Jonathan Duncan wrote:
On a tangent, how is this experiment related to what I assume is a
larger project of creating a "genetically coded web site"? I assume
that every aspect of the web site may start off like this? Is this
the first part of the site that you are experimenting with or do you
already have parts of the web site genome assembled?
What you are wanting to do, seems incredibly huge and unprofitable
from a monetary perspective. I can see how it could be beneficial
perhaps for purely scientific/academic purposes. Unless your goals
are much further reaching than the near future.
Trying to decide the color of a button/link without context might not
give you the results you are looking for. Unless, again, at this
stage all results are purely academic.
Cheers,
Jonathan
Side note: No matter what the outcome, the description, "genetically
coded web site", has some pretty good PR value to it.
Jonathon,
I have just recently started reading about genetic algorithms and
artificial intelligence (see
http://www.ai-junkie.com/ga/intro/gat1.html), it is very interesting.
Most the world thinks of artificial intelligence when they play computer
games. There is a lot of money in the gaming industry also. While I love
computer games, I also have a passion for business.
So, as I thought about AI for the last little while, I started thinking
about how I could apply AI to businesses. The thought came to me as I
was looking over my analytics, that companies spend hundreds of
thousands of dollars to create awesome websites that do what?
Try to get a customer to click a button.
It is that simple. Companies (not necessarilly organizations) exist to
make money. So why couldn't AI be applied to getting a user to click a
button?
Would it be difficult, and a lot of coding? Yes.
Would it make me filthy rich? Most likely not.
Would it work? I belive it could.
Think about a nation-wide website that sells books. With good AI
built-in, it would know (theorizing here), that most consumers out of
California, buy books that are about yoga when showed on a light green
background with a dark tan "Buy Now" button, so why not show yoga books
to Californians on a light green background, and a dark tan "Buy Now"
button.
This kind of information would be cost prohibitive to ab test for all
types of books.
As suggested earlier, if this were built into a CMS, like Drupal, or
OSCommerce, it would greatly enhance the effectiveness of those websites.
Companies, would pay for that.
Justin Giboney
_______________________________________________
UPHPU mailing list
[email protected]
http://uphpu.org/mailman/listinfo/uphpu
IRC: #uphpu on irc.freenode.net
