Re: Reboot: Strategic goals for GNOME

["Jason D. Clinton" <me jasonclinton com>]

On Wed, Mar 3, 2010 at 3:09 AM, Dave Neary <dneary gnome org> wrote:

 vision for GNOME 2.x did.Back in February, I posted the following - it
kind of got lost in the ensuing thread; but I think it's worth breaking
out into a new discussion (marketing list CCed). Like I say, I'm not
happy with the "vision" part of this (GNOME everywhere, and invisible)
because it doesn't offer a destination - it doesn't help anyone make
decisions about what's important - in the way that the "simple, usable,
beautiful" But perhaps it's the beginning of a vision that we can work on?

Hi Dave;

Let me update you on what the marketing team did at the Chicago hackfest in mid-November[1][2] since we haven't yet made our announcements.

The most difficult thing that we did at the hackfest was to settle on a short-term vision for Gnome 3 which will, I think, help shape this discussion about medium-to-long term. Our starting point was this email from Vincent: <http://mail.gnome.org/archives/desktop-devel-list/2009-April/msg00004.html>. We interpreted that directive to mean that all Gnome marketing efforts should be directed at the first point: selling users on the revamped user experience.

This is--perhaps--slightly different from how a corporation marketing department operates: finding what the market wants and then directing engineering resources at that. Instead, Gnome 3 is whatever it will be and it's our job to take that and sell it to our audience. Here is what we (then) saw Gnome 3 as:

There was wide ranging discussion about what GNOME 3.0 is and how much of a change it is. We talked with each other about what each of the proposed technologies will bring to the end user experience: tightly integrated search, application-oriented window management, dynamic workspace workflows, temporally oriented document location (Journal), interruption suppression (message tray), and user interface physicality (via window and Shell chrome animations). There was also agreement that aside from "3.0" technologies, there's a number of long-stewing features coming to fruition that can also be emphasized: Telepathy, for example.

There was an observation that we're achieving a new level of tight integration throughout the desktop environment. Think of IM notifications with rapid dismissal in the messaging tray in Shell. Or, think of Empathy sharing your Desktop with you friend via Telepathy without you needing to open a firewall port. Another example could be how both the Journal and Tracker technologies (potentially) tightly integrated with the Shell UI.

The following morning that was boiled down to:

First, we agreed upon the themes--the Shell, the search work, the new levels of cross-application integration--all substantially advance the release team's primary point of the 3.0 release: a better user experience.

We went through over a hundred themes before we settled on what Gnome 3's marketing theme will be: "Made of Easy." (Pause for dramatic effect.)

Up until this past week, I was really happy with that. When I saw Seth's blog post and the subsequent press coverage, I understood that the Gnome 3 marketing train had just gone off the rails. What, until then, had been a well-defined expansion of our user experience in Gnome 3.0 had just been radically re-cast as what can either be seen as a partially implemented, half-baked version of our long-term vision for Gnome 3.x's usability end-game or, worse, a experiment thrust on our users until we figure out what Gnome 3.x will be.

Let me state this clearly: until last week, Gnome 3.0 had a barely-achievable time frame and a well defined goal for users to look forward to. Today, Gnome 3.0 is looking like a repeat of GNOME 2.0 and KDE 4.0's marketing disasters. We are dangerously close to repeating history, again. Specifically, we had set expectations in the outside word to an achievable level before and now they are completely in-achievable. Gnome 3.0 will disappoint because it doesn't have what Ars and others mistakenly took as our plan.

But we can fix it.

We need to put some reality check in place on what is getting out to the press. Everything in Seth's posts are just ideas. They aren't Gnome 3.0. And they aren't going to happen in time. In fact, production of marketing video assets for Gnome 3.0 begin in a little over 30 days. We knew that there would be risk starting this early but the simple fact of the matter is that video production takes loads of time and resources; we have to start this early. And as a matter of practicality, there's a lot of code to write to get anywhere near what has been proposed at the usability hackfest and not a lot of people available to do the actual writing of that code.

In addition, from a marketing perspective we, publicly, *cannot* promise anything as radical as proposed by Seth (or others) for delivery *after* Gnome 3.0. Here there be dragons:

1. It makes us look like 3.0 is half-baked
2. It makes us look like we're experimenting on our user base
3. It makes our desktop environment look unstable

On the other hand, if such changes happen organically, so be it.

Well, we have a choice between two paths forward:

1. we agree that Gnome 3.x is what was the plan before last week and that any other usability changes beyond that are merely organic and non-disruptive, or
2. we delay Gnome 3.0 until all the disruptive changes we want to make are ready

For better or worse, a dot-oh release is a stake in the ground: "here is what we are committed to." If we aren't committed to what we deliver in Gnome 3.0, then it's not 3.0.

We need to decide, like, this week--seriously. And whatever we decide shapes our discussion of Gnome 4.x.

[1] http://jasondclinton.livejournal.com/77635.html

[2] http://jasondclinton.livejournal.com/78061.html