All the Failure Money Can Buy
I usually stay away from the popular news of the day but this one was so funny I had to say something about it: IBM in Talks to Buy Sun in Bid to Add to Web Heft (Note: This link keeps coming up with an excerpt and a “subscribe” link, but when I opened it from Google search results it gave me the whole article. Interesting, huh?). The short version of this is that the two companies are in talks and there’s a possible price tag of $6.5 billion. As of this writing, no actual deal has been announced. They’re just talking. Interestingly, according to the WSJ article, Sun has been looking for a buyer for some months. This makes sense to me since I’ve considered Sun to be doomed for several years.
I’ve worked with Sun’s Solaris operating system quite a bit. I like it. It’s terrifically stable. In fact, the times when it seemed to fail there was always an underlying hardware problem. Sun hardware is pretty good, too. Some of their Sparc stations run practically forever. That said, I don’t see Solaris to be all that much better than Linux and it costs MUCH more (On any hardware. Linux has been ported to everything except my mailbox but it’s not much of a mailbox). The extra cost for Sun just isn’t worth it.
When I worked at the newspaper the editorial software we put in during my first year ran on Solaris 8 on a pair of V880s the size of one of those mini office refrigerators. At the time, that editorial system depended on Oracle for the db and was not available on Linux. It was a powerful, very stable and fabulously expensive system. That editorial software is now available on Linux. It’s still fabulously expensive but the cost of the hardware and the operating system are a fraction of what they used to be.
Too bad for Sun. Great for the vendor’s customers (I’m avoiding mentioning the vendor’s name because I prefer to keep the amount of profanity on this blog down, btw. They were not the best vendor I ever saw, though they were a lot better than the worst). Sun itself seems to have recognized this since, starting with Solaris 9, they began incorporating Gnu utilities and features in the base system. In other words, they tried to win back users who were going to, or interested in, Linux, by becoming more Linux-like.
Making one of your major products more like something that’s free in order to compete is a sure sign of a problem. A BIG problem. So this leads us to the question: What on Earth does IBM think they will gain from buying Sun??
Remember that the Wall Street Journal article says that IBM intends to increase its “web heft” whatever the hell that means. It says it right there in the headline. But since the term “web heft” has never been used in conjunction with Sun and never will be (and not just because it’s a newly made up term that doesn’t have any meaning and doesn’t make any sense), the question still remains what does IBM think Sun has that’s worth spending billions of dollars on?
Don’t say Java. When Java was first rolled out, they told us it was going to conquer the web. It didn’t because but it was too heavy both to learn and to run. Java is obsolete. Forget it. I’m aware though that people’s opinions vary and it’s possible that the suits at IBM actually think that Java adds to Sun’s value. If this is true, that would not be a good sign for IBM’s future health, either.
Maybe “web heft” is code for “The economy sucks and we think we can get a fire sale price.” That would make some sense to me. Not much but some. There are those who, unlike me, believe an IBM-Sun merger is a good idea. See Why an IBM purchase of Sun would make sense for example. And for a more balanced consideration, try Is it a bad idea for IBM to buy Sun?
To me it seems that the overwhelming majority of analyses of a possible merger between these two giants are ignoring the glaringly important problem that these kinds of mergers (or buyouts, or whatever you want to call them) have a terrible track record of success. See Why mergers fail for a good take on this. There are lots of others.
The upshot is that no matter how good the products or how successful the corporate cultures or even how dynamic the “synergies,” really big companies don’t merge well. I remember a story about the merger of two department store chains that wound up failing because their computer systems – including things like inventory control, and low level stuff like hardware interfaces – were too dissimilar. They spent a ton of money trying to get them to talk to each other but never succeeded.
The story may be apocryphal but it’s plausible. A huge percentage of big tech projects fail, too.
LinkedIn
Technorati Favorites