Re: Baobab

[Emmanuele Bassi <ebassi gmail com>]

On Thu, 2006-07-27 at 01:39 -0700, Jeff Waugh wrote:
> <quote who="Emmanuele Bassi">
> 
> > Neither saying that having a tool that shows how much of your disk your
> > files are taking up belongs to a "power tools" suite makes a lot of sense,
> > given that on Linux you have had the same tool installed as part of your
> > basic set of commands since 1995 or something like that.
> 
> Considering Baobab's approach,

... Which is showing a list after having selected a starting point, so I
don't see what's wrong with that approach and why it's so strikingly
"power-tool-esque"...

>  and the problem it solves for users, I think
> it is entirely reasonable to suggest that it is a 'power tool'.

That's where I don't agree.

For me the system log viewer is a power tool.  My wife doesn't know what
logs are and why should they be important (and my wife is a geek who
wants me to explain her how the dictionary protocol works); but my wife
really would like to know the layout of her Documents folder by size, to
know if she has been copying stuff, or if it's time to remove the old
cruft.

> We generally think about delivering a complete, integrated user experience
> like f-spot rather than a set of random one-shot tools.

Then we should remove gnome-utils entirely, that's my point.
gnome-utils is composed of one-shot tools by definition (utilities).  As
I said, splitting gnome-utils into its applications and then submitting
for approval each and every one of them is a perfectly fine solution for
me.  Hell, I'll do the split up in the week end if the release team
gives the go ahead and the community wishes so.

> I have often used the phrase 'greatest commmon factor' to describe the kinds
> of things we ought to ship in the Desktop suite. It means shipping things
> that are going to be appropriate for most of the different kinds of users
> who use GNOME. How does Baobab fit in with that?

By providing a functionality not available on the desktop up until now,
and providing it integrated in the UI and in the interaction with other
tools (gnome-search-tool and nautils).

>  What is the user goal that
> we are trying to solve by including Baobab,

Showing the folder structure from a starting point, in a way that
conveys not only the hierarchical layout but also the size of every
single node in the structure.

>  and how can we solve those user
> goals more directly?

By having nautilus showing the same layout and having it ordering by the
same criteria the same data.

If someone wants to code a patch for nautilus 2.18 and have it accepted
by nautilus' maintainers, I'll more than gladly remove baobab from
gnome-utils.

>  What are our users trying to achieve?

They are trying to see how their files are big, and where their disk
space went, especially when a notification popup tells them that they
have 88% of their disk full.

> These are the questions we need to answer, and we need good answers to those
> (even if coming up with them is harder)

As you can see, those are easy questions, and they were posed and
answered before baobab was included, because I don't want to add stuff
"willy nilly".

Otherwise, a year ago or so, when Davyd proposed the inclusion of gruler
inside gnome-utils I would have said "yeah, great idea, let's do it".

> I'm not saying this to diss Baobab or your maintenance of gnome-utils, I
> just want to make sure that the things we've achieved over the last six
> years live on in the psyche of new maintainers.

As you can see, those things haven't bounced on my thick skull.  Those
things are why I love GNOME and contribute to it.

Ciao,
 Emmanuele.

-- 
Emmanuele Bassi,  E: ebassi gmail com
W: http://www.emmanuelebassi.net
B: http://log.emmanuelebassi.net