Slashdot Mirror

← Back to Stories (view on slashdot.org)

MacFixIt Details Mac OS X 10.2.8 Bugs

Posted by michael on from the pioneers-get-arrows-in-back dept.

mneptok writes "Premier Macintosh troubleshooting site MacFixIt has just posted a detailed report on the bugs and broken features in Apple's latest point release for MacOSX. As reported previously on Slashdot, the 10.2.8 update was released and pulled within hours earlier this week. Many users upgraded before the update was pulled and are being bitten, and MacFixIt has run down the behavior you can attribute to Apple's goof."

3 of 333 comments (clear)

  1. Another reason to end developer seeds? by buckhead_buddy · · Score: 5, Interesting

    Apple has been hinting of late that it's becoming more uncomfortable with seeding pre-release software to developers. Routinely, leaks of the new features and builds in the latest versions of the software make it to the mac rumor sites within hours of being made available in the seed channels.

    Of course the argument was that it was a necessary evil to put up with this stuff because the feedback that Apple received from developers (both in quality and quantity) helped catch glaring bugs.

    10.2.8 news has been rampant through the rumor channels for a LONG time. Now that it's finally out, there are quick glaring holes that cause the recall of the update and lots of bad publicity for Apple.

    I certainly hope Apple doesn't get paranoid about the release of new software to paying seed developers, but this is just another reason that the scale may be tipping toward the paranoid, closed-testing route than the limited open-seeding way it is today.

    1. Re:Another reason to end developer seeds? by King_TJ · · Score: 5, Interesting

      Well, what disturbed me most was the comment someone made over on MacFixIt when 10.2.8 was first released. He claimed he had good friends working at Apple who told him that the "in house" word was, the new 10.2.8 release still needed some tweaking, and wasn't really quite ready for public release.

      Since it turned out to have all these issues and got recalled, it looks like this guy was correct.

      Assuming this is true, why did Apple go ahead and push 10.2.8 out the door when their own employees knew it wasn't ready yet?

      My theory is, it contained security fixes such as patches for SSH - and Apple felt they couldn't risk waiting any longer to roll it out, and have a Microsoft-like fiasco develop where OS X got exploited before fixes were available.

      This probably begs the question: Should security patches/fixes always be released by themselves, instead of rolled into general system updates? It may not be a bad idea.

      As far as Apple getting paranoid about giving out pre-release software to developers, I think that's mainly due to all the Safari builds that leaked onto Usenet. Apple hates not being able to surprise people with cool new features, and that element of surprise keeps getting stolen out from under them when code gets leaked. (We all knew about Safari adding tabbed browsing long before Apple announced it, for example.)

      Still, you have to balance those concerns with the risk of not having enough people testing/pre-screening your code for serious problems. In the end, writing a solid, bug-free product is the most important goal.

  2. Apple's patch strategy needs work by Otterley · · Score: 5, Interesting

    I run a lab with an OS X server which relies on ssh for remote access, and all I wanted was a point patch to fix the ssh security hole announced two weeks ago.

    However, Apple failed to provide us one. Instead, they rolled the patch into the 10.2.8 release, thus exposing anyone who upgraded solely for the security fix to additional instability caused by changes to other OS components.

    I feel sorry for those server administrators who were unable to upgrade to 10.2.8 because it means the server is still vulnerable to the ssh security hole.