On Wed, 2013-04-17 at 13:10 -0400, Jesse Hutton wrote:
Lets consider a concrete example. Before Gnome Shell was initially released, I (like many others) didn't like the lack of a power off option in the system menu (or anywhere on the desktop). I've been an on and off lurker on IRC for a while. I brought up the concern a few times perhaps. At one point, I got into a small debate with owen about the design/user experience trade-offs of the issue. He made multiple specific arguments *against* having it in the menu and for having suspend (which I found completely unconvincing). I made multiple arguments *for* including it in the menu. It ended with him saying he'd wasted enough time debating the issue.
This example is not an usability study, it was a debate between 2 people having different opinions. I could also make the case in opposite direction, debates that proven to be right with the time, in both 2.x and 3.x cycles. The most famous that comes to my mind is workspaces versus viewports. Today nobody cares. Does this prove anything? I do not think so.
Three release cycles later, all of a sudden, there's a power off option in the menu right where suspend used to be (with the inverse behavior now! Alt-click -> suspends). I'm glad for it; don't get me wrong. But, what I'd like to know is, what arguments were made to finally convince owen and whoever else pushed through the change? Were the arguments he mead before somehow obsolete? I'd be fascinated to know, since I did my best to make a persuasive case before and was ultimately shot down. (Seriously, if anyone knows of a record of this, I'd like to see it)
What makes you think that Owen changed his mind? or what makes you think that he did the changes? or what makes you think the change was done because of your arguments? I do not know, but it could have been all coincidence and -perhaps- the 'I told you so' argument does not apply here. This seems to be the relevant bug: https://bugzilla.gnome.org/show_bug.cgi?id=647441
Is it possible a tool like loomio could help? I don't know much about it in particular, but I think it's clear that the Gnome development process could greatly benefit from more of what it appears to facilitate.
IMVHO, other problems seem more important to solve in the short-term,
such as:
1. Few people are building and testing the whole system since early
stage of development.
2. Plenty of bugs (usability issues included) are reported late in
the development cycle (after the beta period).
There is ongoing work to improve this situation in the long-term, but
having a system for voting does not seem to help in none of these
unfortunately. It could help in other areas, though.
It is different discussing with empirical data than just discussing
different points of view (sometimes with a partial understanding of the
goals, implementations details or restrictions). IMVHO, the former
helps more.
--
Germán Poo-Caamaño
http://calcifer.org/
Attachment:
signature.asc
Description: This is a digitally signed message part