Re: disk drake

[Szakacsits Szabolcs]

On Tue, 13 May 2003, Andrew Clausen wrote:
> On Mon, May 12, 2003 at 09:13:53PM +0400, Yury Umanets wrote:
> > I guess, that ntfs in parted is intersting by itself, but, it makes
> > small sence because of the following reasons:
> >
> > (1) Ntfs is needed only if you are using ntfs on one of yours
> > partitions, that is, you have windows on it.

Potentially over 200 million users and it's growing fast. Most don't switch
from Windows to Linux completely but dual boot. Also a significant part
needs both for their daily job, etc.

> > (2) I'm sceptical about that, anybody will use ntfstools (resize, check,
> > etc) under linux if windows present on the box.

Considering most commercial implementations and Windows technical
limitations it makes a lot of sense :)

> Many Windows users would like to use GNU/Linux also.  (Conversely,
> many GNU/Linux users would like to use Windows occassionally)
> I think it is very important, just I'm short on time :(

I also think Andrew is right. But here are some numbers. Visitors to the
ntfsresize FAQ page since December is over 100,000. Downloads of the
statically linked ntfsresize is over 10,000. Some people reported over 100
box having NTFS were resized by it. Of course I don't have any clue how
much people used Mandrake to resize NTFS and used other sources to get
ntfsprogs.

Last month Linux-NTFS project had over 200,000 page hits. But SourceForge
numbers are unreliable, they lose data. It's growing almost like
exponentially. You might want to take a look at the graph below how fast
the interest grew in the last two years since XP started shipping. Red line
is the interest (decline is due to the incomplete May data)

  http://sourceforge.net/project/stats/index.php?report=months&group_id=13956

> They'll support where there is money.

Think so. But having more potential customers wouldn't result more money?

> Historically, there hasn't been much money in supporting general purpose
> desktop environments in GNU/Linux.  Hence, NTFS and Windows compatibility
> hasn't been a high priority.

For migration, interoperability was always one of strengths of Linux but
today due to the incomplete NTFS support it looses day by day, IMHO.

> (It's too hard and too expensive to enter & compete in that game).

Well, I estimate doing

  - data relocation for ntfsresize is about 1-2 weeks (plus QA)
  - integration with parted 1-3 days (hack), properly 1-2 weeks
  - adding write support to the NTFS driver 1-3 months

Full time. But nobody is doing that. True it's not easy but would it be
really too expensive? I'm quite curious instead of doing the open source
way, how much Xandros pays/paid for PQDisk and SuSE for PartitionExpert.
Wouldn't that cost cover the open source development? Once. Of course
competitiors would also freely benefit ... unless they share the costs.

There are things to (re)learn.

        Szaka