Richard Reynolds wrote: >.. >> For a very long time, ntfs was _maybe_ readable from Linux. > I cant consider it _maybe_, for a long time its been readable reliably. > >> Then, as I understand, it got somewhat better to the extent that ntfs >> filesystems could be created, which made it possible to move the things >> from one partition to another. I'm not sure whether defragmenting was >> required, or perhaps just recommended. I suspect the latter. > I would have to knock my head into the wall a few more times today to > recall the order of things, but at first defrag and checkdisk were > requirements and to some extent they still are > > >> Now for a few months I think there has been some claims of further >> breakthroughs, namely achieving writability. > its been a lot longer than a few months >.. >> "As of late 2006, Captive NTFS is no longer the only Free read/write >> NTFS solution for Linux. ntfs-3g is a FUSE based driver with almost >> complete NTFS write support which unlike Captive NTFS does not require >> any third party proprietary software to function. > the proprietary software is simply the windows drivers, its mode is to > use them in a way like vmware. your not magically granted license to use > them, you are supposed to have a copy of WinHozed version something that > includes ntfs support. but you can download them for free from M$ or > mount the ntfs partition ro and copy them to a location off the > partition to remount rw with those drivers in place. allowing them to do > all the work, this means files get writen and read just as if windows > was doing the work with the wishful thinking that everything works as > expected, which as I have been using it for years can say yea it works > most of the time, when it doesnt, usually it just forgets to check the > "clean unmount" flag so when booting into windows you get the checkdisk > warning. i have not seen a noticeable difference in fragmentation vs > windows which one should expect with the code is being re-used. other > problems exist, most are just general anoyances.overall this is the best > method I have found.
Thanks for de-fuzzying my story. You didn't seem to cast a vote for ntfs-3g vs captive-ntfs. If one uses reverse-engineered code and the other uses actual Windows code, should one trust the Windows one more? Regards, ..jim -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
