Member stats (was: [thesite] [bug] t.e.o. - blah blah)

Madhu Menon webguru at vsnl.net
Sat Nov 17 01:30:28 CST 2001


At 07:37 AM 11/17/2001, the great .jeff spake thus:

>there's nothing that says we can't add this information there as well.
>that's something i've wanted to do for a long time, but haven't come up with
>a sensible way to present the information.

That's my point. Don't add it "there as well"; add it *only* there. Is 
there really any compelling reason that it *needs* to be on the article 
page? Is this information *critical*? I think not. IMO, the only things 
that might make a case for being there are "Articles written" and "average 
rating"


>try this one instead then and see if it looks better.
>
>http://test.evolt.org/article/thesite/17/16286/

Nope. Still a whole bunch of numbers. I'll be frank. When I looked at them 
for the first time (and even now), I sat there for a minute or two trying 
to make sense of WTF was being conveyed to me. Call it poor information 
design if you will, but it takes a while to digest those numbers and make 
*sense* out of them. "23 comments" - what exactly does it tell me? That 
Jeff doesn't comment much? That's he just joined evolt and that explains 
only 23 comments? Is this necessarily worse than Dan's 86 comments? What if 
Dan's comments are mostly "Oh yeah! Great article" (Dan, I'm using this as 
an example only, nothing personal) and Jeff's comment's are higher quality? 
Do the number of comments indicate their quality? If not, what is it trying 
to indicate? If it doesn't indicate *anything*, what's the purpose?
It might make *some* sense if combined with your joining date: 04/15/1999. 
Then I might be able to tell how often you comment. But even that is 
largely inconsequential to me, the reader of the article. It's just a bunch 
of stats and don't tell me much about your knowledge of a subject.

"Avg. rating given: 4.42" - so Jeff generally rates articles high? Or does 
he only rate articles if they're high quality? (Besides, I'm not sure I 
*want* people to know what I rate articles on an average, but that's a 
different discussion)

OK, let's have our steel cage match.
Compare Jeff Howden (http://test.evolt.org/article/thesite/17/16286/)  to 
Madhu Menon (http://test.evolt.org/article/thesite/4090/4299/index.html)

Jeff has received 37 comments on 11 articles. I have received 17 comments 
for 3 articles. Comment for comment, I come out on top with 5.67 comments 
per article over your 3.36 comments per article. Does that make my articles 
better? Yes? But your average article rating is 4.5 compared to my 4.07. 
Huh? So comments are not an accurate measurement of anything? If it's not a 
measure of anything, why have them? I can extend the same argument for most 
of the numbers on the article page.

In short, what *information* is the reader getting from seeing this *data*? 
(The difference: processed data presented in a meaningful format becomes 
information). Look at things from a user-centred perspective.

I compare this to some clueless PC magazines (and there are a fair few of 
them) telling me that the XYZ processor returned a score of 34.8 on their 
benchmark. Is that good? Bad? 34.8 compared to what? Does it make a 
difference?

"23 comments" - good or bad?

Let's not have an orgasm over cool SQL queries. Presenting data is 
pointless. It must make sense to me.

We need to go back and read Tufte's "Visual display of quantitative 
information", and perhaps "How to lie with statistics".

Madhu
PS: This would be a good discussion for theforum. Shall I move it there?

<<<   *   >>>
Madhu Menon
User Experience Consultant
e-mail: webguru at vsnl.net





More information about the thesite mailing list