Page 3 of 4

Posted: Mon Feb 07, 2005 8:14 am
by uniQ
azeemarif wrote:if you push it to the limits even Windows XP has bugs in its ".ext" implemetation (most people don't come across those cases though).
How so?

-Q

PS. If there is an option to enable/disable this, there probably should be an option "Do not decide type is executable file" for safety and confusion.

Posted: Mon Feb 07, 2005 2:17 pm
by Phalanx
toxigenicpoem wrote:
but this is working against the hardware.
Do the research, over all hard-drive failures are under 1% due to mechanical malfunction. Segate reports that 40% of HardDrives returned, are not do to mechanical failure. The most critical point of a harddrives life is during the first inital spin cycle at boot up. Please use google, thats why they made it.... :wink:
Also reading google you will get the world is still square, pigs fly and kangaroos live in Antarctica. The simple fact is that every time a HD fails due to being worn out, you normally will have just passed your warranty and will not be able to return it. Did you notice your facts were returned hard drives? And if 40% are not due to mechanical failure, it still mean there is 60%.
toxigenicpoem wrote:The most critical point of a harddrives life is during the first inital spin cycle at boot up.
Sure, but the wearing out is still while in use. It is just when you start it for the first spins that it fails cause that is when the strain breaks it (and does not mean the damage is done when it starts or is off).

Posted: Mon Feb 07, 2005 3:21 pm
by Toxigenicpoem
Did you notice your facts were returned hard drives? And if 40% are not due to mechanical failure, it still mean there is 60%.
yes I did, but please remember thats 60% actual mechanical failure out of 1% noted as return on mechanical failure.... or 0.6% returns of failed Harddrives all together due to mechanical malfunction.
just passed your warranty and will not be able to return it.
That is a good point.
Also reading google you will get the world is still square, pigs fly and kangaroos live in Antarctica.
:wink: ha, also a valid point. :) But I'm talking about checking out statistical data, LOL...


It is just when you start it for the first spins that it fails cause that is when the strain breaks it (and does not mean the damage is done when it starts or is off).
I've re-read this 20 times, and I still dont have a clue as to what the hell your saying here friend. When the HDD starts to spin is when it gets the MOST stress, and the MOST strain. and there for most of the damage is done during startup. [/quote]

Posted: Wed Feb 09, 2005 6:37 pm
by MadRat
Static files generally have static checksums, right?, so instead of cross checking dates the checksum makes more sense. Isn't this sort of what MS initially planned for their index.dat files they generate with XP's indexing service?

This idea really reminds me more of VAX idea than what I'd think of with Windows.

Posted: Wed Feb 09, 2005 11:27 pm
by mf
MadRat wrote:Static files generally have static checksums, right?, so instead of cross checking dates the checksum makes more sense. Isn't this sort of what MS initially planned for their index.dat files they generate with XP's indexing service?
And checking checksums for 3000 files in a directory takes quite a few millisecs (understatement intended) more than simply matching modified dates.

Posted: Wed Feb 09, 2005 11:41 pm
by NetSlayer
mf wrote:And checking checksums for 3000 files in a directory takes quite a few millisecs (understatement intended) more than simply matching modified dates.
Thing is, creating checksums (or hashes) from files is quite slow, because the algorhythm has to go trough the whole file. If you have some big files in your 3000-file-directory you really have to wait some minutes. ;)

Posted: Thu Feb 10, 2005 6:28 am
by MadRat
mf wrote:
MadRat wrote:Static files generally have static checksums, right?, so instead of cross checking dates the checksum makes more sense. Isn't this sort of what MS initially planned for their index.dat files they generate with XP's indexing service?
And checking checksums for 3000 files in a directory takes quite a few millisecs (understatement intended) more than simply matching modified dates.
Well, a SID for the file is basically a unique value or as near-unique as you'll get and can play the role of the checksum. The file system should handle these tasks without need to map out the entire drive contents everytime their is a simple directory call of a folder. Explorer currently refreshes at a regular pace generally of every few fractions-of-a-second or longer depending on the latency of the task. Only when reading remote system drives, or when the system is under full load, is the latency all that bad.

my 2 cents

Posted: Thu Feb 10, 2005 7:56 am
by Floyd
I personally hate linux not using file extensions. The OS may not need them but I find that it makes it easier for me to find stuff (I often use explorer, KDE Kommander or whatever).

nothing I hate more on linux than messing with a new program and trying to figure which file is the executable. extensions don't matter to the OS, they matter to the users. Besides, if you absolutely have to have to be more "linux-like" just hide the file extensions.

Posted: Thu Feb 10, 2005 3:12 pm
by Mammlouk
Th idea isn't to start purposely leaving out extensions, its to be able to handle the few stray files that don't have them. More particularly for someday when there is a fully functional linux subsystem.

Posted: Thu Feb 10, 2005 6:31 pm
by Phalanx
mammlouk wrote:Th idea isn't to start purposely leaving out extensions, its to be able to handle the few stray files that don't have them. More particularly for someday when there is a fully functional linux subsystem.
Problem is when you start supporting it, you give people a way to be even more lazy and make the problem even bigger. IE is a perfect example of this.

Posted: Thu Feb 10, 2005 6:34 pm
by Phalanx
toxigenicpoem wrote:
It is just when you start it for the first spins that it fails cause that is when the strain breaks it (and does not mean the damage is done when it starts or is off).
I've re-read this 20 times, and I still dont have a clue as to what the hell your saying here friend. When the HDD starts to spin is when it gets the MOST stress, and the MOST strain. and there for most of the damage is done during startup.
It is like the commit aircraft. The strain on the metal came throughout the whole flight. But it is at the point it is highest were the aircraft ripped open in flight. This does not mean the total damage done to the window happened during take off, it just means that is when the breaking point occurred. More damage would have happened during the rest of the flight, just being not as harsh, it does not reach the breaking point of the metal as fast as the take off strain did.

Posted: Fri Feb 11, 2005 7:47 pm
by Toxigenicpoem
It is like the commit aircraft. The strain on the metal came throughout the whole flight. But it is at the point it is highest were the aircraft ripped open in flight. This does not mean the total damage done to the window happened during take off, it just means that is when the breaking point occurred. More damage would have happened during the rest of the flight, just being not as harsh, it does not reach the breaking point of the metal as fast as the take off strain did.
I really dont see this a valid analogy, there are way more forces impending on a plane in the air, vs take off.

Posted: Fri Feb 11, 2005 10:13 pm
by Phalanx
It had to do with the change of pressure as the comet was one of the first pressurised aircraft.

Posted: Tue Feb 22, 2005 6:55 pm
by Toxigenicpoem
11 days later, and I still fail to see your point. Sorry friend I tried.

Posted: Tue Feb 22, 2005 9:48 pm
by GreatLord
I do not if this a good idea in reactos or not.

it scan frist three leters if it find sign MZ something it is a exefile
most file format have a signture at the begin of the file, to determent which file type it is. elf as elf at the begin of the file.

I think it good idea to keep dot and three lettter and implement this idea also. but I have not talk with any other dev about this idea.