[thelist] targeting effectively

[Paul Houle]

    With regards to targeting applications for different browsers:

    Something funny I've noticed is that if anybody is going to be
running an obsolete or funky browser it's going to be one of life's
gatekeepers:  you know,  the professor who's the PI for a project you're
working on,  the venture capitalists that you're trying to get funding
from,  or the person who's interviewing you for a job.  They're going to
be the people who are running Netscape 3,  surfing with cookies turned
off,  or have the flash plugin disabled.

    So if the 5% of people you exclude include the right person,  that
can sink your project as surely as if you excluded 95% of them.

    If you study the usability of even a very good site you're going to
get depressed.  For instance,  as part of a user registration process we
worked very hard to get an 80% success rate.  This was all a matter of
finding various reasons why 2% of users were getting lost here,  and
another 1% of users were getting lost here.  (One big thing we can't fix
is that a lot of users give fake e-mail addresses even though we tell
them that registration won't work if they give a fake address.)  Now,
 even if the use of (flash | DHTML | whatever) costs only 5% of users
here,  it's only one of a number of aspects of the site that are going
to confuse users.  So the success rate of a task goes,  say,  from 65%
to 60%.

    If that was all of it,  it wouldn't be so bad.  But the kind of web
designer who'd be willing to throw out 5% of the user base so they can
win design awards and be able to make windows pages (not web pages,
 because they're coding for windows only) by only pointing and clicking
rather than learning how to code HTML by hand is also going to be the
kind of web designer that's going to throw away 3% of users here because
they've got a fetish to use the latest feature in IE and 7% of users
there out of sheer ignorance,  and pretty soon the success rates are in
the 20-30% range and the conversion ratio is 0.1% and then they're
coming around and complaining that there aren't any profitable web sites.

    That all goes double for the clients who think they're doing
themselves and their investors a favor making a page that's difficult to
use,  expensive to develop,  and expensive to maintain.

    The most difficult part of making a web site in 2002 isn't making
the web site,  it's getting visitors to come.  So many sites are
screaming for attention that it's almost impossible to get a foothold.
 Every visitor that shows up a site has to be treated like gold.

    In any case,  the vast bulk of work I've done in Flash is real
dreck.  Flying letters,  spinning logos,  dancing credit cards,  all
that junk.  That's not to say that people can't do great stuff with
Flash,  but that's less than 5% of what people are doing with it.