🚀 go-pugleaf

In article <478@vaxb.calgary.UUCP> bonham@calgary.UUCP writes:
>However, there must be *some* reason Sun, and many other window
>system developers, chose the upper left with Y increasing downward.
>Perhaps it is because their systems were intended primarily for ease
>in programming text  displays.  Most typesetting / page description
>languages use the inverted, Y-positive-down coordinate system.

So far as adopting a convention goes, it really doesn't matter,
so you might as well provide programmers etc. with one that
follows mathematical/engineering conventions (1st-quadrant).
Whether this is appropriate for an end-user depends a lot on
what the user's background is.

HOWEVER, the order that bits are stored in a file DOES matter;
there are observable consequences, such as the order in which
an image is "painted" on the display.  It should be obvious
why text-oriented systems store the bits near the top of the
image first, and consequently tend to adopt 4th-quadrant address
conventions.  I also think that it's aesthetically more pleasing
for graphic images to roll onto the display from top to bottom,
but not everyone agrees with this.

Image storage order and coordinate conventions have only a loose
connection and can be chosen independently if desired.

Henry stated that the Blit was 4th-quadrant; I believe the Blit
was originally 1st-quadrant, changed to 4th-quadrant for the DMD
(which is NOT a Blit!).  There is evidence for this in some of
the older DMD software.

In article <7291@utzoo.UUCP> henry@utzoo.UUCP (Henry Spencer) writes:
>To quote from "Hardware/Software Trade-offs for Bitmap Graphics
>on the Blit", Pike/Locanthi/Reiser, Software-Practice&Experience Feb 1985:
>
>" ... All bitmaps share a single
>coordinate space, with (0,0) in the upper left. ... "

Ah, but in the ACM TOG, Apr 1983 Rob used 1st-quadrant.  It is worth
noting that almost all the algorithms are independent of this convention.
(Exercise:  Find the one that isn't.)

Note that I'm not arguing which is BETTER, since I've already pointed out
that that should be decided on the basis of observable consequences at
the user interface.  In my corner of the world, text-oriented stuff
should appear from left-to-right, top-down.  This argues fairly strongly
for at least STORING raster information in that order, but it has
somewhat less impact on the design of raster software/hardware internals.