Updated Fedora Workstation PRD draft

[Matthew Garrett]

On Wed, Nov 27, 2013 at 03:00:11PM -0500, Máirín Duffy wrote:
> On 11/27/2013 02:39 PM, Matthew Garrett wrote:
> > To do that we need an audience who's able to evangelise to a wider base. 
> > Are developers going to do that? And if they are, how do we attract them 
> > in the first place?
> 
> Well, wasn't it developers and sysadmins evangelizing from the bottom up
> that got Linux taken seriously and deployed within enterprises from a
> server POV? So I think yes, if they like something they will evangelize it.

At the time, Linux was competing against operating systems that cost 
hundreds of dollars (if not thousands) a seat. Were people evangelising, 
or were they just pointing out that it was cheaper?

> We're more attractive than OS X I think in that we're built on top of a
> base that is closer to the platforms they're deploying to (assuming a
> web / server developer of course not desktop / mobile.) Even better, if
> the three product Fedora.next plan goes well, they'll have server and
> cloud versions of the environment to deploy to - there is no equivalent
> that I am aware of for OS X, except for - again - desktop and mobile app
> devs that are targeting OS X and iOS. Even better, those server/cloud
> versions are from the same family as the leading enterprise Linux
> product already widely used in the market.

We build a Linux-based product. Most of my coworkers run OS X. Client 
code is all in Python and runs fine on OS X without modification. For 
server code, they just work remotely. My experience is that this isn't 
unusual. Running Linux locally might provide some marginal benefit in 
code testing, but that's at the cost of running an OS that they just 
don't like as much. They're developers, not sysadmins. Satisfying their 
development requirements is easy. But they also want to be able to 
upgrade their iphones, watch movies, listen to music and play the 
occasional game. *I* can't convince them to switch to Fedora. Whatever 
its merits as a development platform, they just don't find it a 
sufficiently useful operating system.

> Our multi-monitor story is pretty poor as well (eg
> https://bugzilla.gnome.org/show_bug.cgi?id=712325 .) It seems common or
> at least desirable for developers to have a multi-monitor set up, from
> my days doing contextual interviews on site with RH customers, and
> walking around Google's office in Mt. View and even Red Hat's office
> here in Westford. If we focused on making a kickass multi-monitor
> experience that bettered OS X, that would be a good step in the right
> direction.

I agree that merely being as good as OS X isn't sufficient and we need 
to think about ways that we can offer concrete benefits, but that still 
means we need to offer an experience that's approximately as good as OS 
X. And that means we need to think about more use cases than just 
development, because these days the development laptop is often also the 
casual use laptop.

> Anyway we could do a review of complaints devs have about other desktop
> systems then do an affinity map of the complaints, block them out into
> different focus areas (e.g., multi-monitor), and then attack them
> one-by-one.

That sounds like an entirely worthwhile thing to do.

-- 
Matthew Garrett | mjg59 at srcf.ucam.org