Re: [Usability] Save Icon



On Sat, 28 Jan 2006, Steven Garrity wrote:

> Date: Sat, 28 Jan 2006 10:25:47 -0400
> From: Steven Garrity <stevelist silverorange com>
> To: Sarah Berry <sberry grex cyberspace org>
> Cc: usability gnome org
> Subject: Re: [Usability] Save Icon
>
> Sarah Berry wrote:
> > Thank you for your responses to my other suggestions.  Another
> > thought that has occurred to me is that the "Save" icon in GEdit and
> > other applications may no longer be appropriate.

There are regular calls for Gnome to completely kill off the concept of
manually saving.  The idea would be to save automatically like how
programs do on handheld devices, or smaller programs like Tomboy or the
Sticky notes applets do.

The immenant demise of Save has been predicted for decades now (I think
Alan Kay of Smalltalk fame mentions it regularly) but it is much easier to
implement manual save.  Hardly any (any?) large applications have
implemented full autosave and it tends to be used more to make regular
backups in case of failure. If you are going to provide autosave it must
be completely reliable if users are going to trust it.  I really dont
think save is going away anytime soon.

> > As floppy disks are rapidly becoming archaic and are no longer
> > included on many systems, perhaps a new icon could be created?

I never would have guessed I'd still be using floppies in 2006 but
somtimes for smaller files it can be an adequate solution, and it is
sometimes handy to have a bootfloppy around.

> > Unfortunately, I cannot offer any constructive criticism in the form
> > of what a suitable replacement image might consist of.

Tbat is the tough part isn't it :)

Any new icon will inevitably cause problems for existing users.

> Sarah,
> We've been debating this issue over at the Tango Project [1]. For now,
> the best alternative we've come to is an arrow pointing into a hard-drive.

That isn't a bad idea but it isn't ideal either, it seems more like
Download to me than save and that is assuming I recognise the Harddrive
icon.

> Here is a summary of some of the issues we have been considering:
>
> 1) The floppy disk isn't common anymore. There are lots of younger
> people who will have grown up using a PC and have never seen a floppy disk.

For me any new icon is the worst of both worlds, you alienate those who
already know what they are doing and therefore make it very difficult for
new users to get help.  (On an application by application basis I would
try and prod them to see if they could implement various bits of auto save
functionality, and see if they could reduce the need for users to manually
save things but for some programs this simply will not be possible.)

I wouldn't say the existing floppy icon is perfect either.  I did hear one
user ask "What is this other computer icon for" as she failed to recognise
the stylized floppy disk icon for what it was supposed to be (I think a
more plainly drawn icon would have been okay, so this is more a commmen on
the drawing than the icon metaphor).  It is best to both show and tell, to
have both icons and text labels shown by default as Glyphs/ideographs are
notorious difficult to idenifty otherwise.

> 2) The floppy disk is a very common save icon - people are used to it
> (probably - I don't know of any research).

It has been in use for eons and use by Apple Microsoft and just about
everyone else.

> 6) "We shouldn't have to save at all." Probably/maybe.

Nice idea, not at all easy to do properly and reliably but I hope more
developers will try. (I could do with a nice mspaint equivlant with
autosave, I should check what GPE are doing maybe run one of their
applictions on my desktop, but I digress.)

> Doesn't mean we should try to find a better icon in the meantime.

Trying to find one is good, but switching anything will be difficult and I
seriously doubt it would really be worth changing and we would really want
strong evidence of it being signficantly* better before switching.  (*
difference should be substatantial to offset the penalty of confusing
existing users)

> These are just my collection of thoughts - I don't have any research
> behind any of this (maybe BetterDesktop.org could help?).

OpenUsability may also have discussed the issue.

Sincerely

Alan Horkan



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]