[J3] [EXTERNAL] Private enumerators

[Ondřej Čertík]

Dear Malcolm,

On Sun, Oct 27, 2019, at 7:44 PM, Malcolm Cohen via J3 wrote:
> > Can you please explain why you strongly disagree with this claim?
> 
> 
> There are two parts to that claim: that extensible enumeration types 
> make no sense, and that private enumerators make no sense. As it 
> happens I disagree with both, but you ask only about the second part.
> 
> 
> > his argument is convincing
> 
> 
> I did not find it so.
> 
> 
> > I would like to learn what the arguments are for private enumerators.
> 
> 
> And I don’t want to get bogged down in technical arguments right after 
> the meeting.
> 
> 
> IMO the version of enumeration types that the meeting decided on is 
> completely unacceptable, and grossly violates our charge from WG5. 
> Others obviously have different opinions. But it is so bad (again IMO), 
> that I would certainly not agree to the inclusion of the current form 
> in the standard.
> 
> 
> But I do NOT want to get down into the nitty gritty of tricky technical 
> arguments now! We are not about to revisit any of the (IMO very 
> suboptimal) decisions or otherwise, that is for the next meeting. Where 
> I would want to have time to fully prepare my arguments, time that I do 
> not have now. Hashing/rehashing/rearguing stuff from the meeting over 
> the email list now is, IMNSHO, mostly a waste of time, as those 
> arguments will need to be made again, and made more clearly, at the 
> next meeting.
> 
> 
> Seriously, now is not the time to do this. And moreover, I have no time 
> available to do this beyond indicating where some of my disagreements 
> lie.
> 
> 
> And now I’m out of time even for writing emails.

Thank you for explaining your perspective. I think I better understand now why you don't want to discuss technical details over email.

Unfortunately, that doesn't work for me personally --- in order for me to get up to speed and prepare better arguments at our next meeting in February, I absolutely have to be discussing these technical details over email or at GitHub before we meet with others. Just thinking on my own without discussing with anyone does not work for me.

So I am inviting anybody who wants to continue the discussion to do so here:

https://github.com/j3-fortran/fortran_proposals/issues/11
https://github.com/j3-fortran/fortran_proposals/issues/46

That arrangement might work for everybody: people like me who want to discuss the details can do so, and others who do not want to can ignore the discussion. And this J3 mailinglist will not be "spammed" with low level details.

Ondrej