[TriLUG] Re: Linux for the desktop

[Ilan Volow]

On Thursday, November 6, 2003, at 12:34  PM, Maria Winslow wrote:

> On Thursday 06 November 2003 12:00 pm, trilug-request at trilug.org wrote:
>> I have switched to Linux in my home PC about one year ago (RH 8, right
>> now)- Even during the first days of using it, my only issues have been
>> installing new hardware - I have always find the software I needed.
>>
>> I had several people using my computer at different times to do things
>> like browse the web, write letters, or play, without giving them more
>> instruction than "the web browser is called mozilla in this machine, 
>> in
>> the menu on the left" - In my experience, none of them had more issues
>> using it than they would have had using a WinXP pc. Some of them 
>> cannot
>> even tell the difference between Word and Open Office! (yes..I've 
>> asked
>> them out of curiosity).
>>
>> I guess what I am trying to say is that there is nothing wrong or
>> not-usable on Linux for the desktop right now - once it is properly
>> running and configured. The complaints I always hear are always 
>> related
>> to installation or configuration of hardware/software. Those are two
>> very different issues.
>>
>> Just my $0.02
>>
>> Carlos-
>
> If you haven't seen it yet (in English), then some of you may want to 
> check
> out the Relevantive Linux usability study. Relevantive, the 
> Berlin-based user
> interface specialists, released a study this year that compared the
> productivity of average computer users on the SuSE Linux desktop versus
> Windows XP. Neither group received any training, nor did they have 
> previous
> experience with their assigned desktop environment. The Linux users 
> were
> nearly as productive as the Windows XP users in a series of typical 
> office
> tasks. In fact, most of the Linux group felt that they would reach 
> their
> usual level of proficiency within one week. An English translation of 
> the
> report is now available at http://relevantive.de/Linux.html (click on
> "English Translation").
>

That study had some extremely questionable stuff going on in it. From 
what I can remember, one group consisted of something like 20 windows 
folks, whereas the second group consisted of 60 linux folks; such 
imbalance in numbers are going to screw up any statistics generated by 
such a report. It also tells you something about the folks doing the 
study that they index their lists of their report beginning with 0; 
programmers tend to do this sort of thing, and programmer's usually 
haven't a very good track record of evaluating the usability of 
anything.

Another problem with the study is that it looked at the users over a 
very short period of time. To really see how productive a user is going 
to be, you have to take a look at usage over the long-term. A properly 
done usability study on linux vs. windows would look at productivity 
over the course of a year or more.

An article having a fairly balanced summary of the study can be found 
at http://zdnet.com.com/2100-1104_2-5060709.html

One additional thing that bothered me about the study (that the article 
mentions) is that problems caused by something like KDE's use of the 
term "Directory" is something that any usability person worth their 
salt would have been able to recognize as being problematic even 
without user testing. "Directory" is system-oriented jargon, and it 
breaks consistency by not matching metaphor (the folder icon) with 
terminology (the word "folder"). For those who would accuse me of "not 
liking the word 'Directory' because it's not like Windoze", they are 
more than welcome to go to usability guru Jakob Nielsen's website and 
see that this 'Directory' is clearly in violation of usability 
heuristic #2. 
http://www.useit.com/papers/heuristic/heuristic_list.html. 
Incidentally, it's kind of depressing that Apple got their terminology 
right in 1984 when the used the word "folder" with the first macintosh, 
but after so many years KDE can't even make a simple change in 
terminology that would introduce no bugs and greatly improve the user 
experience.

Don't get me wrong. I'm all for incorporating usability into linux, as 
are all the usability folks who for the last 5-6 years have been 
screaming at apathetic linux programmers/desktops/distros to make the 
needed changes to make linux usable. But I believe it should be done 
without unix geek bias and by people who know what they're doing.

--Ilan

--
Ilan Volow
Ergonomica Auctorita Illico!