[Proposal for Vote] Approve Draft 6 of the PRD

[Owen Taylor]

On Tue, 2014-01-14 at 14:41 +0000, Matthew Garrett wrote:
> On Tue, Jan 14, 2014 at 07:50:28AM -0500, Josh Boyer wrote:
> > Per our governance I'm calling for an official vote on the sixth draft
> > of the PRD that Christian has provided.  I've attached it again for
> > reference.  WG members have one week from today to vote.  Missing
> > votes after one week will be counted as abstains.
> 
> This version still contains an explicit statement that the target 
> audience is made up of developers. I've explained why I think this is 
> damaging, and Christian obviously disagrees with me. However, I'm 
> disappointed to see very little feedback from any other members of the 
> working group.

I think it's a mistake to have a secret agenda - to declare one thing
and actually work on something else. Nobody is taking Fedora and
building a consumer desktop with it. For structural reasons, it's hard
for Fedora itself to be a slick out-of-the-box consumer desktop - Fedora
isn't set up to pursue relationships with consumer content providers and
OEMs, and has historically been unwilling to ship DRM and closed source
software. On the other hand, Fedora *is* used as the basis of a
workstation product for developers and related types of users (RHEL) and
is also in itself widely used by all sorts of developers.

A secret agenda is hostile to people coming into the project - they
don't understand what they should work on, why people are working what
they are working on, why decisions are being made. So we shouldn't be
saying that we are building a desktop with a target of everybody unless
that's what we are *actually* putting resources on. Vague aspirations
don't cut it.

But beyond that, one big problem that Fedora has is that a lot of Fedora
community members think that they are far from the target audience -
that whatever they say is being ignored because the desktop is being
designed for school kids and grandparents using tablets to play games on
Facebook. 

Centering the PRD around developers is a statement of the common purpose
that *actually exists*. A desktop designed primarily playing games on
Facebook would be useless for all the people contributing to Fedora -
for Red Hat and for the community.

Of course, if "for developers" is read as "can have wires sticking out
all over the place", then we're in trouble. But to me, having polish and
coherency and quality is something that proceeds and is a prerequisite
for any sort of target audience. Targeting developers doesn't mean
giving up on the basic design principles we have:

 * The user has better things to do with their time than fiddle with
   the operating system.
 * Configuration options have cost.
 * Understand the real problem that the user is having, don't
   stop at their feature request.

And so forth.

Many of the other objections you've raised I think come down to
misunderstanding the type of developers that we're primarily targeting.
There's no idea that there will be an ecosystem of native application
developers developing native applications for each other. And certainly
we're not building an operating system for people who build operating
systems.

Server side application development is really the primary target, though
other types of application development - embedded, hobby, etc. are
within scope. 

(It might be clearer if the PRD separated developers by what they were
dong rather than concentrating on their employment status.)

In the end a PRD that simply said that Fedora Workstation is for
everyone would be of little utility in decision making, or even of
forming an identity of the product. A strict focus only on developers
would be limiting, but a soft focus in that direction seems to me to be
a productive way to guide the initial development of the product.

+1 to approve Draft 6 of the PRD as stands

- Owen