[E-Lang] New Page: Partially Ordered Message Delivery

[Tyler Close]

MarkM wrote:
> Due to limited time, and to the difficulty people have had
> learning Hydro
> well enough to give me feedback,

The only person I ever heard from was MarcS. He was having trouble
constructing new containers. Consequently, I introduced the
Initializer class to simplify this. He seemed pretty happy after that.

> I expect that, as of 1.0,
> E will still be
> mainly using the current collections API ((Const, Flex) x
> (List, Map)), and
> the Walnut will be mainly documenting their use and their
> dangers.

This basically means that E will forever be using these containers as
the built-in containers. That sucks. I've explained at length why, and
I thought we'd achieved agreement.

> I'm truly sorry about this.  But you can't imagine how
> thinly stretched I
> feel lately.  (At least there's one dimension on which I'm thin!)

Currently, my main interest in E is to use it as a language to
facilitate the construction of Droplets applications. As it stands, I
write code in my own E dialect and then manually do a translation to
the corresponding Java code. The difference in file size is typically
10 to 12 times. Obviously, I need a language wrapper. To this end,
I've been putting a lot of effort into making my Droplets semantics
uniform with E's. For some of these issues, I've been trying to
persuade you to change E's semantics. The treatment of collections and
sturdy references are the two major ones. I appreciate that you've got
a lot on your plate, but if I'm going to fail in this task because of
your time constraints, then I might be better off just writing my own
compiler. This seems like a poor long term strategy, so I'd rather
not. What do you suggest I do?

Tyler

PS.

Note that Droplets with an E language wrapper would be a lot closer to
E 1.0 than the erights.org E currently is. I've already got the
persistence done in spades.