Seriously can we put this statement to bed yet? It has been several years (think, five or so) since this statement has even been slightly accurate. Yes, many writes can destroy a drive, but the number is in the (upper) hundreds of millions - performed on one single sector.
Today flash hard drives levy on technology used in older embedded devices that relied on flash, called "wear leveling".
Because each write is spread out throughout the entire disk, you don't physically write to the same sector X thousands of times when updating a cache file or whatnot.
Even if you had something thrashing the SSD continuously, you would not destroy the drive within the reasonable lifespan of a comparable rotating media drive.
Why your post is modded insightful, I have no idea.
>> First, if it was really OSX, why would they need Google's help to implement Google Maps? It would just run.
Why did any company need Apple's help and vice versa with the transition to a new processor? If it's a different architecture, then there is a good chance some porting techniques will need to be applied and Apple and Google got together to do just that.
>>Second, the interface is obviously significantly different.
Define significantly different. What would OSX look like if you replaced Finder with X application managing the phone capabilities? No one knows. You can't really say at this point it is significantly different.
>>Third, it's hard to believe a handheld would have the resources to run OSX.
Why? Have you ever run OSX on a G3 iBook ~ 500Mhz? Some of the fancy things are gone, but it's still gorgeous to look at and quite usable. In four years time it's not hard to imagine that a dedicated graphics accelerator in conjunction with the (whatever) processor could easily outdo what an old iBook could do, at 640x480 resolution
nonetheless.
>>Finally, if it was really OSX, then any OSX app would run on it (in theory).
Why can't it? Checkbox Intel, PPC, and perhaps ARM architecture in Xcode and compile your application. Surely there are going to be some gotchas, but after the first transition and everyone using Xcode to write UB apps, if the code was designed for portability, it should just work.
Well I love Ubuntu, but the upgrade was *not so painless* for me and a few other users according to the forums. The upgrade process seems to fail on specific motherboards upon detecting that PCMCIA does not exist. If you are upgrading from Breezy, the kernel has no support to restart the detection or for this instance, fail out. Upon restarting the system, you are left with a hodge-podge Breezy/Dapper install that fails even in safe mode due to the newly introduced PCMCIA junk. If anyone is seriously considering using the GUI upgrade system, check that your motherboard works. If someone runs into this problem -- Boot using any live cd and just rm everything related to PCMCIA in etc so the system fails PCMCIA check, but at least does not hard lock. From here you can dist-upgrade the remaining packages from Dapper and everything should work fine. Once the new kernel has been installed, PCMCIA detection isn't a problem...
Other than that, it seems speedier than Breezy and I'm loving the fact that SMP kernel is now integrated into the main -- saves me a reboot or three.
Unfortunately, the universe would collapse if these were introduced. Millions of people suffering alcohol poisoning daily... And worst of all... the nickname sliders would have to go.
Student Perspective->
I go to a school where the laptop program is required to enroll. At first glance, the extra cost sucks.. Each quarter involves paying a fee of $360 for the use of the laptop and the installed software. The program works so that after the first two years, you return the laptop and receive a new one which you will be given for *free* after you graduate, if you graduate. All in all, you end up paying:
$360 * 3 trimesters * 4 years of school = $4320
Given the fact(s) that:
1) classes don't have to obtain lab time
2) laptops all contain standard software and no one has to hunt around installing new things and wasting time in class
3) professors can *expect* the software is there and plan lessons around it
4) the laptops are not locked down in any way
5) complete accidental damage coverage is provided (including beer soaked laptops)
The laptop program is actually a good deal. Plus, upon graduation you can quickly recover at least half of the costs of the laptop.
Administrator Perspective ->
At one point I worked for the school building the images and cloning to multiple laptops. The way we separate the images is by clumping the majors that are most likely to use specific software together, and sort out liscencing after. It is an engineering university and the laptops that are provided are usually top of the line models that cost around the price point of $4500, with a return from HP of $500 for any laptops failing to be associated with a graduating student.
