I thought I read somewhere that with Snow Leopard (or was it Safari 4.0?) that when a plug-in crashes, as Flash does a lot, only the plug-in would crash and not all of Safari. Does anyone see this? My Safari is still crashing when Flash does.
MBP Core Duo (1,1)
10.6.2
Safari 4.04 (but 4.04 hasn't crashed yet)
Fully uninstall Flash then restart, then reinstall it. See if that helps.
Did anyone do a permissions repair after this update and find that a TON of permissions had to be repaired? I did. And I just repaired permissions the other day with the 10.6.2 update.
Only a few... like only enough that its gonna take 30 MINUTES TO FIX... what happened with this update...?
And I did a permissions repair before and after 10.6.2 (nothing came up either time)....
I just discovered that Webkit does not support MathML ( the xml tags used to render mathematical notation) Firefox does though. I just landed a nice job to write an application for such so I'm a little bummed. I usually use Safari but no dice on this job. For anyone one else interested, a new JS library is in the works to be released soon to fix the lack of support in Webkit. It is called MathJax
Seriously if this is something you think is important to support, you need to log it with Apples bug tracking system. Further letting the big guy at the top know what is important to you is very important.
I'm not saying you will see an instant change with respect to the support of MathML but you may have a chance of getting MathML on to a future feature set list. YOu might also want to look at WebKit betas, I don't use MathML myself so I'm not sure what they might be implementing. It is pretty obvious that Apples primary focus lately have been performance improvements and HTML5. However the development of Safari won't stop just because those items get fixed.
Simply put, you?re not just updating Safari.app or QuickTime.app when Apple?s issue an update, they are also updating the frameworks that are loaded when you start your machine.
So why does updating these frameworks require a reboot? Even the last iTunes update didn't require a reboot, whereas pre-Snow Leopard it would ask for a QuickTime update in lock-step and that would require a reboot.
So why does updating these frameworks require a reboot? Even the last iTunes update didn't require a reboot, whereas pre-Snow Leopard it would ask for a QuickTime update in lock-step and that would require a reboot.
Go to System/Library/Frameworks/. There is a WebKit and QuickTime framework. iTunes uses the QuickTime framework and others, but until those are updated iTunes doesn?t require and update. Now you are probably wondering why it?s a WebKit framework and not a Safari framework. Well, Safari is just and app but the WebKit engine is part of the OS. There is a reason why Mac OS X has long had options like QuickLook that allow for reading documents and playing media without launching a specific app. These are all part of the frameworks built into the subsystem. The solution is simple. If you don?t like an integrated OS don?t use Mac OS X. Apple is going to tie more things together, not less, so this will only become more of a problem for you int he future.
I was running the latest update a little while ago and it must be said it looks snappy. It makes me wonder just how fast Safari can get, it sure does make IE look really bad.
The next thing we could use is a Mail update. I'm still using Thunderbird on my Mac as it just seems more logical and stable.
Seriously if this is something you think is important to support, you need to log it with Apples bug tracking system. Further letting the big guy at the top know what is important to you is very important.
I'll direct the scientific community to review your proposal and advise that their suplication to the 'big guy' should include a prayer for html5 support of MathML.
I'm pretty sure that I've seen the WebKit MathML page at least 3 years ago for the first time and that it hasn't changed much since then. It doesn't look like it's high on the priority list of the WebKit developers (the science community will thank them for that).
Quote:
Fully? This Wikipedia page paints a different picture. Granted, it?s Wikipedia and it?s a complex page that could easily be out of date, but all the main browsers seem pretty on the ball.
2009-03-23 Microsoft released version 8 of Internet Explorer, with full CSS level 2 support, plus some internationalization features from level 3. (Windows, free)
Microsoft developed and submitted thousands of new tests for CSS 2.1 during the development of IE8 and caught the other browser vendors pants down after all of their self-proclaimed superior standard support. Instead of chasing after the newest shiny thing (CSS 3, CSS animation, WebGL, etc.) maybe the Gecko and WebKit developers should sit down and fix the existent and widely used technologies first. Actually, Microsoft provided a very good example how absurd the focus on the ACID test suites is as they cover only small subsets of the tested standards (something I now at least some Gecko developers complained about).
Quote:
PS: Acid3 wasn?t designed to test for everything. It?s designed that way. Acid4 will likely work the same way.
See the above paragraph. It's actually almost ironic that Microsoft is doing the tiresome work of completing the CSS test suites, something I would have expected from our self-proclaimed champions of standard conformity for the sake correctness and verifiability.
