SPEC : Write Support For NTFS

[Florent Mertens]

Le vendredi 27 octobre 2006 à 11:15 -0700, Matt Zimmerman a écrit :
> The trouble with this, as I noted on the spec, is that the upstream authors
> of the relevant tools state quite clearly that they are neither ready nor
> intended for prime time.  We shouldn't take risks with our users' data with
> software which is specifically contraindicated for general use by its
> authors.
> 

I actually wrote this spec with the help of Szabolcs Szakacsits (the
author of ntfs-3g) so he is aware of it. Saying that ntfs-3g is not
ready for prime time is right and wrong :

- wrong because read/write support with ntfs-3g is quite stable now and
he, szaka, don't like to hear the word experimental for it driver :

http://lists.debian.org/debian-wnpp/2006/10/msg00323.html

- right, because this is still a beta release, because of the reasons
stated by the authors :

"> People are keep asking why ntfs-3g isn't marked stable yet, why it's
still
> in beta when for example the stable Captive NTFS crashes and corrupts NTFS
> by running this simple test:
>
>         for i in `seq 1 200` ; do touch $i ; done
>
> The answer is that, I'd still like to solve some issues described in the
> README file (e.g. transparent unicode handling, fix posix timestamps,
> running my "mission-critical" workstation Linux for at least a week on NTFS
> root (everything being on only NTFS), finish completely with the posix
> conformance, LTP and some other testsuite, etc). Some of the solutions
> aren't trivial, and the constant testing needs quite a lot of time as well.
> To be honest, it's a bit unbelievable how stable the driver performs during
> general use for so many people ..."

and for others reasons like the fact that there is no equivalent to chkdsk for
linux yet (which could be quite problematic).


But, in any way, i don't think that this specs is too much ambitious (but i might be wrong). 
The intention is not to make it default (which should be a big error), but
- to make it easily installable ( the 'easily' can be discuss)
- to make possible to mount external device with ntfs-3g, which should fix

https://launchpad.net/distros/ubuntu/+source/pmount/+bug/63271

And the implementation plan start with a condition :
"as soon as *a stable release is provide*"


The "not intended for prime time" is an other story. In fact it's quite clear now that
all the needed improvement made by szaka will not be merge in ntfsmount, because
of technical disagreements between him and the main author of ntfsmount, Yura Pakhuchiy.
This might change in the futur, but actually it's not his intention, and :
- there is an active developpement on ntfs-3g with on the other part, no or 
few merge in ntfsmount. the diff begin to be big.
- Szaka, state it clear to me, he will maintain ntfs-3g in the future, "unless somebody makes, 
maintains a more reliable and usable driver."

So to be honest, i think myself that this is a kind of decision that need time and
reflexion, and i don't think that it's ready "right now". But 6 month could be long. i think
you should wait and see how goes the stable release, that will probably come in this period.

This could be a big improvement for users.

Regards,

Florent Mertens

[It seams that this mail get lost somewhere the fisrt time i send it. I resend it to the list]