The liscencing we do here involves thousands of dollars of programs installed on each laptop, again, sorted by major. The Architectural / Mechanical engineers may need SolidWorks, so it is installed. The liscencing is taken care of through liscence servers. All the majors are provided MatLab and general electric circuit analysis tools.
Administering 1000 laptops vs. 300 desktops will definitely be a larger task. Without a good place for the students to go to get their laptops repaired, no questions asked, the program will fail. Expenses aimed at making the student pay for every repair or damage to the laptop make them angry and unwilling to participate. The only thing my school does not cover is a stolen laptop, but a deductable of $800 still shields the student for the full cost (As long as there is a police report).
Overall, if the latops are good enough desktop replacements, the software is available and not costly to the student, and there is a service center on campus providing decent customer support, then the desktopless environment for school is an excellent choice no matter what the concentration.
Anyone know what sort of performance hit there is running the current Firefox release under Rosetta? I mean, do the Flash ads stutter or anything?
Firefox under Rosetta performs... poorly at best. From my initial tests it bounces short of a dozen times in the dock, finally loads, and then crashes within the first five minutes of casual browsing. Whether or not this is due to the missing Java and Flash plugins, I'm not sure. Overall, it left me removing Firefox shortly after I installed it. March can't come soon enough. Another interesting point is the non mozilla community builds compiled for intel mac. While these run natively and are quite fast, there are still some bugs to be worked out with random crashes. Again, I look forward to the March release of an official Firefox build.
Not only that.. But you can use the VMware image, a live cd of any linux flavour, and dd to create a native booting version of OSX. Unfortunately.. it seems the Broadcom GiB-Extreme support is missing. After playing with adding Darwin kexts to the extension library manually.. I gave up.
It looks pretty. It plays well.
It is useless without a network card of any kind.
Slashdot Mirror
I think you have microsoft confused with apple.
Seriously can we put this statement to bed yet? It has been several years (think, five or so) since this statement has even been slightly accurate. Yes, many writes can destroy a drive, but the number is in the (upper) hundreds of millions - performed on one single sector.
Today flash hard drives levy on technology used in older embedded devices that relied on flash, called "wear leveling".
Because each write is spread out throughout the entire disk, you don't physically write to the same sector X thousands of times when updating a cache file or whatnot.
Even if you had something thrashing the SSD continuously, you would not destroy the drive within the reasonable lifespan of a comparable rotating media drive.
Is this more a problem of UBUNTU or a problem of XORG? Don't knock something that isn't UBUNTU's fault.
Why your post is modded insightful, I have no idea.
>> First, if it was really OSX, why would they need Google's help to implement Google Maps? It would just run.
Why did any company need Apple's help and vice versa with the transition to a new processor? If it's a different architecture, then there is a good chance some porting techniques will need to be applied and Apple and Google got together to do just that.
>>Second, the interface is obviously significantly different.
Define significantly different. What would OSX look like if you replaced Finder with X application managing the phone capabilities? No one knows. You can't really say at this point it is significantly different.
>>Third, it's hard to believe a handheld would have the resources to run OSX.
Why? Have you ever run OSX on a G3 iBook ~ 500Mhz? Some of the fancy things are gone, but it's still gorgeous to look at and quite usable. In four years time it's not hard to imagine that a dedicated graphics accelerator in conjunction with the (whatever) processor could easily outdo what an old iBook could do, at 640x480 resolution nonetheless.
>>Finally, if it was really OSX, then any OSX app would run on it (in theory). Why can't it? Checkbox Intel, PPC, and perhaps ARM architecture in Xcode and compile your application. Surely there are going to be some gotchas, but after the first transition and everyone using Xcode to write UB apps, if the code was designed for portability, it should just work.
Just my two cents.
Thanks, way to include a text version of the blasting music on your myspace page.
But in all seriousness, this is going to bring the "web-rip" scene of trash on P2P networks to a whole new level.