Could somebody please explain to me why this update requires a reboot, despite the fact that no other browser on any other operating system I know of requires a reboot after install?
I'm on Snow Leopard, for God's sake! You'd think Apple would have worked out the legacy issues, including QuickTime 7 and any Carbon libraries.
Yeah, there's a minute and a half you'll never get back, huh? ... poor baby.
Did anyone do a permissions repair after this update and find that a TON of permissions had to be repaired? I did. And I just repaired permissions the other day with the 10.6.2 update.
I just installed and did permissions repair .... got nothing, nada, but I'm still on Tiger, if that means anything.
Go to System/Library/Frameworks/. There is a WebKit and QuickTime framework. iTunes uses the QuickTime framework and others, but until those are updated iTunes doesn?t require and update. Now you are probably wondering why it?s a WebKit framework and not a Safari framework. Well, Safari is just and app but the WebKit engine is part of the OS. There is a reason why Mac OS X has long had options like QuickLook that allow for reading documents and playing media without launching a specific app. These are all part of the frameworks built into the subsystem. The solution is simple. If you don?t like an integrated OS don?t use Mac OS X. Apple is going to tie more things together, not less, so this will only become more of a problem for you int he future.
Again, why do updates to the framework require a reboot? It's not like a security or system update where the kernel itself is being overwritten.
A true Unix geek brags about their uptime, and OS X is Unix with a poor uptime
Seriously, anyone who brags abut uptime has got be a complete douche.
Quote:
Originally Posted by JavaCowboy
Again, why do updates to the framework require a reboot? It's not like a security or system update where the kernel itself is being overwritten.
Let try to make this even simpler. Frameworks are loaded into memory when an application needs them. Safari is not the only application that uses the WebKit framework. They are shared across many applications. You would need to make sure that all applications that were using the framework have quit before being able to restart a new Framework. Do you not see how this can be an issue for a consumer machine if a Framework was forced quit while another app was using it? It?s all pretty simple stuff. It?s a system level component, so the proper protocol is to have the user restart the system when it?s convenient for them.
Seriously, anyone who brags abut uptime has got be a complete douche.
Let try to make this even simpler. Frameworks are loaded into memory when an application needs them. Safari is not the only application that uses the WebKit framework. They are shared across many applications. You would need to make sure that all applications that were using the framework have quit before being able to restart a new Framework. Do you not see how this can be an issue for a consumer machine if a Framework was forced quit while another app was using it? It?s all pretty simple stuff. It?s a system level component, so the proper protocol is to have the user restart the system when it?s convenient for them.
So why not ask them to log out and log back in? That would seem to be the simpler solution than forcing a reboot, and would be much less of a pain.
If it weren't for all the reinstalls that required reboots, no. Otherwise, I'd never have to reboot OS X, contrary to XP and previous versions of Windows, where practically every application install requires a reboot, and the colloquial "solution to all Windows problems" is a reboot. By contrast, only kernel updates seem to require a reboot in Linux.
So why not ask them to log out and log back in? That would seem to be the simpler solution than forcing a reboot, and would be much less of a pain.
Quote:
Originally Posted by JavaCowboy
If it weren't for all the reinstalls that required reboots, no. Otherwise, I'd never have to reboot OS X, contrary to XP and previous versions of Windows, where practically every application install requires a reboot, and the colloquial "solution to all Windows problems" is a reboot. By contrast, only kernel updates seem to require a reboot in Linux.
Mac OS X is not Linux.
Mac OS X is not trying to be Linux.
Mac OS X is a consumer OS built to sell Apple’s PCs.
Apple has made some very distinct changes in the way system level components are installed between Leopard and Snow Leopard. With Snow Leopard these major updates are done without the user being able to use the system.
Logging out doesn’t mean that frameworks are still not being utilized. If you put yourself in Apple’s shoes as a company supporting consumers perhaps you can imagine this as a fail safe to make OS upgrades less problematic. What you call “simpler” may not be the case for Apple’s technical support. If you’ve ever taken a business courses you may understand some of the pitfalls of growth.
It’s a bit annoying to have to do, which is why I don’t do an update because it’s there. Apple clearly informs you which updates require an restart. If you don’t like it then don’t update or don’t use the OS. They have decided the how and why and I’m certain it was done for actual reason that inevitably has a financial reason more so than to piss you off.
Comments
I thought I read somewhere that with Snow Leopard (or was it Safari 4.0?) that when a plug-in crashes, as Flash does a lot, only the plug-in would crash and not all of Safari. Does anyone see this? My Safari is still crashing when Flash does.
