Tiger Tomorrow?

It’s been rather quiet on the Mac front. Well, that’s not entirely true, we have the continuing saga of Apple versus the rumor sites. Very little has been decided in the courts, but that hasn’t stopped the Mac community at large from giving their fifteen cents. Unfortunately, the battles being waged are being clumped into one, making the debate wandering and unfocused. We have one group shouting “This is a freedom of speech issue!” while the other says (quite accurately) “This particular decision is not.” A pissing contest ensues.

But I digress. That quiet seems to be the calm before the storm that will be the release of Tiger. It’s making me nervous, much like being alone in a house for the first time. Boo! With rumor sites including the now infamous Think Secret predicting a release announcement tomorrow, the buzz may start to return to the Mac community. Then again, most of that buzz will likely be “Have you put your order in yet?” Now that I think about it, the release following the announcement by only a couple weeks seems a little strange. That leaves very little lead time for ordering. That’s significant because it will make it harder to estimate initial demand for the OS.

As for whether it’s appropriate to be releasing Tiger now, that’s entirely up to Apple and their software development people. However, by reading many messageboards, you’d think that the decision should be decided by Apple fans. We have overzealous fans who have heard through someone else that there are major bugs in Tiger and as such, they should delay until May.

There’s a few things wrong with that. First, it’s coming through a third-party source, which amounts to “friend of a friend of a friend” many times. “My brother knows this guy that has a friend that’s a developer that has access to Tiger builds and he says they are buggy” is not a reputable source for information. I’ve heard conclusively that Tom Cruise is gay that way.

However, let’s assume that the source isn’t made up. The next part of the argument that fails muster is the definitive way in which Tiger is being pronounced “buggy”. A developer is saying this. We are assuming that the program the developer is working on doesn’t make assumptions about the behavior of a particular API (which happens all the time) or use a deprecated API call. Maybe, just maybe, the program is buggy, and not the OS. This of course goes back to the fallacious logic gem “it worked before, therefore the blame lies with what changed”. This doesn’t always apply when programming.

If an API call is faulty in some way, we are making the gross assumption that its severity is high, affecting daily computing. Further, it makes the assumption that the bug actually causes something to fail or even does something that’s not cosmetic. Also ignored is the number of bugs. What about the complexity of the issue as far as diagnosing, fixing, and testing go? Software engineers working on Tiger itself are most likely incapable of properly estimating these things. Their managers might better be able to do this, but then again, it’s only an estimate (and may not be right).

That’s a whole lot of assumptions. I’ll put them into a list to better summarize them.

• The source is:
  • Trustworthy
  • Right
  • Aware of the scope of the issues
• The bugs affect usability.
• The bugs are non-trivial to:
  • Diagnose
  • Fix
  • Test

That seems like a rather tenuous position to take now, doesn’t it? I’d love to see Tiger’s release announced tomorrow, but I won’t freak out if it isn’t either.