<p dir="ltr">I asked a very simple question, which was not about primitives, was not about the need of primites or anything like tgat. you wrote a nice email with lots of details, but you didn't reply that very simple question:</p>
<p dir="ltr">In principle, should it be possible and useful to encode each possible *calligraphic* variants separately, or not?<br>
Is that one of the things Unicode should be about or not?</p>
<p dir="ltr">And this is not a question I am asking to you in particular, but to everyone, egyptologist and non egyptologists.</p>
<p dir="ltr">And I think you should be interested in the answers as well</p>
<div class="gmail_extra"><br><div class="gmail_quote">On Jul 30, 2016 1:30 PM, "Mark-Jan Nederhof" <<a href="mailto:mn31@st-andrews.ac.uk">mn31@st-andrews.ac.uk</a>> wrote:<br type="attribution"><blockquote class="quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">There are motivations for the particular choice of primitives we have now,<br>
which was the result of half a year of concentrated study by the folks<br>
from TLA, Ramses and St Andrews, with constructive criticism from some<br>
participants in Cambridge, and some further ideas discussed on this email list.<br>
<br>
Is the current collection of primitives and the way they are to be used set in<br>
stone? No.<br>
<br>
Do we have a single easy one-line answer why we need this set of primitives<br>
and no other. No.<br>
<br>
Could the theoretical and empirical support in favour of our set of primitives,<br>
or any other, ever be exhaustively explored? No.<br>
<br>
Does that mean our set of primitives is completely arbitrary and baseless?<br>
If you think that you have not read the text of our proposal and have not<br>
followed the discussions on the list.<br>
<br>
For the rest you are contributing no new thoughts. Life is too short to<br>
have to repeat the same discussions over and over and over again.<br>
I hope you excuse me but this is for me my last response on this thread.<br>
<font color="#888888"><br>
Mark-Jan<br>
</font><div class="elided-text"><br>
On Saturday 30 Jul 2016 12:46:12 Marwan Kilani wrote:<br>
> We are back to square one because you (and I say you as you are suggesting<br>
> the proposal) have not approached one of the basic questions yet:<br>
><br>
> what is a spatially relevant feature that need to be encoded/displaed, and<br>
> what is not?<br>
><br>
> stéphane gave some scattered example, but there is no definition yet, no<br>
> general coherent frame about *how egyptian work spatially* and therefore<br>
> about what is meaningful and what is not has been suggested. Only random<br>
> examples taken here and there (some even borderline cases, like your Dd=f,<br>
> where you are essentially discussing and issue that *does not exist*<br>
> because the difference between the to is not meaningful in any way, as far<br>
> as i know) have been pointed out. Or at least this is what I have seen.<br>
><br>
> What you describe is a calligraphic variant: it is important to be able to<br>
> represent calligraphic variants or not?<br>
><br>
> This is an *important* question that *has to be tackled with* because it<br>
> has consequences on all the the suggestive steps.<br>
><br>
> just to say: if you want to be able to display calligraphic variants, then<br>
> you have to include corresponding precomposed glyphs in your font (because<br>
> control characters dont make up anything on their own), with all the<br>
> problems this will generate (and beside the fact that unicode is not for<br>
> calligraphic variants)<br>
> if instead you don't want to do that, then you have to make explicit that<br>
> calligraphic variants *are not* meant to be handled with unicode, and this<br>
> can have consequences on the features you need in your encoding system, and<br>
> also brings up the next questions: if calligraphic variants are not meant<br>
> to be encoded then 1) what is a calligraphic variant and 2) what else could<br>
> be left out?.<br>
><br>
> This is an important question, as important as understanding what is a real<br>
> glyph that should have an independent unicode slot, and what instead is<br>
> just a graphic variant.<br>
><br>