MBP Core Duo (1,1)
10.6.2
Safari 4.04 (but 4.04 hasn't crashed yet)
Fully uninstall Flash then restart, then reinstall it. See if that helps.
Did anyone do a permissions repair after this update and find that a TON of permissions had to be repaired? I did. And I just repaired permissions the other day with the 10.6.2 update.
Only a few... like only enough that its gonna take 30 MINUTES TO FIX... what happened with this update...?
And I did a permissions repair before and after 10.6.2 (nothing came up either time)....
I just discovered that Webkit does not support MathML ( the xml tags used to render mathematical notation) Firefox does though. I just landed a nice job to write an application for such so I'm a little bummed. I usually use Safari but no dice on this job. For anyone one else interested, a new JS library is in the works to be released soon to fix the lack of support in Webkit. It is called MathJax
Seriously if this is something you think is important to support, you need to log it with Apples bug tracking system. Further letting the big guy at the top know what is important to you is very important.
I'm not saying you will see an instant change with respect to the support of MathML but you may have a chance of getting MathML on to a future feature set list. YOu might also want to look at WebKit betas, I don't use MathML myself so I'm not sure what they might be implementing. It is pretty obvious that Apples primary focus lately have been performance improvements and HTML5. However the development of Safari won't stop just because those items get fixed.
Dave
Simply put, you?re not just updating Safari.app or QuickTime.app when Apple?s issue an update, they are also updating the frameworks that are loaded when you start your machine.
So why does updating these frameworks require a reboot? Even the last iTunes update didn't require a reboot, whereas pre-Snow Leopard it would ask for a QuickTime update in lock-step and that would require a reboot.
So why does updating these frameworks require a reboot? Even the last iTunes update didn't require a reboot, whereas pre-Snow Leopard it would ask for a QuickTime update in lock-step and that would require a reboot.
Go to System/Library/Frameworks/. There is a WebKit and QuickTime framework. iTunes uses the QuickTime framework and others, but until those are updated iTunes doesn?t require and update. Now you are probably wondering why it?s a WebKit framework and not a Safari framework. Well, Safari is just and app but the WebKit engine is part of the OS. There is a reason why Mac OS X has long had options like QuickLook that allow for reading documents and playing media without launching a specific app. These are all part of the frameworks built into the subsystem. The solution is simple. If you don?t like an integrated OS don?t use Mac OS X. Apple is going to tie more things together, not less, so this will only become more of a problem for you int he future.
The next thing we could use is a Mail update. I'm still using Thunderbird on my Mac as it just seems more logical and stable.
Dave
there was a HUGE amount of permission problems with mine...seriously seemed like atleast 100
Just language localizations and such within the app contents. Nothing to be concerned with.
Seriously if this is something you think is important to support, you need to log it with Apples bug tracking system. Further letting the big guy at the top know what is important to you is very important.
I'll direct the scientific community to review your proposal and advise that their suplication to the 'big guy' should include a prayer for html5 support of MathML.
Firefox looks to only support partial MathML. If you want real support you?ll have to go with Opera. Note that it?s coming to WebKit.
I'm pretty sure that I've seen the WebKit MathML page at least 3 years ago for the first time and that it hasn't changed much since then. It doesn't look like it's high on the priority list of the WebKit developers (the science community will thank them for that).
Fully? This Wikipedia page paints a different picture. Granted, it?s Wikipedia and it?s a complex page that could easily be out of date, but all the main browsers seem pretty on the ball.
http://www.w3.org/Style/CSS/
2009-03-23 Microsoft released version 8 of Internet Explorer, with full CSS level 2 support, plus some internationalization features from level 3. (Windows, free)
Microsoft developed and submitted thousands of new tests for CSS 2.1 during the development of IE8 and caught the other browser vendors pants down after all of their self-proclaimed superior standard support. Instead of chasing after the newest shiny thing (CSS 3, CSS animation, WebGL, etc.) maybe the Gecko and WebKit developers should sit down and fix the existent and widely used technologies first. Actually, Microsoft provided a very good example how absurd the focus on the ACID test suites is as they cover only small subsets of the tested standards (something I now at least some Gecko developers complained about).
PS: Acid3 wasn?t designed to test for everything. It?s designed that way. Acid4 will likely work the same way.
See the above paragraph. It's actually almost ironic that Microsoft is doing the tiresome work of completing the CSS test suites, something I would have expected from our self-proclaimed champions of standard conformity for the sake correctness and verifiability.
