Re: [Usability] Re: [Epiphany] epiphany toolbar/bookmarks

[Marco Pesenti Gritti <mpeseng tin it>]

I sent a reply a few minutes ago from a web client, but it seem to have
been lost. Hopefully I'm not going to post two times a similar thing.

> Two of the above statements are both crucial and, IMHO, unsubstantiated, 
> namely:
> 
> 1) that the ephy-bookmarks-unhappy are just a "very low part of user base";
> 
> 2) that the new bookmarks system is an improvement for most users.
> 
> With regard to 1) I can't but wonder how is it possible to say that 
> since 99% of all browser users (on most platforms) are used to 
> hyerarchical bookmark menus. Perhaps it's not optimal, perhaps is under- 
> and misused, but this is what people expect to find in a web browser: 
> you can easily move a Window user from IE to Mozilla (and then perhaps 
> to Linux altogether), but I doubt that new and ex-Windows users would be 
> less than puzzled by a completely different bookmarking method *without* 
> having the chance to import their bookmarks and getting on thanks to a 
> compatible system.

I tried to substantiate this is in the first part of my mail, when I
explained because we decided to do something different.

The fact that most browsers have a hierchical bookmarks system doesnt
really imply everyone is using that feature.

In my limited experience it's exactly the opposite:

- Someone doesnt use bookmarks at all
- Someone doesnt organize bookmarks at all
- Few people create 1 level of folders, which is still not a hierarchy

(The same for file manager, everyone put stuff on the desktop or in the
documents folder)

I talked with several people that tries to watch users and no one can
claim to have seen a non technical IE user to use multiple levels of
folders. Surely this isnt very scientific, but we have not resources for
extensive user tests, and we need to base our rationales on something.

Additionally I think it's proved by usability studies that people doesnt
grok hierarchies well and that submenus are phisically hard to use.

> Wrt to 2), I really don't know, all the more so because I still haven't 
> played much with Epiphany. It could well be, though, that there are 
> people out there who won't ever like the new method: I, for one, like 
> the idea of a well organized hyerarchical bookmark menu/system, which is 
> why perhaps I haven't used Epiphany a little more. Judging from the 
> reactions on gnomedesktop there are many users who find the new system 
> unwieldy: at this moment, it is not possible to rule out the idea that a 
> large part of present and future users won't feel completely at ease 
> with the new system. In some cases you just can't decide in favor of the 
> One True Way because it simply doesn't exist: which explains why window 
> managers offer a "focus follows mouse" option without forcing it on 
> their users.

Please notice the IHMO I put there.

Again, I dont have resources to do user test, which could allow me to
remove the IHMO, but I try to build a rationale about things I design.
The whole mail I sent was about this rationale, I'd like to keep the
discussion about specific points.

Do you really think a "I for one like the idea ... there are flames on
gnomedesktop ..." approach is going to produce something positive ?
If we followed gnomedesktop flames we would be back at GNOME 1.0 and we
would be using Velocity as filemanager.

Summarizing the positive points again:

- We dont use deep hierarchies and submenus which are proved to be bad
for usability
- We designed it with user tasks in mind, instead of following personal
tastes, coolnes etc ...
- We didnt do a bad copy of other applications or operating systems,
like happen for most of linux applications. We tried to build our design
over a rationale, with the non technical user in mind.
- Using a window has the advantage to make browsing phisically easier,
to allow more powerful filtering and to allow editing in place (without
context menus on menus that are totally crack)
- Our metadata/database approach is line with what is planned for the
GNOME future
- We are pushing system integration of bookmarks by allowing them to be
opened directly from the panel and the desktop
- We know that there are some usability problems with the current
system, some of them due to the compatibility with the old. But we tried
to show how there is space for improvements and that we have some ideas
to solve these issues.

The similarities with apple design and the users that started using
bookmarks only with epiphany are a confirmation that there is something
positive in the direction we choosed.

It would be really good to move away from an ideological battle between
the new and the old.
There are some differences between a hierarchical system and a metadata
based one. They are not completely different though, the bookmarks
hierarchy contains metadata (title, url, the folder itself can be seen
ad metadata).
Evaluating the differences in relation to user tasks (which of them it
improves, which make slower etc...) is more likely to produce a better
interface then worrying about gnomedesktop flames :)

Marco