Re: [tlug] Browser share in Japan?

["Stephen J. Turnbull" <stephen@example.com>]

Francois Cartegnie writes:

 > - 40% of development time/costs is just for dealing with IE 
 > incompatibilities...

Nobody said otherwise that I can recall.  Developer time is known to
be a scarce resource, and nobody is suggesting that it be used for
supporting non-revenue-generating users.

 > - Today, IE6 is officially deprecated and at most 5%.

Of *what*?  That's the whole question here.  My point is simply that
"what" != "market" in any useful sense for the figures you are quoting.

 > - Mobile web grown over 5% this month. [1]
 > 
 > Depending of the dev/revenue ratio, it's probably smarter to use
 > developer's time to bet on mobile and not on IE6.

Did your mother put asparagus in your bento again?  I've never seen so
many people eager to give away their lunches!

No, it's not "smart" or necessary to "bet" on mobile.  What's smart is
to consider doing market research to find out what browsers your
actual customers (if you have any yet) and potential new customers are
using, and what the trends are for adopting new browsers and/or
platforms in those groups.

Admittedly, it's also expensive to do that.  Sometimes it does make
sense to take global numbers as a less expensive, sufficiently
accurate proxy for the information about the actual market segments
you're interested in.  Darren Cook made a good case for using global
numbers as a proxy in the case of his client.  But this needs to be a
conscious decision by the responsible manager.  And most often, the
appropriate way to make the actual decision is on the basis of a
combination of current numbers (gathered by your own webserver),
global numbers (as a proxy and check on potential customers), some
amount of market research (into potential customers and trends into
the future), and manager intuition.

Loose use of the term "market share" tends to obscure that.