Re: intent -driven search

14 Jun 2005 - 8:22pm
9 years ago
3 replies
377 reads
Kesava Mallela
2005

Intention behind the new search tool sounded good. But the slider and
results were bullshit.

First of all, how much sense did the slider make? What does it mean
when I slide it midway towards shopping? Does it mean that I have
assigned .75 weight for shopping and .25 for research? What do those
weights .75 and .25 mean anyway? I mean, how do you differentiate them
from .95 weight on shopping + .05 weight on research? Does it mean
that somethings are inherently more commercial (and offer lesser
important information)?

When I slide it towards the research end, results which normally
appear in pages 8 or 9 show up. So does research mean looking for
stuff that is hidden deep inside?

I think that shopping/research preferences should have been more of
black or white(radio buttons) instead of the slidebar.

I thought Google's personalized search (though broader than sales and
non-sales) was better. It solved the above problem with check boxes.
:).

Choice is good. But more of it may not necessarily mean it is better.

Kesava Mallela
Interaction Designer
http://theonewhere.blogspot.com/

-------------Original Message from--------
vimal sharma vimalsharmaa at yahoo.co.in
Sun Jun 12 22:55:53 PDT 2005

hi all,
latest from yahoo's research lab is yahoo mindset which adds to
yahoo search engine as a slide control at the top of the page,the left
hand shows shopping and rightend is researching having entered your
search u can use the mouse to slide the button to the left to get more
commercial results.sliding to the right gives you more analytical or
informative results....
......................."this is a great tool to seperate the wheat
from the chaff of the online search results. What is wheat and what is
chaff is of course,your personal preference."the hindu.
http://mindset.research.yahoo.com

as philips says "lets make things better"

regards
Vimal Sharma
IDC
IIT Bombay

Comments

15 Jun 2005 - 6:12pm
Jonathan Grubb
2004

Kesava wrote: Intention behind the new search tool sounded good. But the
slider and results were bullshit.

Now Kesava, just because someone works for an impersonal faceless global
corporation doesn't mean they don't have feelings. George's feature is not
bullshit. If you take a moment to really look at it you can see that.

Try searching for Interaction Design on yahoo mindset:
http://mindset.research.yahoo.com/search.php?p=interaction+design

At 50% shopping / 50% researching you get:
#1: The companion website for the book Interaction Design
#2: Interaction Design Inc, an IxD consultancy
#3: Interaction Design Patterns -- maintained by Tom Erickson

At 100% shopping / 0% researching
#1: A book on amazon (formerly result #46)
#2: another book on amazon
#3: an IxD consultancy

At 0% shopping / 100% researching you get:
#1: Stanford HCI group
#2: InterctionDesign.org
#3: Interaction Design Patterns -- maintained by Tom Erickson

So obviously the thing is working. But do we *really* need a slider? Why not
just use checkboxes as you suggest?

As I slide from 100% researching to 90% researching Stanford HCI drops from
#1 to #3. Interaction-Design.org jumps up to #1. To me, this implies that
Stanford HCI is slightly more academic and interaction-design.org is
slightly more practical (no judgment implied - both are pretty interesting).
After a 2 minutes of research I see that this is pretty accurate; Stanford
has hard research while interaction-design.org.

Kesava wrote: I thought Google's personalized search (though broader than
sales and non-sales) was better. It solved the above problem with check
boxes.
:).

On google personalized I'm asked to set up preferences saying what I'm
interested in. There's no checkbox for Interaction Design, but I can say I'm
interested in computers>> internet, mobile computing, and software. Setting
these preferences caused no change at all in the results (though it does
remove all sponsored links).

I went back and checked every option under Computers and tried again. I was
honestly surprised to get A SLIDER BAR! Kesava was wrong. Google also uses a
slider bar. Sliding all the way to maximum personalization gives two links
to welie.com and one to an interview on extreme programming.

Kesava wrote: Choice is good. But more of it may not necessarily mean it is
better.

Usually when shilling for Google it's safe to say "google product x is
better because it's simpler." In this case it doesn't work out; the yahoo
product is simpler and, in this Yahoo shill's opinion, it gives better
results.

