John Brady's Weblog

I've laid off making comments about Sun for a while, but with the recent 2 announcements I just had to add my two pennies worth.


First Jonathan Schwartz decides to change the stock ticker symbol for Sun from SUNW to JAVA. Boy, you wonder how many MBA classes he studied to come up with that idea to save Sun and turn it back into a growing profitable company?

By his own admission, this change has no bearing or impact on the value of Sun as a company or of how people will decide on whether to purchase shares of Sun or not. So why bother? What kind of argument is he trying to put forward here? On the one hand this will have no net impact at all, but on the other hand ... nothing. So lets change our stock symbol anyway, at great cost (it cannot be free to do this), confuse everybody who knew our old symbol, and provide no benefit to anyone as a result. Must have been a quiet quarter there at Sun, Jonathan.

But the really amazing thing to me, is that Jonathan clearly does not understand what Sun is as a company, and how it makes profit and so a return for shareholders. It does this through being a computer hardware company. Sun gets revenue from selling computer hardware, and not from selling computer software. Yes, it has lots of software, but these don't make money for it.

Sun has open sourced and given away almost all of the software it has, and certainly all of the key ones. Sun's recent history includes a long list of software companies it has bought, not known what to do, failed to be able to make money from the software, and then given it away for free and tried to wash its hands of it. StarOffice, N1, and now Solaris and its Java Application Server (GlassFish). It has even given away the software it got from hardware companies it bought, such as Cobalt.

Sun also has bought many different software companies all doing the same thing. At one time it had 3 different Java Application Servers (NetDynamics, Netscape and Forte). It has bought several File System companies, only to do ZFS to replace all of them. And several Cluster related companies. So Sun has had a lot of software products over the years.

So, in spite of the fact that Sun only makes money from computer hardware and has pretty much given away for free all the software it has ever had, Jonathan thinks that the new stock symbol should be based on a piece of software. Solaris? No. Java.

Why pick on Java? In reality it is actually more of the odd one out. The one piece of new software that was successful, compared to all of the many, many others that have been quietly brushed aside and forgotten about. And Java is not even a product. It is a technology. Which was invented by engineers at Sun, but is now available within many different products from many different companies.

Jonathan has given this bizarre argument about how this new stock symbol will 'push the brand of Java', but Java isn't a brand it is a technology. It stands for a very specific piece of technology. And when he goes on about people using Java, the majority of them are not using it on Sun hardware - whether computers or mobile phones or anything else. So Sun is really a small part of the Java story now, and Java doesn't make any money for Sun but does for a lot of other companies. By using Java he is actually watering down and distracting from everything else that Sun does well, such as the new multi-threaded SPARC chips.

If I didn't know more about Sun, I would presume that this change in stock symbol was heralding its exit from computer hardware and signifying a move into pure software. Which couldn't be further from the truth. Sun only makes money from the computer hardware it sells, and nothing from the software it makes. The only money it makes from software is from selling support services for that software, which is really turning it into a subscription service rather than a product.


Second, Sun merges its Storage division with its Server division. So having paid $4 billion for StorageTek about 2 years ago, which presumably had good products and knew how to sell them, Sun has managed to destroy whatever added value was present in there and now has nothing it can do with it other than just throw it in with the Server division and lump them all together. Haven't we seen all of this before with other hardware companies such as Cobalt? So just why did you buy StorageTek in the first place then?

Personally I have always said that Sun has failed miserably in the storage marketplace, and it should just keep it simple and stop trying to play with the big boys. It tried to get into storage too late, after everyone else had made the move; tried to catch up but couldn't; and tried to make people believe it had truly open storage products when it didn't. It seems to have fooled itself into believing that if it told everyone often enough that Sun was good at storage, then eventually everyone would believe them and start buying their storage products. But it has never happened, and never will.

Simply put, Sun's storage products have never been truly open. They have never really worked in a truly multi-platform environment, and have mainly been made to work against Sun servers running Solaris. Any support of other platforms has always been very restrictive. And as a result, in spite of buying many companies and doing OEM deals, and launching product after product, Sun has never been successful in storage. Many of the comments on Jonathan's blog entry reflect this, and also the Register says it too.


And Sun continues to lay off people. And so it will go on.