Re: Soft ejects
Quoting Fish Smith (dyson_sphere_explorer@yahoo.com):
> >It is not only newbies that can make stupid mistakes,
> >and remove a
> >floppy disk that is currently mounted...
>
> I was taught in kindergarten /never/ to remove a disk
> when the light was on, and I never do it. Removing
> while it is mounted but not currently being read or
> written isn't very damaging--you just get an error
> message, have to unmount and remount.
While agreeing about the light, removing a disk when the light is
off is generally less damaging if you're (a) not in a hurry, i.e.
you leave it a while if writing, and (b) it's a FAT disk rather
than ext2. This is one reason why I only use FAT floppies. (The
other is I'm worried about people thinking ext2 floppies are
corrupted and throwing them away when they can't read them.)
> >Perhaps the real problem with soft ejects is that
> current
> >implementations make it to easy to override, eg when
> the power is off.
> >Personally, I think I would much prefer the risk of
> not being able to
> >eject a disk, rather then the risk that someday I
> will accidently
> >currupt an important disk by ejecting it when it is
> still mounted.
>
> This is more or less the same as saying "personally I
> would prefer not to be able to delete a file, rather
> than the risk that I someday will accidentally delete
> something important." Should your OS not allow you to
> delete files manually? It is generally good policy to
> ask "are you sure you want to delete this file" but
> there is always going to be a chance of deleting files
> you need, no matter how many precautions are added,
> short of simply not allowing user deletion of files.
I don't agree with the comparison. If I decide to delete a
file, I make that decision using the evidence of the moment.
I don't need to have retained extra invisible state information
in my brain.
If I remove a floppy disk with the button, I need to
remember whether I wrote to it recently. I would also
need to remember if it was ext2 (were that possible in
view of my earlier decision not to use them).
I am willing to use ext2 jaz and zip disks because the
computer does the remembering for me. That's what computers
are good at.
> >These protection devices not need to turn you into a
> windows[1] user, I
> >think it is just plain common sense. Other protection
> mechanims
> >already exist in Linux, eg you can't eject a CDROM
> that is mounted (I
> >guess this protects programs from crashing that are
> currently using
> >it), you can't e2fsck a mounted filesystem, etc.
>
> But if you had a hard eject button, you could eject
> the CDROM while mounted, (even if the OS didn't like
> it) something I have needed to do plenty of times but
> have been unable.
I thought that's what the pinhole was for.
> >Note:
>
> >[1] Dos/windows copes with this problem in a
> different (IMHO broken)
> >way - it keeps track of which disk is inserted, and
> if it needs to
> >read/write to another disk, it complains to the user
> to reinsert the
> >original disk. Why is this mechanism broken? For
> starters: some games
> >will automatically eject a CD-ROM and ask you to
> insert the next
> >CD-ROM. For some reason, windows will often decide
> that it still
> >needed the original CD-ROM, and ask you to reinsert
> it!!! It even goes
> >as far as to suggest that the CD-ROM might be dirty.
> Now thats what I
> >call "machine is smarter"!!!
>
> Everything windows does is (IMHO) broken. That's my
> point. However, DOS doesn't have a problem with
> taking media out whenever, as long as it doesn't have
> the light on. I've never had DOS ask for a different
> disk (except for individual applications which ask for
> the disk they need, not the last one in.)
That's not my experience. I have hardcopy of the evidence
in front of me for two particular occasions (there have
been plenty more). With the then current disk in the drive,
DIR A: gave a correct listing of the disk. On changing
disks, DIR A: complained thus:
Invalid disk change reading drive A:
Please insert volume { serial 0070-06F4
Abort, Retry, Fail?
The "{" differed on each occasion - another said "~Z&tp"
but the serial number was always 0070-06F4, which, to the
best of my knowledge (and all my floppies were catalogued
under their VSN if they had one, label if not) I did not
and never had possessed.
On replacing the then current disk, DIR A: would work again.
Rebooting would make the problem go away.
Just thought I'd share that with you.
> The bottom line I'm getting at here is the idea that
> these machines are here for our effective use. They
> make plenty of mistakes, always have and always will.
> So do we, of course, but /we/ are the ones paying for
> our /own/ mistakes, whereas if we give the machine
> power over the decisions, presuming it is infallible,
> /we/ pay for /its/ mistakes. If I'm paying for a
> mistake, it damn well better be a mistake I made. One
> thing I've learned in life, don't put yourself in a
> position of depending on someone else unless you're
> sure they'll come through. By that token, I also hate
> others depending on me because if I don't come
> through, somebody else is paying. I should be the
> only one paying for my mistakes, and only for mine.
I personally think you're mistaking power of control
with the convenience of allowing machines to remember
things for us, but providing manual override where they
get it wrong.
By the way, none of your emails seem to contain any
header information about who/what you're replying to.
Would it be possible for you to add a line like the
Quoting Fish Smith (dyson_sphere_explorer@yahoo.com):
above so that we know who you're replying to. (Perhaps
your email client could be set up to automate this.)
Cheers,
--
Email: d.wright@open.ac.uk Tel: +44 1908 653 739 Fax: +44 1908 655 151
Snail: David Wright, Earth Science Dept., Milton Keynes, England, MK7 6AA
Disclaimer: These addresses are only for reaching me, and do not signify
official stationery. Views expressed here are either my own or plagiarised.
Reply to: