> Your are blowing this problem out of proportions.

What problem? I was just responding to Ken's assertion that
"the development model of POV-Ray was pretty much at the whim
of the developers", when that is not actually true. I'm not making any
connection between this particular subject and the scanline stuff;
I'm just pondering it independantly as an interesting socioeconomic
phenomenon in its own right.

Roughly speaking, whoever supports a file format best gains the
majority of the user base and relegates other supporters to the sidelines,
or even displaces them entirely. But the users decide what constitutes
"best", so whatever feeling of control developers have is illusory.

The only point I can make relating that and the scanline feature
is that *if* the majority of users decided scanlining was desirable,
then it wouldn't matter what the developers thought; the feature
would inherently find its way in. OTOH, sometimes users don't know
what they want, and they realize only in retrospect that they badly needed
a certain feature (it explains those "How did I ever live without X?"
sayings).
Software is prone to that phenomenon because myths build up easily about
software, and the only way to change thinking is to just develop working
code and demo it. So I risk failure to demo scanlining just in case
it can only be desirable retroactively.

The point about the different unofficial POVs is interesting too,
actually. They demonstrate what happens when a large but minor part
(say 30% or 40%) of the user base prefers a feature. A parallel fork
winds up being separately maintained. A scanlining feature could
end up like that too.

As for my narrow lines of text, I find reading long lines difficult
(it's why newspaper columns are narrow -- imagine if newspaper articles
ran their text all the way across a page. It's also why all the
main news sites like Salon, ABC, MSNBC, etc. use columns). I'd rather
scroll (I just hit the down arrow key) than keep losing track of
which line is next.

Ray

Post a reply to this message