Hi Leonard,
Are you sure, you know where OOo falls short?
Well sure... but I do think I can make reasonable assumptions.
I am not that sure that *I* myself know it. I may have some thoughts,
but *these* are still only my thoughts, NOT very hard evidence. I am a
guy who worked extensively in the *Evidence Based* field, and believe
me, there are various grades of evidence. Currently, OOo fares worst
that the lowest evidence level. (Yes, this is possible, too.)
This sounded a bit abstract to me, I mean there are various degrees of
'evidence'. When I want evidence, I am thinking of (results of) an
experimental set up... otherwise you're just looking at figuring out
correlations and try to deduce cause effect relationships... that's at
least what I am afraid this is leading to... but...
*Who's the competition?*
This is indeed an important question. And yes we can learn from
competition as well. And more recognized competitors leaves us indeed
more to learn from (I am trying to summarize you here).
When comparing websites, what do we want to compare? What do we want to
do better? Where do we fare bad?
Well I think this is a question that is keeping us busy at the moment :)
The end-user requirements document is an attempt to see how we want to
tackle them. But if I understand you correctly you also want to know why
we suggest changing things, relate it e.g. to competitors? Could indeed
be a wise thing to do, since only by giving proper arguments you can
reason instead of just guess? My counter argument however would be that
it is very clear that e.g. clicks to get to the place should be reduced,
that is something too obvious to me.
Implementing something is not the issue. What to implement is the issue.
Agree 100%
Without knowing what sites to look at, you don't know what to improve.
And without knowing what to improve, there won't be any real improvement.
There I disagree, I think you do not have to agree about sites. IMHO
everyone has plenty of experience with other sites, and takes his or
hers knowledge with him/her. And one of the basic things we want to
strive for is going for simplicity now, reduce clicks -> if we can show
we have reduced clicks & mental effort for most people, that is evidence
for me.
All decisions will be based on the feelings of a single user (or of
those who reply to the mailing list), and they would not capture the
true requirements.
Well I am, and I think many more here, are also in contact with normal
users.
Even IF you had a group of experts, and drew conclusions by consensus,
that would make it only to a level 3 evidence, still far away from a
level 1A evidence in evidence based research.
Hmm... actually I didn't knew about these levels :) ... totally not
something I am dealing with normally (not the codes)... but ok. Found an
overview [1], and think would consider building a better OOo most of a
therapy, but I find it really hard to see similarities in building a
website and trying to find the best alternative, there are simply (too)
many /competing/ qualities a website can have (most important features
vs. simplicity).
More important is the question: what do we want to accomplish? Only then
we can start something that comes closer to clinical trials.
a) just market domination (Word)
b) make people happy when using your package (iWorks)
c) offer as many features as possible (WordPerfect)
d) be free and ... (OpenOffice.org)
e) have a cool identity (iWorks / Firefox)
f) ...
> Please take my rant as a friendly advice. I have some experience with
> studies, and the only consistent way to improve something is through
> meticulous study and application of hard proven methods.
If we can agree on goals, we might be able to set up some kind of
experimental set up. But comparing with websites of competitors leads to
too many confounding variables, which cannot really contribute to
truth-finding, imho. But maybe I have misunderstood you (I am always
eager to learn about interesting new approaches)
g.,
Maarten
[1] http://www.eboncall.org/content/levels.html
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
