[Ur] questions on the demos

[Adam Chlipala]

On 04/06/2015 10:11 AM, mozert1 mozert1 wrote:
> 1. For such ur/web programming, the best is to start with the actual
> profitable thing, so I went over the demos at
> www.impredicative.com/ur/demo/
>
> And listEdit.ur is a bit too folded , And chat.ur is also somewhat
> folded. I am attaching the diff files of the revision to this mail. Or
> may be I don't get the sense of things yet.

I don't understand what you mean by "folded," so I can't comment more.

> 2. Nomenclature thing: could the [signal] monad be renamed to
> [permaction] ((dynamic) permanent action), so to parallel the
> [transaction] (transient action) monad. and the [signal] operation of
> the [signal] monad be renamed to [dynget] (dynamic get). and maybe
> push all the way and replace [<dyn source=] by [<dyn permaction=] or
> [<dyn connectfrom=] or  [<dynpermaction connectfrom=] and replace
> [<ctextbox signal=] by [<ctextbox connectto=] ...

"Signal" is established terminology from functional-reactive 
programming, so I think it remains a good name for this monad.  I 
haven't seen "permaction" before.

> 3. Precisely why inria ocsigen is never ever mentioned on the website ?

There's no claim of completeness for comparison with other frameworks, 
on the Ur/Web site, I think; so that's the reason! Ocsigen is definitely 
mentioned in the research paper on the design of Ur/Web.

> or even on the TechEmpower comparison ?

This one is even easier to answer: because no one contributed an Ocsigen 
implementation of the benchmarks yet. :)  Anyone is free to do so on GitHub.