> It is the same principle that make people at the workshop arguing against<br>
> encoding into the unicode set signs "at random": you need first to know<br>
> what is a really meaningful variant, and what instead is an allograph to<br>
> have a manful set of hieroglyphs.<br>
><br>
> With spatial distribution should be the same: what is meaningful and what<br>
> is not.<br>
><br>
> What is the general theoretical frame we are working in? If any?<br>
><br>
> Essentially, what (in absolute terms, in concepts, not in random examples)<br>
> are we trying to encode and what are we not trying to encode?<br>
> And possibly why (meaning-wise, quantity-wise etc?)<br>
><br>
> And this stands true whether you want to use ligatures, control characters,<br>
> or whatever else.<br>
><br>
><br>
> Marwan<br>
><br>
><br>
><br>
><br>
><br>
> On Sat, Jul 30, 2016 at 12:18 PM, Mark-Jan Nederhof <<a href="mailto:mn31@st-andrews.ac.uk">mn31@st-andrews.ac.uk</a>><br>
> wrote:<br>
><br>
> > On Saturday 30 Jul 2016 10:52:00 Marwan Kilani wrote:<br>
> > > "For example take Dd=f, with cobra + hand + viper. Sometimes you see<br>
> > > hand:viper clearly entirely<br>
> > > inside the bounding box of the cobra. Sometimes you see the hand in the<br>
> > > cobra, while the viper<br>
> > > is entirely below, with the tail of the viper extending below the tip of<br>
> > > the tail of the cobra. Sometimes,<br>
> > > the head of the viper is inside the bounding box of the cobra, while the<br>
> > > tail of the viper extends<br>
> > > below the tail of the cobra."<br>
> > ><br>
> > > Sorry, to go back to the same thing, but.. what is the practical need of<br>
> > > encoding such a distinction?<br>
> > ><br>
> > > having the inside or outside the cobra has (I guess I should say "as far<br>
> > as<br>
> > > I know") no meaning and no importance whatsoever. There is no linguistic,<br>
> > > no semantic, nothing..<br>
> > ><br>
> > > it is just a graphical variant, a calligraphic choice of the scribe.<br>
> > ><br>
> > > Unicode should be about standardized transcriptions, not about<br>
> > paleographic<br>
> > > details. The important thing here is that the "f" comes after "D+d".<br>
> > That's<br>
> > > all.<br>
> > > Why should we encode such calligraphic variants in the first place?<br>
> > ><br>
> > > What is the utility of that?<br>
> ><br>
> > I must have been spectacularly unclear.<br>
> ><br>
> > If one accepts that an encoding should contain primitives that describe<br>
> > the approximate spatial arrangement of signs, then it is inevitable there<br>
> > will be<br>
> > boundary cases where it is unclear how to encode some text. That holds<br>
> > for a system with 20 control characters as well as for a system with 3<br>
> > control<br>
> > characters.<br>
> ><br>
> > Of course, if one does not accept that an encoding should contain<br>
> > primitives<br>
> > that describe the spatial arrangement of signs, then we're back to square<br>
> > one. I suggest you reread Stéphane's messages on the subject, who<br>
> > motivated time and time again why at least one prominent potential user<br>
> > community most certainly needs to have access to the graphical realisation<br>
> > of a text. Stéphane explained this with extraordinary detail and above all<br>
> > patience. I don't see any need to restart this. There are diminishing<br>
> > returns<br>
> > for repeating the same discussions ad infinitum.<br>
> ><br>
> > Mark-Jan<br>
> ><br>
> ><br>
> ><br>
> ><br>
> ><br>
> ><br>
> ><br>
> > _______________________________________________<br>
> > Egyptian mailing list<br>
> > <a href="mailto:Egyptian@evertype.com">Egyptian@evertype.com</a><br>
> > <a href="http://evertype.com/mailman/listinfo/egyptian_evertype.com" rel="noreferrer" target="_blank">http://evertype.com/mailman/listinfo/egyptian_evertype.com</a><br>
> ><br>
<br>
<br>
_______________________________________________<br>
Egyptian mailing list<br>
<a href="mailto:Egyptian@evertype.com">Egyptian@evertype.com</a><br>
<a href="http://evertype.com/mailman/listinfo/egyptian_evertype.com" rel="noreferrer" target="_blank">http://evertype.com/mailman/listinfo/egyptian_evertype.com</a><br>
</div></blockquote></div><br></div>