this debacle, there's a real
risk the impact will be amplified across the next few customer budget cycles.
- a lot of valuable Sun people, and not many deadweights, are updating their resumes
and talking to friends about new opportunities.
And, in addition, none of the old problems have gone away:
- quarterly performance pressure drives many executive decisions, forcing short
term responses -like the volume customer focus in sales and announced layoffs-
sabotaging long term strategies;
- the Wall Street shell game in which players sell blocks of Sun shares
at a small loss per share to drive up the value of a collaborator's investments in competitors,
continues - and, of course, the only guarantee coming out of the "indigo" mess is
that, absent significant change, variations on this will happen again.
- most of the market, and many among Sun's own staff, have no idea how the SPARC
products line up against x86; no appreciation for the differences between Solaris
and Linux; and no interest in learning how unloved products from Sun Ray to
the CMT cryptology processors can produce huge savings and/or productivity
improvements for customers.
There's a vicious cycle here: as quota pressure drives salespeople to sell what's selling,
and mass media support drives customers to favor dumb solutions, the people who should
be explaining the cost, security, and productivity advantages that go with SPARC/Solaris
become increasingly unwilling and unable to do so - and so more and more SPARC sales go
to fewer and fewer large customers whose technical strengths lead them to demand the
product - and whose technical people are under pressure from within data processing to
buy IBM and from the executive suite to buy Wintel.
The basic bottom line on strategies to date is that appeasing Wall Street doesn't work, that
selling to a diminishing number of ever larger customers invokes enormous business risks, and
that having great products doesn't count for anything if neither the customer nor your own sales
people want to know about them.
So what to do? How about:
- spin off the x86 business as a separate company - tied to the original through
agreements on early technology access, OpenSolaris, and big customer marketing;
- look into the feasibility of setting up the open source software businesses
as a charitable foundation;
- take the rest of the company private via a leveraged employee buyout;
- build a new marketing division aimed at selling in ones and twos to smaller
- work with the tens of thousands of "joe's computing" shops run by capable people
around the world to sell directly to smaller companies looking for precisely the
savings, power, reliability, and security offered by the combination of SPARC/Solaris hardware,
open source software, and local support;
- train up "local joe" by providing extensive business and technical support;
- motivate "local joe" by helping him build pride in his work while
selling more support and services to the customer - paid for
by the customer's savings over the default Wintel decision.
- develop a couple of longer term, high visibility, showcase projects for both the technology and
the company; and,
- launch one other, major, initiative I don't want to discuss in public.
And one more thing: a lot of people don't share my interest in Sun technologies and tell me
they'd just as soon see the company go under. Setting aside those with personal motivations
(usually some form of fear of competence) for that position, most of the people I've talked to
who believe this won't affect them simply haven't thought it through.
The big reason you should be working to help Sun survive and prosper even if you never want to
use any open source product it contributes to, or any product it sells,
is that competition is good for the industry.
Witness, for example, Intel's "new" Nehalem line - it's
great right? By Wintel standards, it is - but if Sun hadn't lent AMD some people to design
x64 and then supported the company by building its own motherboards to demonstrate what x86 multi-core
could do, you'd be paying HP's prices for Itanium desktops - Itanium performance, and Itanium security.
If Sun dies, a lot of innovation dies with it - and you'll be paying a lot more for a lot longer
to Intel, IBM, and Microsoft.
So what's the bottom line? Simple, Sun can get its act together - and you, whoever you are, should
help because their success, even if you never buy from them, forces others
to lower their prices while increasing the value
you get from the people you do buy from.
Paul Murphy wrote and published The Unix Guide to Defenestration.
Murphy is a 25-year veteran of the I.T. consulting industry, specializing in Unix and Unix-related