Re: [boost] [random] new threefry random engine

Subject: Re: [boost] [random] new threefry random engine
From: John Salmon (john_at_[hidden])
Date: 2014-05-08 13:43:26

On Thu, May 8, 2014 at 12:27 PM, Thijs van den Berg <thijs_at_[hidden]> wrote:

>
>
> > I think my version now provides you with the 'plain old engine' concept
> > that you're looking for.  It also *allows* you to do interesting things
> > with restart(), but you can ignore those methods if you wish..
> >
> For me that’s a good compromise, but it’s up to Steven and (perhaps some
> others?) to decide what he wants with the interface. In an earlier version
> I gave public access to counter manipulators and those were questioned
> because it was non-standard interface, and so I made them private.
>
>
Yes, I'd like to hear what others think of the new
'bona-fide-engine-with-restart-extension' design.

> Your version had a nice feature that mine lacked so I adoptied it:  the
> > ability to control the output type, independently of the choice of
> > pseudo-random function.  With this addition, I can produce 32-bit output
> > from a prf that internally uses 64-bit arithmetic (or vice versa).  The
> > template is now:
> Thanks. I wanted to provide both 32 and 64 bit random numbers because 32
> bit is still used a lot. My first implementation used a fast
> reinterpret_cast<> but that was non endian invariant, and so I had to fix
> that. I think the interface and consistent behaviour is more important than
> speed for boost random, and I agree with that.
>

My implementation is intended to be endian-independent.  But as they say -
"if it hasn't been tested, it's doesn't work", and I haven't had a chance
to test on a big-endian machine, so proceed with caution..

> I think the most enjoyable way forward would be join effort into a single
> submission instead of competing ones? For that you will need to fix the
> copyright and license. What’s your view on this?  Are you doing this in
> corporate time or personal time?
>

Yes, I think it make sense to merge.  Perhaps we should start using the
recommended workflow, https://svn.boost.org/trac/boost/wiki/StartModMaint.
Especially if that would make things easier for Steven.

As you can see, my code is released under a fairly permissive license that
is very close to the BSD 3-clause license.   But it's not the Boost
Software License, and I understand why that's a problem.  I'm working on
getting permission to release it under the Boost Software License.

John

>
> _______________________________________________
> Unsubscribe & other changes:
> http://lists.boost.org/mailman/listinfo.cgi/boost
>