(j3.2006) (SC22WG5.4394) [ukfortran] Result of informal ballot on draft TR

[N.M. Maclaren]

On Dec 8 2010, Aleksandar Donev wrote:
>
>> The main one is 10-235r1, which I do NOT feel has been properly
>> considered, and is as problematic as the TYPE(*) issue and even more
>> pervasive.  Unfortunately, explaining why (beyond what is in 10-235r1,
>> which clearly wasn't adequate) is non-trivial.
>
>I do not find an r1 of this paper on the J3 server?

>I took a quick look at 235. It has some good points, but I must say that 
>any paper that begins with "we need to start from scratch", i.e., undo 
>everything else others have done, and implement your favorite approach, 
>does not get sympathy from me regardless of technical content (which has 
>some good points but also a lot of exaggerations).

I suggest that you look at it with a less prejudiced eye.  Not merely
does it NOT say or propose that, it starts by explaining why I was forced
into that conclusion.

As I posted before, I started by trying to draft wording to constrain
use enough to ensure both implementability and usability.  After a day's
work, I gave up.  I couldn't.  As I have also posted before, now there is
a complete draft, I shall try to produce actual code and demonstrate how
nasty the problem is.

>That said, just vote No, if that is what you want. In the end, 
>everyone's work will be wasted, including yours.

I do not think that you have realised the seriousness of the semantic
inconsistencies.

Regards,
Nick Maclaren.