Re: [idn] copy & paste

["James Seng" <jseng@pobox.org.sg>]

Hi Eric,

> You know, I never understood such a position. It is ridiculous that we can
> bring an elephant into a room and then say it's not our problem. Of course
> it's in-scope, at least as an open concern. We don't have to solve it but
> we shouldn't try to ignore it.

The bigger scope of the clipboard, copy & paste of strings, images, sounds
etc are out of scope. but the smaller one involving how domain names is copy
& paste may be of interest to look at. this is what I am doing here.

> Does anybody actually expect this to happen? Any clipboard that tries to
> transliterate on-the-fly is doomed to failure. I mean, you couldn't even
> copy-n-paste an example string with such a clipboard.

I shall assumed the clipboard is dumb, ie, accept what the application gives
it and just hand it over to the other one for this examine.

> > Case 1: resolving works, display works.
> > Case 2: resolving fails, display works on A1 but not A2.
> > Case 3: resolving works, display works on A2 but not A1.
> > Case 4: resolving works, display fail for both A1 & A2.
> >
> > Is there anything else I missed?
>
> Is this an exercise or is this to be used as proof of a point?

Just an excercise. Lets make sure we cover all the cases first.

Let me do a bit of tweaking.

           Resolution    Display
           A1    A2      A1    A2
Case 1 :  okay  okay    okay  okay
Case 2 :  okay    ?     okay    ?
Case 3 :  okay  okay    fail  okay
Case 4 :  okay  okay    fail  fail

? I made a correction here: Case 2, A2 is likely to fail in the resolution
altho A2 may be or may not able to display the string correctly. We cannot
predict what an non-IDNA applications going to do to the domain names. I
replace it with a "?"

Have we cover all our bases?

> I can think of several other ways to analyze this if that's
> what you're looking for. For example, Case 4 represents 100% of the
> current time data-point, will reach 0% at or near infinity. Cases
> 2 and 3 will be dominant for some period of time between now and
>infinity. One good question would be "when does case 1 become
> majority?"

I agree.

Although one can not predict how fast applications support IDN, i dont
expect to see 100% Case 1 happened in overnight and very likely going to
take infinity time as yo said. Case 2 to 4 is going tobe dominate.

> Also, case 2 is where the mandatory transliteration problem visibly
> occurs, although case 3 probably has some interesting issues as well.

Can you eleborate? Have I missed something in my analysis?

-James Seng