WD-DALI-1.1 polygon winding direction

[Arnold Rots]

It all depends on what expression is used for calculating the area.
The sum of determinants of successive vertex vectors requires
flipping the sign of the result for left-handed coordinates; but one
needs to be careful about what to do with cos(dec) and it will not
provide exact precision.
On the other hand, using the sum of interior vertex angles gives
a correct result independent of handedness.

Yes, this will not change.

Cheers,

  - Arnold

-------------------------------------------------------------------------------------------------------------
Arnold H. Rots                                          Chandra X-ray
Science Center
Smithsonian Astrophysical Observatory                   tel:  +1 617 496
7701
60 Garden Street, MS 67                                      fax:  +1 617
495 7356
Cambridge, MA 02138
arots at cfa.harvard.edu
USA
http://hea-www.harvard.edu/~arots/
--------------------------------------------------------------------------------------------------------------


On Mon, Sep 12, 2016 at 1:30 PM, Patrick Dowler <pdowler.cadc at gmail.com>
wrote:

> In STC you standardise the "on-the-left" rather than the sign of the area.
> OK.
>
> And just for clariy, the common way to compute polygon area via
> triangulation gives a negative value for left-handed coordinate
> systems (eg celestial) and a positive value for right-handed
> coordinate systems?  Is this something to mention explicitly? I note
> you said "may depend"... what does that mean exactly?
>
> Thanks for the paragraph; will add it to the next revision. I assume
> that this is consistent between STC-1.x and STC-2 and I can make the
> reference just to the latest STC...
>
> Pat
>
>
>
>
> On 12 September 2016 at 10:13, Arnold Rots <arots at cfa.harvard.edu> wrote:
> > No. There is no telescope that looks at the sky from the outside; it's
> CCW
> > regardless of handedness.
> >
> > The following may not be brief, but I think it is clear and complete.
> >
> >   - Arnold
> >
> > Polygons are defined by traversing their vertices in counter-clockwise
> > direction.
> > In other words, the inside of the polygon is always on the left side
> during
> > such
> > a traversal, regardless of whether the coordinate system is right- or
> > left-handed.
> > Note that, consequently, expressions for the area of the polygon and
> whether
> > a
> > particular location is contained within the polygon may depend on the
> > handedness
> > of the coordinate system; celestial coordinate systems are left-handed.
> > The sides of a polygon are considered part of the inside.
> > Also, polygon sides in spherical coordinate systems are great circle
> > segments
> > and segments have to be shorter than 180 d; if longer sides are called
> for,
> > an intermediate vertex needs to be added.
> >
> > ------------------------------------------------------------
> -------------------------------------------------
> > Arnold H. Rots                                          Chandra X-ray
> > Science Center
> > Smithsonian Astrophysical Observatory                   tel:  +1 617 496
> > 7701
> > 60 Garden Street, MS 67                                      fax:  +1 617
> > 495 7356
> > Cambridge, MA 02138
> > arots at cfa.harvard.edu
> > USA
> > http://hea-www.harvard.edu/~arots/
> > ------------------------------------------------------------
> --------------------------------------------------
> >
> >
> > On Mon, Sep 12, 2016 at 12:09 PM, Patrick Dowler <pdowler.cadc at gmail.com
> >
> > wrote:
> >>
> >> So just to be clear and minimalistic, I think the only correction to
> >> the text in rev 3530 is that the view if from outside looking toward
> >> the origin. Is that correct? I will also add a reference to STC but I
> >> think this change would make the DALI text minimal and consistent.
> >>
> >> Pat
> >>
> >> On 9 September 2016 at 01:02, Marco Molinaro <molinaro at oats.inaf.it>
> >> wrote:
> >> > Hi Pat, Arnold, all,
> >> >
> >> > 2016-09-08 18:56 GMT+02:00 Patrick Dowler <pdowler.cadc at gmail.com>:
> >> >> So, should I change the wording to say "when viewed from outside the
> >> >> unit sphere" and add STC reference? I want to say the minimum and
> >> >> leave the definitive specification in the hands of STC, but if we
> make
> >> >> readers go look it up in STC they will be annoyed.
> >> >
> >> > I agree on referencing STC (even if it's not trivial because it would
> be
> >> > nice
> >> > to have a pointer to 2.0 or general STC), but summarizing it in DALI.
> >> > I'm ok with the wording, is the group agrees (warning: if no one
> protest
> >> > means agreement).
> >> >
> >> > Maybe, Arnold, can you suggest something
> >> > short and clear to put in DALI?
> >> > The reference text you replied is probably bullet-proof but too long
> for
> >> > the DALI spec (at least that's my opinion).
> >> >
> >> > Cheers,
> >> >      Marco
> >> >
> >> >>
> >> >> Pat
> >> >>
> >> >> On 8 September 2016 at 01:00, Felix Stoehr <fstoehr at eso.org> wrote:
> >> >>> Dear all,
> >> >>>
> >> >>> after some side-discussions with Marc I believe the following is
> true:
> >> >>>
> >> >>> - STC-S declares clockwise and counterclockwise with respect to the
> >> >>> celestial coordinate frame (north up, east to the left as seen from
> >> >>> earth)
> >> >>>
> >> >>> - this is consistent with the footprintfineder.py output and thus
> all
> >> >>> the MAST/ALMA/ESA spectra
> >> >>>
> >> >>> - area calculations will be correct in this definition and
> coordinate
> >> >>> frame
> >> >>>
> >> >>> - it is a bit counter-intuitive, because it means that
> anticklockwise
> >> >>> polygons in that definition are clockwise if you look up on the sky
> >> >>> from
> >> >>> earth.
> >> >>>
> >> >>> It might be worthwile to declare in WD-DALI-1.1 the coordinate
> system
> >> >>> in
> >> >>> which the clockwise/anticlockwise are defined, i.e. in which system
> to
> >> >>> "hold the clock".
> >> >>>
> >> >>> Best regards,
> >> >>>
> >> >>> Felix
> >> >>>
> >> >>
> >> >>
> >> >>
> >> >> --
> >> >> Patrick Dowler
> >> >> Canadian Astronomy Data Centre
> >> >> Victoria, BC, Canada
> >>
> >>
> >>
> >> --
> >> Patrick Dowler
> >> Canadian Astronomy Data Centre
> >> Victoria, BC, Canada
> >
> >
>
>
>
> --
> Patrick Dowler
> Canadian Astronomy Data Centre
> Victoria, BC, Canada
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ivoa.net/pipermail/dal/attachments/20160912/9852da82/attachment.html>