- - - - - - -
Jonathan Grubb
Yahoo! Mobile
- - - - - - -

15 Jun 2005 - 10:21pm
Kesava Mallela
2005

On 6/15/05, Jonathan Grubb <jgrubb at yahoo-inc.com> wrote:
>
> Kesava wrote: Intention behind the new search tool sounded good. But the
> slider and results were bullshit.

First of all, apologies for the swear word.

Now Kesava, just because someone works for an impersonal faceless global
> corporation doesn't mean they don't have feelings. George's feature is not
> bullshit. If you take a moment to really look at it you can see that.
>
> Try searching for Interaction Design on yahoo mindset:
> http://mindset.research.yahoo.com/search.php?p=interaction+design
>
> At 50% shopping / 50% researching you get:
> #1: The companion website for the book Interaction Design
> #2: Interaction Design Inc, an IxD consultancy
> #3: Interaction Design Patterns -- maintained by Tom Erickson
> ....
> As I slide from 100% researching to 90% researching Stanford HCI drops
> from
> #1 to #3. Interaction-Design.org <http://Interaction-Design.org> jumps up
> to #1. To me, this implies that
> Stanford HCI is slightly more academic and interaction-design.org<http://interaction-design.org>is
> slightly more practical (no judgment implied - both are pretty
> interesting).
> After a 2 minutes of research I see that this is pretty accurate; Stanford
> has hard research while interaction-design.org<http://interaction-design.org>
> .

Those are some of the subjects with which you are extremely familiar...and
the results probably made sense.

When I was looking at a less familiar subject, the slider made me less
comfortable.
Thats because I am not fooling around...and I want my results. Slider only
adds to my confusion...as I am not sure what midway exactly means.
So I tend to put it in extremes just to make sure that I am not wasting my
time in searching.

When you are familiar with the subject, honestly, whats prompting you to put
the slider midway?
The slider might work but how is your 'intention' behind 90% research
different from 80% research? I mean, how differently do you anticipate?
how differently do you quantify?

Kesava wrote: I thought Google's personalized search (though broader than
> sales and non-sales) was better. It solved the above problem with check
> boxes.
> :).
>
> On google personalized I'm asked to set up preferences saying what I'm
> interested in. There's no checkbox for Interaction Design, but I can say
> I'm
> interested in computers>> internet, mobile computing, and software.
> Setting
> these preferences caused no change at all in the results (though it does
> remove all sponsored links).
>
> I went back and checked every option under Computers and tried again. I
> was
> honestly surprised to get A SLIDER BAR! Kesava was wrong. Google also uses
> a
> slider bar. Sliding all the way to maximum personalization gives two links
> to welie.com <http://welie.com> and one to an interview on extreme
> programming.

Well, I didnt realize the slidebar in google until Jonathan pointed out.
Google uses the slide bar to mention a degree of personalization.
They are not using it, in asking you to make a choice of subject (or
intention) of search. They still use check boxes for the subject.
I believe you get the difference.

Kesava wrote: Choice is good. But more of it may not necessarily mean it is
> better.
>
> Usually when shilling for Google it's safe to say "google product x is
> better because it's simpler." In this case it doesn't work out; the yahoo
> product is simpler and, in this Yahoo shill's opinion, it gives better
> results.

I should partially agree with Jonathan. Yahoo has smaller no.of choices in
terms of areas (shopping and research). Google has lot more there.

--
Kesava Mallela
Interaction Designer
http://theonewhere.blogspot.com/

16 Jun 2005 - 12:15pm
George Olsen
2004

On 6/15/05 5:12 PM, "Jonathan Grubb" <jgrubb at yahoo-inc.com> wrote:
> Now Kesava, just because someone works for an impersonal faceless global
> corporation doesn't mean they don't have feelings. George's feature is not
> bullshit. If you take a moment to really look at it you can see that.

Actually, just to clarify, I had nothing to do with intention-based slider
-- I helped design Y!'s My Web -- so my feeling weren't hurt a bit. Even if
I did design it, attacking my designs isn't attacking me.

Personally, I agree it's kludgy and there are far better solutions -- and
Google's equivalent ain't one of them.

George
___________________________________________________
George Olsen Principal, Interaction by Design
650 329 1728 george at interactionbydesign.com

Syndicate content Get the feed