[Techtalk] writing to ntfs?

Caroline Johnston johnston at biochemistry.ucl.ac.uk
Fri Feb 6 16:35:37 EST 2004


Hi again,

Just to let you know I've not been put off, I'm happy to give captive a 
go. The only reason I need my windows partition is for internet access
and CM4 so it's nothing I can't replace if I break it. (all this faffing
about is just to rescue about 200 hours of CM4 save games ;). 

I thought it wasn't possible to read a linux partition from XP. Any 
recommendations for software that'll do it? Guess that might be less 
likely to break stuff.

Thanks again for the help.

Cxx

On Sat, 7 Feb 2004, Rasjid Wilcox wrote:

> On Friday 06 February 2004 22:10, Yaroslav Fedevych wrote:
> > Rasjid Wilcox wrote:
> > > There are two sites to look at:
> > >
> > > The Linux-NTFS Project (http://linux-ntfs.sourceforge.net/) and
> > >
> > > Captive (http://www.jankratochvil.net/project/captive/)
> >
> > I can only say: Linux can deal ~70% well with NTFS 4 partitions to
> > write, and will warn you that you should try and check it every time
> > you've finished with writing job. Not impressive. In order to gain read
> > support for NTFS, there was a hack, since docs that leaked are poor. For
> > writing to NTFS, the hacks were even dirtier, IIRC, since it was reverse
> > engineering (English: shooting). As NTFS 5 (used in 2k+) contains even
> > more "features" which are proprietary and not hacked still -- hence,
> > Linux won't allow you to write on that -- just too risky.
> >
> > The solution is that you should keep a FAT32 partition for shared data...
> 
> Yaroslav, unless you can confirm otherwise, I going to assume that your 
> comments are directed at the the Linux-NTFS project, which is the well known 
> one, and has always(?) stated that write support was experimental.
> 
> The Captive project is a *completely* different project, and states that is 
> *does* have working write support.  Please don't discourage please from 
> looking at both projects, and please don't imply that the Captive project is 
> being dishonest unless you have at least some evidence to back up your claim.
> 
> I make no representation either way, as I have used neither.  But I do think 
> that working NTFS write support is important to Linux.  Therefore since, as 
> far as I can tell, the Captive project has have very little exposure[1] I 
> think that it is important that people know about it, and test it out.  If it 
> really does work as well as claimed, then the good news should be spread, 
> damit.  A _prudent_ user might set up a test NTFS partition to write to and 
> read from, and test it out for a bit, just to be on the safe side.
> 
> On the other hand, if you do know that the Captive NTFS write support is 
> substantially buggier than they claim, then please confirm and give a few 
> details.
> 
> Thanks,
> 
> Rasjid.
> 
> [1]  I don't know why - politics perhaps, or perhaps the authors don't know 
> how to promote the project?
> 
> 




More information about the Techtalk mailing list