Victor,

>That's qutie interesting, and I guess the more I think about it, the more it makes sense.
>Question now is what can we do with the information? I mean obviously we all like to keep our apps running as quickly as possible, but at the same time I'm not ever going to recommend to a client that they quit defragging their hard-drives because if they do it'll slow down their VFP applications....they'd look at me like I was nuts and I'd start hearing more of the old 'fox is dead or should be' stuff again.
>I haven't used it in a few years, but I wonder if http://www.diskeeper.com/diskeeper/diskeeper.asp allows you to select only certain parts of your drive to defrag, or perhaps we're doomed to put our data on a seperate partition?
>And who's the bozo that came up with this scheme anyway? I mean MS provides us with a OS tool to speed up our hard-drives and help increase the life-span of our hardware, and at the same time the same tool from MS screws us on something like this? grrrrrr talk about shooting yerself in the foot...geeze!

Unfortunately, the link you provided didn't work for me. I'll try later on that.

I certainly agree that there isn't much that we can do about fragmentation in the general case, given the tools (that I know of) that are available to us at this time.

But I don't fault MS on the tool provided, because I do think that for the majority of PC usage it will make a difference. Perhaps, though, there is too much emphasis on keeping a HD defragged. I feel that very intermittent defragging (once a year?) would be fine for most workstations unless the HD is known to be very full already. Then I might defrag more often as I wait for a newer bigger HD to be installed.

But I do feel very strongly that someone needs to make some kind of tool available to give us some significant control over fragmentation, even to the point of deliberately fragmenting specific groups of tables/files and even to control 'dispersal' of fragments within the group(s)!
I would also like Explorer to have more information about file fragmentation available, even if only displayable through a rt-click popup option or additional pages of the "properties" dialog. I would like that to include things like # of fragments for a file, HD "addresses" of those fragments (cylinder/head/sectors) and to have that information also displayable when more than 1 file is selected.

I would also like to see some options available to us as regards the allocation of free clusters on any NTFS volume. For instance, it would be helpful to be able to choose from a variety of free space allocation options for the volume as a whole. I'd like to see an option that specifies categorically that all allocations are to always proceed by looking first at largest contiguous chunk of free space and always allocating the requested space from there, and always from its lowest available cylinder/track/sector. Should that fail, then the next option should be to fragment the file in question but looking for appropriate free clusters FROM the largest contiguous free space (LCFS) area INWARDS towards the hub of the drive (it sure seems to go in the other direction today). If that fails to provide all of the needed space then it should outward from the LCFS to the outer edge of the spindle.

In the meantime I do think that the recommendations I made should be practical in the general sense. I will consider writing something up to help to "prove" the case that fragmentation of VFP tables/files is GOOD, not by using numbers but by using the layout of a HD and how files are allocated and records are written.

I think it would also make a HUGE difference if more and more people started TELLING people (anyone, but most especially MS and software houses that write related HD utilities) that fragmentation is very helpful for some applications - especially fast response situations! The axiom that fragmentation is universally bad needs to be challenged at every opportunity. It is that, more than anything else in my opinion, that holds software designers back from doing things WITH fragmentation rather than against fragmentation.

cheers