Well I love Ubuntu, but the upgrade was *not so painless* for me and a few other users according to the forums. The upgrade process seems to fail on specific motherboards upon detecting that PCMCIA does not exist. If you are upgrading from Breezy, the kernel has no support to restart the detection or for this instance, fail out. Upon restarting the system, you are left with a hodge-podge Breezy/Dapper install that fails even in safe mode due to the newly introduced PCMCIA junk. If anyone is seriously considering using the GUI upgrade system, check that your motherboard works. If someone runs into this problem -- Boot using any live cd and just rm everything related to PCMCIA in etc so the system fails PCMCIA check, but at least does not hard lock. From here you can dist-upgrade the remaining packages from Dapper and everything should work fine. Once the new kernel has been installed, PCMCIA detection isn't a problem... Other than that, it seems speedier than Breezy and I'm loving the fact that SMP kernel is now integrated into the main -- saves me a reboot or three.
Unfortunately, the universe would collapse if these were introduced. Millions of people suffering alcohol poisoning daily... And worst of all... the nickname sliders would have to go.
I feel as if I was just verbally assaulted by an informercial.
Student Perspective-> I go to a school where the laptop program is required to enroll. At first glance, the extra cost sucks.. Each quarter involves paying a fee of $360 for the use of the laptop and the installed software. The program works so that after the first two years, you return the laptop and receive a new one which you will be given for *free* after you graduate, if you graduate. All in all, you end up paying: $360 * 3 trimesters * 4 years of school = $4320 Given the fact(s) that: 1) classes don't have to obtain lab time 2) laptops all contain standard software and no one has to hunt around installing new things and wasting time in class 3) professors can *expect* the software is there and plan lessons around it 4) the laptops are not locked down in any way 5) complete accidental damage coverage is provided (including beer soaked laptops) The laptop program is actually a good deal. Plus, upon graduation you can quickly recover at least half of the costs of the laptop. Administrator Perspective -> At one point I worked for the school building the images and cloning to multiple laptops. The way we separate the images is by clumping the majors that are most likely to use specific software together, and sort out liscencing after. It is an engineering university and the laptops that are provided are usually top of the line models that cost around the price point of $4500, with a return from HP of $500 for any laptops failing to be associated with a graduating student. The liscencing we do here involves thousands of dollars of programs installed on each laptop, again, sorted by major. The Architectural / Mechanical engineers may need SolidWorks, so it is installed. The liscencing is taken care of through liscence servers. All the majors are provided MatLab and general electric circuit analysis tools. Administering 1000 laptops vs. 300 desktops will definitely be a larger task. Without a good place for the students to go to get their laptops repaired, no questions asked, the program will fail. Expenses aimed at making the student pay for every repair or damage to the laptop make them angry and unwilling to participate. The only thing my school does not cover is a stolen laptop, but a deductable of $800 still shields the student for the full cost (As long as there is a police report). Overall, if the latops are good enough desktop replacements, the software is available and not costly to the student, and there is a service center on campus providing decent customer support, then the desktopless environment for school is an excellent choice no matter what the concentration.
Anyone know what sort of performance hit there is running the current Firefox release under Rosetta? I mean, do the Flash ads stutter or anything?
... poorly at best. From my initial tests it bounces short of a dozen times in the dock, finally loads, and then crashes within the first five minutes of casual browsing. Whether or not this is due to the missing Java and Flash plugins, I'm not sure. Overall, it left me removing Firefox shortly after I installed it. March can't come soon enough. Another interesting point is the non mozilla community builds compiled for intel mac. While these run natively and are quite fast, there are still some bugs to be worked out with random crashes. Again, I look forward to the March release of an official Firefox build.
Firefox under Rosetta performs
Not only that.. But you can use the VMware image, a live cd of any linux flavour, and dd to create a native booting version of OSX. Unfortunately.. it seems the Broadcom GiB-Extreme support is missing. After playing with adding Darwin kexts to the extension library manually.. I gave up. It looks pretty. It plays well. It is useless without a network card of any kind.