What site do you use to search the internet?
— Carl 
I liked Nick Bradbury’s post on the topic of #10 in commercially produced software (even if it is now freeware) http://nick.typepad.com/blog/2008/07/how-to-kill-a-b.html
Barry Schwartz’s TED talk on the paradox of choice mentioned research that showed the more choices you gave people, the less likely they were to actually pick one.
This was in the context of company pension schemes, but there are a lot of other examples in there as well.
LUG radio had a long discussion on the Linux desktop and whether to provide different choice for each type of program. I’d have loved to chip in and say that: If you throw in lots different programs, it will put typical end-users off. Regular people want to achieve a task at minimal mental cost, and having three different text editors simply doesn’t make sense to someone who uses it perhaps once a week.
I have to agree with the Firefox approach, if less than 15% of people are likely to use an option, hide it.
The options dialog is a history of arguments the developers had over UI that they didn’t settle. Its the software equivalent of the Wikipedia Talk page
Whenever I have to add user configurable option to software I’m writing (which these days tends to be various web browsers) I think I’ve failed. If we can’t determine the right value for the option or expose different, important, valid usage patterns naturally through the UI then we’re almost always doing it wrong.
I know the Evince developers are very proud that they don’t have any preferences ui. Compare that to the leading brand PDF viewer.
> If we can’t determine the right value for the option or expose different, important, valid usage patterns naturally through the UI then we’re almost always doing it wrong.
I’ve heard this same sentiment over and over during internal discussions about Google Project Hosting. The restriction of licenses is a slightly different thing, though. UI options are a local fight; license proliferation is a global fight. I think we’re making the right choices on both fronts, though.
— Mark
Proliferation, or ulterior motives?
http://www.theregister.co.uk/2008/04/11/google_bans_aero/
— Stir
I think Steven Den Beste summed it up best when he wrote “… if you give away ice cream, eventually a lot of people will complain about the flavors, and others will complain that you aren’t also giving away syrup and whipped cream and nuts.”
Fighting proliferation is a great choice. I have fought proliferation myself by retiring licenses Sun initiated[1] at OSI. But like others I can’t help suspecting the reasons for hating Affero and MPL are connected with other agendas. The OSI is the authority on open source license validity, not Google.
[1] http://blogs.sun.com/webmink/entry/addressing_proliferation_deeds_not_just
Well, you fight it in your way, and we’ll fight it in ours.
— Mark
Universal Feed Parser a) can be forked b) is useful; (a) mitigates any problem with (b).
Both statements must be negated for Google Project Hosting. Therefore the inequality changes direction: (b) mitigates any problem with (a).
Fight away.
In the 1960s IBM talked about selling their hardware and giving away the software to run it. Their view, then, was that they were a hardware manufacture. We humans seem make good use of “only incidentally useful” items. I appreciate my fingernails, but only notice them when I need them as a tool.