Re: [HIG] Recent files

[Gregory Merchan <merchan phys lsu edu>]

On Sat, Aug 10, 2002 at 12:13:42PM -0500, Seth Nickell wrote:
> On Sat, 2002-08-10 at 01:38, Gregory Merchan wrote:
> > On Fri, Aug 09, 2002 at 05:57:28PM -0500, Seth Nickell wrote:
> > > <snip>
> > > So basically, I was already aware of your card sort results, but I don't
> > > think they were the right tool for answering the questions we need
> > > answered.
> > 
> > Which is why I did more than just ask people to sort.
> 
> You asked them questions. *buzzer* Sorry, play again. As Nielsen,
> Norman, and Cooper say at different points, you cannot effectively
> understand a problem by asking users questions. You need to watch what
> they're doing, and *THEN* ask them the questions. The questions you
> cited before were asking people to list things they didn't like about MS
> Word, which is fine, but not necessarily a useful tool for directly
> making a design decision.

I had one person tell me things she didn't like about MS Office.
I did not ask for such a list. You are either confused or a liar.


<snip>
> Huh? Recent files experienced a lack of of use ONLY IN GEDIT WHICH WAS
> FOLLOWING YOUR RECOMMENDATION. Dude, people used Recent Files like crazy
> in Microsoft Word, and Excel. At the time I didn't notice people weren't
> using the feature in GEdit because its not what I was looking into that
> day.

Different people may use different programs in different ways. There are
help sites on the 'net showing how to remove the list, make it longer, how
to move it to an Open Recent menu, and how to make a similar top-level menu 
for some MS products.

Even a person who uses it in MS Word might not use it in GEdit. The programs
are not tuned to the same kind of work. Adobe Photoshop uses Open Recent
according to some sites I saw and so does Gimp 1.3.

If one rarely uses a program he might not consider the list to be of any
consequence.

You keep saying that you didn't notice and you weren't paying attention.


<snip>
> I don't have a big enough sample to pull up real reliable numbers like
> "90%" and "10%", so those are bullshit. But that's the effect I both
> theoretically expect to see, and the effect actually observed in a small
> usability test.

You said that you didn't notice and weren't paying attention.
It is not surprising that you now recall what you would expect.


<snip>
> I'm not caring too much what Matthew would actually comment on this
> issue, so I'm not caring what context he put this argument in. What I
> care about is that this argument has no outside dependencies. It doesn't
> *need* context to be valid. If a seperate submenu is used, users will
> habitually go into the open recent submenu *just in case* the file they
> want is there, before exiting and moving back up to the 'Open...' item.
> I want to avoid this problem. I think its more important than listing a
> lot of items, because what I'm seeing is even worse than what Matthew
> predicted. I'm seeing people not *using* recent files at all.

You're arguing for a context broader than Matthew's. That is not the same
as summarizing or reordering his argument.

You still need a context even if it's as broad as "stuff in the UI".

You have gone from not even noticing something to seeing it. Fascinating.


> > > but that is not a can of worms I'm willing to open again in this release
> > > cycle (which should be over in less than a week and a half).
> > 
> > Lovely. Everyone should be aware that if you want a change you'll make 
> > it yourself near release time rather than discussing it and possibly
> > being shot down.
> 
> Right, that's what happened. I've been itching to make this change for
> months and I'm waiting until near release to avoid being shot down.
> (FWIW, what happened is that I've been avoiding working on the Menus
> section because picking up whole sections isn't so much fun, but I
> finally realized it needs to be finished before HIG release (it has the
> highest FIXME density right now)).

You have a convenient conspiracy of realization. I have often wondered
whether people who act as you do realize that they do so. It seems they
rarely do.

Recent files wasn't a FIXME, yet you jumped at it. There's no bug and no
visible complaint about it. The menu section may have the most items
marked FIXME, but there is no dialog section at all. There is nothing
about regular dialogs in the HIG.


> > I included that information because I'm arguing for what has shown itself
> > to be better and not for my own personal preference. 
> 
> Very weak argumentation technique. "I'm not arguing for my own opinion,
> I'm just arguing for what is known to be right"

You seemed to think that I was arguing for the submenu to discourage it;
I was clarifying. Indeed, such a technique would be weaker than just lying.


> > My own preference is to have reliable type recognition and handling, to
> > be able to easily create any kind of object (file, folder, launcher, etc.),
> > to have folder windows of reasonable size and shape to organize my work,
> > to be able to make folders based on queries, to be able to take a folder
> > of some kind of work and turn it into a template for later work of the 
> > same kind. If you're thinking people don't work that way, note that they
> > can't.
> 
> I also like this idea(s), at least some variant of it. We were pushing
> to make things like folders based on queries for Nautilus (using Medusa)
> but didn't have time for 1.0. I'm slowly working on pushing a generic
> templating mechanism into GNOME. Etc.

Good luck.


> > Keep reversing changes hashed out a year or more ago. Keep supporting them
> > with illogic and appeasement. I'll try not to give a damn; at least I don't
> > have to use it.
> 
> Yup, that's me. Glad to see you're still melodramatic and petty.
> It's just like old times.

I suppose that by calling me melodramatic you are accusing me of being
insincere in my anger. I don't know what bearing that would has on any
of this even if you could prove it.

Petty I'll take either as a general insult or as meaning "concerned with
petty things", which is a common use. For the latter I will have to agree.
After all, I was concerned about you.


As for the matter of where the list goes, I don't care. Now that something
more than "I think this is totally broken" has been said, I find the data
inconclusive. Even with something conclusive, the procedure is wrong as
it was with the label alignment change. But I had no significant investment
in that and was presented with just the change instead of "totally broken".
If just the comment had been "further studies favored not using a submenu"
or something like that, this thread probably would have been avoided.


Cheers,
Greg