On Mon, Jun 07, 2004 at 09:26:39PM +0200, Michael Schumacher wrote: > Carol Spears wrote: > > >in the united states, there exists a condition where people are being > >educated for a test. teachers that do not teach the content for the > >test are removed. > > For usability tests? I doubt that this is the case, though it might > explain the usability of the software produced by a very large company > in the US... > > >i think this sort of thing is being introduced to gimp development. i > >think it is very good to have the testers show what their purposes were > >before much of the good developers time is taken. especially when you > >are dealing with such high quality developers and free software. > > Hm, I think that I'm at least partly able to understand what you're > trying to say - if someone want to do something with GIMP, he should ask > for advice before he demands GIMP to be changed. Or at least that he > should learn and shut up if he is told the correct way afterwards. > > Good advice for anyone on a mailing list or on irc, but not in a rather > controlled usability test. > no, i really meant what i said. i am actually very nervous about this because i had in real life a roman who failed a test that was made for testing purposes. this was one of the most intelligent, well read and roundly educated human male beings i had ever encountered. too young to date but just one of those kids you wished you had recognized when you were younger, a world changer and one of the right ones to do it. because he failed a test like this in the money world he lost some good cash that he could have used in college and the money was rewarded to a little creep, whose intelligence i also recognized. but without the good heart. then there is gimp, where our roman has volunteered for somethings i cannot support. i want to keep roman and get rid of the stuff that screwed things up for this other roman. if you were to tell me that this roman is being paid for using that crappy docbook set up and spending time with crappy tests i certainly would feel better about these developers spending their time with the test results. roman did really good work within a very bad set up. this is the sort of thing you should be paid for. i am still under the idea that this is free software and i am dealing with volunteers. this is not a test for volunteer work. > >show what you the human being wanted the gimp to do is a very very good > >question. > > Hm? Not sure if I understand correctly, but are you thinking of > questions like "What is the best way to...?". I always wanted to start a > wiki page about this - finding the absolutely fastest way to perform a > task in GIMP - but couldn't find a real catching name for this page. Any > suggestions? > well like find someone who has just started to use linux and the gimp who has a photo they would like to cut the background from. see how long it takes them to do this. you have to actually want to use the gimp for an actual outcome. it is made for production. then you see how well it goes. "how easy is it to use" i hope we would fail this. "how well does it complete actual tasks". we won this with gimp-0.54 i think. long before i started to use it. carol
