Use Advanced Search to search the entire archive.
Re: On Quantity - Measurement relationship
- From: Werner Keil <
>
- To: "
" <
>
- Subject: Re: On Quantity - Measurement relationship
- Date: Sat, 1 Nov 2014 17:19:52 +0100
The most important argument on the ME side is space.
declaring dozens or hundreds of concrete classes (even if they got
significant math from a base class like current BaseQuantity) is a
nightmare there.
People would likely end up with "Wildcard" types given the current approach
of Quantity<?> for unknown types would no longer exist
(the only one who'd be extremely happy might be Otavio[?]) so instead of
Quantity<?> developers in their pain and need to save space would declare
something like AnyType implements Any (sort of like Scala does btw. in its
own typesystem[?])
Werner
On Sat, Nov 1, 2014 at 5:12 PM, Martin Desruisseaux <
>
wrote:
>
Le 02/11/14 01:06, Werner Keil a écrit :
>
> You have to ask Antoine, a Spec Lead of CDI 2 who made the change.
>
> (he's actually in Paris most of the time;-)
>
I'm in Japan right now (so I should already be sleeping). But anyway, he
>
may have good reason for CDI, but that doesn't means that those reasons
>
apply to us too. For now Proxy is a technical solution that works - with
>
performance cost, but we can mitigate it with concrete classes for some
>
types. It could happen that analysis of the technical arguments against
>
Proxy shows that the inconvenient are important enough for disqualifying
>
Proxy technology on JavaSE. But we can not said that without seeing the
>
arguments.
>
>
Martin
>
>
Attachment:
329.gif
Description: GIF image
Attachment:
347.gif
Description: GIF image