Could somebody please explain to me why this update requires a reboot, despite the fact that no other browser on any other operating system I know of requires a reboot after install?
I'm on Snow Leopard, for God's sake! You'd think Apple would have worked out the legacy issues, including QuickTime 7 and any Carbon libraries.
Yeah, there's a minute and a half you'll never get back, huh? ... poor baby.
Did anyone do a permissions repair after this update and find that a TON of permissions had to be repaired? I did. And I just repaired permissions the other day with the 10.6.2 update.
I just installed and did permissions repair .... got nothing, nada, but I'm still on Tiger, if that means anything.
Yeah, there's a minute and a half you'll never get back, huh? ... poor baby.
A true Unix geek brags about their uptime, and OS X is Unix with a poor uptime
Go to System/Library/Frameworks/. There is a WebKit and QuickTime framework. iTunes uses the QuickTime framework and others, but until those are updated iTunes doesn?t require and update. Now you are probably wondering why it?s a WebKit framework and not a Safari framework. Well, Safari is just and app but the WebKit engine is part of the OS. There is a reason why Mac OS X has long had options like QuickLook that allow for reading documents and playing media without launching a specific app. These are all part of the frameworks built into the subsystem. The solution is simple. If you don?t like an integrated OS don?t use Mac OS X. Apple is going to tie more things together, not less, so this will only become more of a problem for you int he future.
Again, why do updates to the framework require a reboot? It's not like a security or system update where the kernel itself is being overwritten.
A true Unix geek brags about their uptime, and OS X is Unix with a poor uptime
Seriously, anyone who brags abut uptime has got be a complete douche.
Again, why do updates to the framework require a reboot? It's not like a security or system update where the kernel itself is being overwritten.
Let try to make this even simpler. Frameworks are loaded into memory when an application needs them. Safari is not the only application that uses the WebKit framework. They are shared across many applications. You would need to make sure that all applications that were using the framework have quit before being able to restart a new Framework. Do you not see how this can be an issue for a consumer machine if a Framework was forced quit while another app was using it? It?s all pretty simple stuff. It?s a system level component, so the proper protocol is to have the user restart the system when it?s convenient for them.
and OS X is Unix with a poor uptime
Really?
Seriously, anyone who brags abut uptime has got be a complete douche.
Let try to make this even simpler. Frameworks are loaded into memory when an application needs them. Safari is not the only application that uses the WebKit framework. They are shared across many applications. You would need to make sure that all applications that were using the framework have quit before being able to restart a new Framework. Do you not see how this can be an issue for a consumer machine if a Framework was forced quit while another app was using it? It?s all pretty simple stuff. It?s a system level component, so the proper protocol is to have the user restart the system when it?s convenient for them.
So why not ask them to log out and log back in? That would seem to be the simpler solution than forcing a reboot, and would be much less of a pain.
Really?
If it weren't for all the reinstalls that required reboots, no. Otherwise, I'd never have to reboot OS X, contrary to XP and previous versions of Windows, where practically every application install requires a reboot, and the colloquial "solution to all Windows problems" is a reboot. By contrast, only kernel updates seem to require a reboot in Linux.
So why not ask them to log out and log back in? That would seem to be the simpler solution than forcing a reboot, and would be much less of a pain.
If it weren't for all the reinstalls that required reboots, no. Otherwise, I'd never have to reboot OS X, contrary to XP and previous versions of Windows, where practically every application install requires a reboot, and the colloquial "solution to all Windows problems" is a reboot. By contrast, only kernel updates seem to require a reboot in Linux.
Mac OS X is not Linux.
Mac OS X is not trying to be Linux.
Mac OS X is a consumer OS built to sell Apple’s PCs.
Apple has made some very distinct changes in the way system level components are installed between Leopard and Snow Leopard. With Snow Leopard these major updates are done without the user being able to use the system.
Logging out doesn’t mean that frameworks are still not being utilized. If you put yourself in Apple’s shoes as a company supporting consumers perhaps you can imagine this as a fail safe to make OS upgrades less problematic. What you call “simpler” may not be the case for Apple’s technical support. If you’ve ever taken a business courses you may understand some of the pitfalls of growth.
It’s a bit annoying to have to do, which is why I don’t do an update because it’s there. Apple clearly informs you which updates require an restart. If you don’t like it then don’t update or don’t use the OS. They have decided the how and why and I’m certain it was done for actual reason that inevitably has a financial reason more so than to piss you off.