The $100 laptop

[FoulDragon at aol.com]

In a message dated 10/10/2005 10:27:40 PM US Mountain Standard Time, 
plug-discuss at stcaz.net writes:
>Furthermore, none of that supports your premise, since there's no incentive 
to write the >software to run ONLY on the proposed laptop, in fact there's 
strong incentive to ensure it >runs on a wide variety of systems, so that the 
investment isn't wasted if someone donates >a large group of more capable systems 
to, for instance, schools.

Okay.  Two points.

1-  In all likelihood, the $100 laptops will be the slowest machines in wide 
circulation.  So designing "for this machine and many others" will mean 
designing for this machine.  Most of the time.

b-  There are some things where the optimal answer for some systems is 
suboptimal for others.  For example, I had two portables with CGA (640x200) 
displays.  One had a 4:3 screen and another a 16:5 screen.  Rendering code designed 
for the 4:3 screen will produce oval circles on a 32:10 and vice-versa.  Indeed, 
I recall an early wysiwyg DTP suite that let you select the aspect ratio for 
that reason.

If 90 percent of your potential customers have one specific system, is it 
worth the engineering effort and overhead to try and appeal to the last 10 
percent who have different hardware?

It's perhaps not a matter of "there's an incentive to write ONLY for the 
laptop" but "there's little incentive to actively write for OTHER machines if 
these become a market-shaping force"


The screen sounds interesting though:  I'd love to be able to turn my LCD 
into a 2560x2048 black-and-white display.