[CharlesS]

Dude, it rarely ever happens, and it's not like this is a crippling problem. It just means that DiskWarrior will have to replace the directory in the same way as every other disk utility does it. In the extremely improbable case that there's not enough contiguous free space to replace the directory when the disk isn't nearly full, you'll still be fine as long as the power doesn't go out during the 5 seconds it takes to replace the directory. You've probably got a better chance of getting struck by lightning.

[Hal Itosis]

Originally Posted by CharlesS 

Dude, it rarely ever happens, and it's not like this is a crippling problem. It just means that DiskWarrior will have to replace the directory in the same way as every other disk utility does it. In the extremely improbable case that there's not enough contiguous free space to replace the directory when the disk isn't nearly full, you'll still be fine as long as the power doesn't go out during the 5 seconds it takes to replace the directory. You've probably got a better chance of getting struck by lightning.

 
But you missed the point. It wasn't *just* about that DiskWarrior example.

Folks like to say that OSX does all the defragmenting we would ever need.
I merely pointed out that it *doesn't* defragment free space.

All I was saying was: that even at 75 to 80% of capacity (i.e., not necessarily even "nearly full"),
free space fragmentation can impact performance, and perhaps be a fairly undesirable condition
if some event causes mega-swapfile generation or log-file explosion. All I'm saying is it's not all
as rosy as some folks may think, and therefore occasional defragging (however one wants to do it)
is probably a healthy practice... not something to scoff at condescendingly.

If you want to criticise that viewpoint and recommend that users should just remain oblivious, fine.

So we disagree (I guess?).

Big whoop.
What else?

[analogika]

possibly helpful? maybe.

"healthy"? Probably not, since it's hardly worth the risk of losing your data if anything goes wrong during defragmentation, and free space fragmentation is only problematic in extremely rare circumstances.


At any rate, none of this has anything to do with the notion that keeping icons on the desktop is "BAD".

[Tomchu]

The notion that defragmenting is such a dangerous activity that your data is virtually on the brink of no return is quite ignorant.

Take a look at Diskeeper.

Actually, take a look at iDefrag and its "Verify Writes" feature. And then consider what it means to "move" a block of storage.

[Chuckit]

Consider that the risk from having fragmentation is effectively zero and the risk from defragmenting is nonzero.

[CharlesS]

Indeed. I have seen far more problems caused by defragging then I've ever seen solved by it.

[TETENAL]

Consider that visual clutter on the desktop has absolutely no correlation to file or free space clutter on your hard drive. None at all.

[Tomchu]

Chuckit: Indeed. And yet some people care about performance (and free space fragmentation sucks for write performance). That is, after all, one of the factors you weigh in when purchasing a computer, yes?

The risk from defragmenting is effectively zero as well -- and it has been for a number of years thanks to better technologies and deeper hooks into the OS. Deal with it. :-P

[peeb]

You are dealing on the margins of performance except for very rare cases - this has virtually no relevance to the vast majority of users.

[Tomchu]

Lollers. Okay.

[Hal Itosis]

Originally Posted by Tomchu 

The risk from defragmenting is effectively zero as well

 
idunno.

From looking at posts around the web for years now, one could argue that
running system updates is a dangerous activity. Let's just buy our Macs,
check our email... and then put them to sleep, ASAP.

The less disk I/O, the better.

--

Wasn't the "desktop items" discussion concluded anyway?
What more did someone want to add about that topic?