On 10/08/2009 12:54 PM, Tom "spot" Callaway wrote:
* I am troubled by the subtext of the repeated questioning of "What is Fedora", because I really don't think that is the question that is being asked. Instead, I think the question being asked (or at least, the question being answered) is: "Who is Fedora (the Linux Distribution) for?" If indeed, we are attempting to answer that question, it implies that there are users for whom Fedora is not, and will never be, a good fit. Even if this is practically true, it bothers me that we are saying that out loud. We're telling folks, go somewhere else if you don't fit into this nice box, don't bother trying to improve Fedora in those areas. If the Board were to say, that there are specific areas where Fedora would like to see more contributions, or even, primary areas of interest to the Board, I could probably be on board with that, as long as it was phrased in a way that made it obvious that it was not exclusionary.
I agree. When we define a target set of users for Fedora, there is a risk of being exclusionary. We shouldn't and don't need to be exclusionary, however. At least not from a design-needs perspective. :)
An example I tell people about a lot is Bob Plath, the inventor of rolling luggage . He was an airline pilot and he designed it specifically with flight crew in mind - not passengers. He wasn't looking to make luggage that was better for everybody. He was looking to make luggage that made life easier for flight crew specifically. That focused his design to be good enough that it just so happened to be generally useful to everyone. I think it would have been hard to come up with a good design when trying to please everyone - you end up in this kind of hell: http://blog.monochrome.co.uk/2009/02/if-architects-had-to-work-like-software-developers/
The 'Fedora is for everyone' path also has a big risk though, and that is the other thing I warn people about - the Alan Cooper car. (see attached.)
This is why I think the best approach is to pick a primary target set of users, design to them - putting your primary users at the top of the priority when making design decisions - but don't exclude anyone else. Keep things open, but definitely draw a line in terms of where your priorities lay so you can make good consistent decisions that support your primary user rather than wily-nily decisions all over the place and end up with an Alan Cooper car.
Hypothetically, if the Desktop SIG answered "Who is the target audience for the Fedora Desktop Spin?" with "Experienced Linux users and developers", I could easily see the value in organizing a "Fedora Simple" Spin, where the answer is "New Linux users", with a separate spin. And honestly, in that hypothetical situation, I'd strongly consider pushing for the "Fedora Simple" spin to be our default offering. I do not feel that good usability and new-user-friendlyness are at all in conflict with the needs of experienced Linux users and developers.
I agree. I think that more experienced users and developers are a little frightened of a Fedora Simple - there's certainly a risk of making it annoying for them, but I don't think that will necessarily be the case. I think if we do things right, in the end a Fedora Simple would be something good for advanced techies to use as well.
E.g., I don't think OS X's primary target is developers, but I think there are plenty of highly-technical developers and other folks who enjoy using OS X.
I think the thing that leaves me unsettled is that all of this seems to stem from the fact that some new users are not having a great experience with Fedora, and rather than analyze that problem and work on usability improvements, we are choosing to let those users go somewhere else, whether that is Mac OSX or Ubuntu or Windows. I think that such a decision is terribly short-sighted, and will result in a long-term loss of contributors, community, and possibly the eventual irrelevance of Fedora.
Absolutely agreed. ~m http://money.cnn.com/magazines/business2/business2_archive/2005/03/01/8253083/index.htm
Description: PNG image