[Theforum] Re: Sub-categories

[Madhu Menon]

At 12:55 PM 12/16/2001, Ben Dyer typed these words:

>>Typically, classification ala Yahoo becomes necessary when you have far 
>>too much content to place in one category. Putting 300 articles into one 
>>category becomes unwieldy.
>
>Agreed...*but*, Yahoo! has how many items in its database?  How many does 
>evolt have (or will ever have)?  If you don't know where Yahoo! classified 
>something, you're in deep poop if you can't find it after a few 
>tries.  With evolt, there are far fewer choices.

Which is exactly what I said. Yahoo has several levels because of their 
content. We don't. Hence my suggestion that we not classify evolt too much.  :)

>[snip an example]
>
>This is true, *but*, this is based on the thought that this is how users 
>will search for articles on the site.

If I were searching for any article *without* using the site search, what 
else could I use besides the categories?


>I think that one-level-deeper categories may be more useful in the 
>following capacity:
>
>Say you read a great article on ColdFusion (I'm more comfortable there 
>than ASP. ;) and you think to yourself, "That was a great ColdFusion 
>article, I wonder what other information is here on ColdFusion."

Yes, that's a good use for them too. I often do this.


>If they back up to "Code", it's hit or miss in finding what they're 
>looking for, but "Code > ColdFusion", they get exactly that.

You're just agreeing with me :P


>But, then again, others may not be using the site in this way, I'm just 
>describing how I might use it.  Regardless, how users use the site is the 
>first thing we should find out, and the survey will help some with that.

Actually, we didn't ask such questions in the survey. I don't recall seeing 
any questions about whether they thought the structure was intuitive, 
whether they used the search a lot, etc. So don't expect the results to 
tell you much.

In any case, those are best done in actual usability tests where you 
*observe* people using the site.

>Very true.  But again, there is also a difference between how users act 
>versus how they want to act.  The only way to find out that is to ask them.

There's a bigger difference between how they *say* they act, and how they 
actually do. Asking is not very reliable.
Uncle Jakob is right about this one.


>So, bottom line is, I guess I'm saying that everything keeps coming back 
>to the survey results.  Once we pore through those, then we'll have our 
>answers.

Maybe. We won't find *all* our answers, however.

>And...I'm spent.

I have that effect on people ;)

Cheers,

Madhu

<<<   *   >>>
Madhu Menon
madhu at members.evolt.org
Blog: http://madman.weblogs.com