Java 6 Update 2 Rolled Out With Show Stopper Bug
As soon as people got their hands on Update 2 of Java 6 at the beginning of July, people running Java on Windows found what I would say is a show stopper bug. It’s not surprising it didn’t take long to show - it’s a biggie. I noticed it myself within a few seconds of installing the update. The bug is - the FileChooser is often so slow that it renders any application where the user needs to open a file totally unusable. See the discussion that took place on JavaLobby. Here’s the filed bug, which doesn’t really indicate how serious this problem is; honestly, this one is a show stopper.
Anyway, I had assumed that Sun would not roll Update 2 out for automatic update on Java.com. But it looks like they have. Doesn’t seem the right move to me. They should have fixed this bug before updating everyone’s runtime. Let’s be clear - Windows is the major client platform for Java. It’s not OK to ship with problems like this.
I wonder if too many of the key people in Sun’s Java GUI team use Macs, and so don’t really care about what is actually the most important platform for Java as far as GUIs are concerned? If that’s the case, this doesn’t bode well for the Consumer JRE. This is really “IT Strategy 101″ stuff - your core development platform should be your most important deployment platform.
Hari Jayaram wrote:
Wow this is a biggie! , whats also annoying is that if what you say about most of the developers at Sun using macs is true, then why is the apple official Java 6 on Mac still stuck at the developer preview b88 stage. You would think the powers that be at SUn would prod apple along to updating Java on their favorite platform
Equally annoying was the recent OSX Quicktime 7.2 update killing microsoft office 2004 apps on the mac if you had Java 1.6 installed. Those apple guys possibly were developing on Windows!
Posted 01 Aug 2007 at 4:43 pm ¶
simon wrote:
Hari, I don’t know if most of the developers at Sun use Macs. I was just wondering if some key people do, because I struggle to understand how this problem could have made it through the door.
Posted 01 Aug 2007 at 5:27 pm ¶
Asam Bashir wrote:
LOL - Use of Macs at Sun leads to quality-control problems with Java on Windows.
Well at least they’re not too dumb then - even if they are using Macs they have access to test Windows code in a VM or dual boot.
Posted 02 Aug 2007 at 12:24 pm ¶
Aaron wrote:
Interesting points. However, I would guess that the developers were on a *nix based system for either of two reasons: security or because the majority (I assume) of Java Applications are running on *nix/Sun servers. I agree with you though, but think it should have been using the core platform. Just my two cents, but great blog.
Posted 04 Aug 2007 at 3:08 am ¶
simon wrote:
Aaron, yes, the majority of Java software on servers may be running on Unix or Linux. And certainly, I would expect Solaris to be the first-tier development platform for Java on the server side inside Sun. I was actually meaning Swing client applications (I should have been more specific). I see a lot of Swing “thought leaders” at Sun choosing to use Mac OS X, rather than Windows.
Posted 04 Aug 2007 at 3:20 pm ¶
Asam Bashir wrote:
Just a small insignificant point, Leopard is a full certified port of UNIX 03 now,
http://www.pcworld.com/article/id,135352-page,1/article.html
*nix now = Mac OS X
Posted 07 Aug 2007 at 4:20 am ¶
Asam Bashir wrote:
http://www.mollypages.org/
Posted 21 Aug 2007 at 8:18 pm ¶