18 comments

[ 4.4 ms ] story [ 51.1 ms ] thread
Is this really suprising? A boxed release has to be done weeks (months?) before it ships. Its not like they stop working on it after it gets done.
NOW I'll consider installing it.
It's a bit arbitrary, no.

If you look at the list of fixes in 10.6.1, do you see anything that would've kept you from making 10.6.0 work on your Mac?

Oops, no sarcasm tag. :-)
Although compay is now repudiating his comment, I'll bite: Even if there's nothing that was fixed in 10.6.1 that would have been important to me, the fact that we got to the first point release without such a fix is better evidence that there was no such problem in 10.6.0, something I would only know, otherwise, had I taken the chance and installed it. Therefore, in the sense that matters, it doesn't make any difference whether there was an important fix in this release, because if there was, it was fixed, and if there wasn't, it's evidence that there was no problem to fix.

Either way, it pays to wait if you're risk averse. And I am.

I'm not going to install it. I'm going to wait until the next MBP refresh, then buy that and manually copy my files. At that point I will rebuild all my unix stuff from scratch and put it in ~.

Right now I'm running a byzantine mess of a 10.3 upgraded to 10.4 upgrade to 10.5 and then transferred through 2 machines via Time Machine.

I totally agree that it's a good rule of thumb to always at least wait for the first point release.

But it just got me thinking that we must understand these heuristics and not just follow them blindly.

I wasn't saying the OC did a bad thing, just thinking out loud about how "wait for the first point release" can be arbitrary since with such a popular software, you could read the feedback on various forums to see how stable it is.

If you're going to be having a major problem, chances are others will have it too. And if it's a minor problem that almost nobody has except you, it might not be fixed in the first point release.

I was planning to wait until SIMBL was working on 10.6, but it appears that it already is if you tell the app it's modifying to open in 32-bit, so I might upgrade shortly.

I agree that with such widely distributed software, browsing forums and news aggregators might well be as useful as waiting to see what's fixed.

SIMBL will probably fall into obsolescence with input managers being disabled for 64-bit apps.

PlugSuit, on the other hand, is SIMBL-compatible and uses a completely different method of injection. The author just needs to work around some new limitations introduced in Snow Leopard before it works with 64-bit apps. I would keep your eyes peeled for an update.

If only they would enable copy-on-select as a regular preference for Terminal, I'd have no further need of SIMBL. :)
I think Apple wanted to play down releasing a security update so shortly after the initial release because the media went crazy about 10.6 shipping with a non-secure Flash player.

So they mixed in a few small improvements and shipped it as a point release. It played well, I haven't seen any articles in e.g. computerworld that Apple finally fixed the security problem.

For what it's worth, I'm running 10.6 on one of my Macs, and I still have some Ports that don't work on it.

In particular, I've had issues with ghc and R. Most other software (both common and uncommon) has been okay, but I'd recommend most people wait another month or so, unless 10.6 fixes something important to you, or is otherwise significant.

ghc is one of a handful of MacPorts which does not work with Snow Leopard, although this should be fixed soon (32-bit only, though): http://trac.macports.org/ticket/20132
I have a friend who's been fighting through installing it from source because he's sick of waiting (apparently cabal was having trouble... not really sure). Hope macports gets things sorted out soon.
"Includes an update to Adobe Flash Player plug-in version 10.0.32.18"

How about just updating Safari?