[Edu-sig] turtle coordinates: Y increase downards - as in most graphics API?

[Andre Roberge]

HI,

There seems to be a lot of misunderstanding about this issue.  Hopefully,
the following can help clear it up.

0. The turtle module includes functions named left() and right() which
result in the turtle rotating in the corresponding direction on the screen.
1. The turtle implementation includes a function, setworldcoordinates(),
which allows one to have user-defined coordinates (both for the width and
height of the world) including a choice as to which in which direction the
coordinates are increasing.   The issue has nothing to do with what choice
is "best"; any user is free to use whatever they want.
2. In the current implementation, if one choses some particular orientation
for the coordinates, the role of left() and right() are inverted and no
longer correspond to what the user sees on the screen.

The submitted fix simply ensures that, no matter what one chooses for the
world coordinates via setworldcoordinates, when right() is called in a
program, the turtle turns right and not left.

André


On Tue, Jun 19, 2018 at 2:29 PM Yoshiki Ohshima <Yoshiki.Ohshima at acm.org>
wrote:

> I also stick my nose here.
>
> The notion of "in computer graphics Y goes down" is not as true as some
> people think.  The graphics you are looking at right now on your computer
> most likely to have gone through some programs that use "Y goes up" scheme
> (such as GLSL).
>
> Only if you are talking about pixels and raster graphics, you could say
> that there is a dominant convention to make Y go down, but in with vectors,
> there is no such dominance.
>
> For end-users, why does it matter how actually pixels are stored in the
> actual memory?  Computer is medium that can simulate anything; so the
> decision should not be based on what computer does at the lower levels.
>
> Once we leave from what  computer does at low-level, there are math text
> books, which are dominantly "Y-up".  Why do they have to know two different
> conventions? Unless, of course, learning the fact that the coordinate
> system is just a convention and can be different.
>
> Also there may be a case where the particular turtle graphics
> implementation is tied to the lower level pixel representation. But is
> basically mixing different abstraction levels, and unless you want to teach
> about mixing abstraction levels, it is a bad starting point.
>
> In summary: the statement "in computer graphics Y goes down" is a wrong
> argument for turtle graphics to use Y-down coordinate system.  Then, for
> turtles and for end-users, there is an existing convention to make Y go
> down.  So that should be a better default.
>
>
>
>
> On Mon, Jun 18, 2018 at 9:54 AM, Kevin Cole <
> kevin.cole at novawebdevelopment.org> wrote:
>
>> Sticking my nose in where it don't belong. ;-) But that's never stopped
>> me before. ;-)
>>
>> First "I are not a teacher". At least, not in any formal sense of the
>> word. Second, some would say "I are not a programmer". I don't listen to
>> those people.
>>
>> In spite of the common "in computer graphics Y goes down", does that make
>> sense to new learners? I've gotten used to various coordinate systems, but
>> the first time I had to work with graph paper and plotting (and, for that
>> matter, maps) one goes "up" ("north", "forward") for Y. Then again, maybe
>> the average learner isn't quite as flexible these days, and telling them
>> after they've gotten used to a system that they have to flip everything
>> upside down and backwards isn't a great idea. I attribute my own
>> flexibility not to any innate ability but rather to learning during a time
>> when architectures OS's and languages were changing fast enough that there
>> wasn't time to become set in one's ways.
>>
>> ​[Note: Grammatical "mistakes" above are meant as weak -- very weak --
>> humor. At least, most are.]​
>> --
>> *Kevin Cole*
>> <http://novawebdevelopment.org>
>> NOVA Web Development Co-Op
>> http://novawebdevelopment.org/
>> Arlington, VA
>>
>> _______________________________________________
>> Edu-sig mailing list
>> Edu-sig at python.org
>> https://mail.python.org/mailman/listinfo/edu-sig
>>
>>
>
>
> --
> -- Yoshiki
>
> _______________________________________________
> Edu-sig mailing list
> Edu-sig at python.org
> https://mail.python.org/mailman/listinfo/edu-sig
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/edu-sig/attachments/20180619/f6b4016a/attachment.html>