On Sat, Feb 18, 2012 at 4:13 PM, Rene Rivera <grafikrobot@gmail.com> wrote:

On 2/17/2012 11:32 AM, Beman Dawes wrote:

...

The C++ committee met last week. There was lot's of discussion and plans to markedly increase the size of the standard library. Boost has an important part to play; Bjarne Stroustrup hoped for "doubling the output of Boost."

Is that output meant to be accomplished by doubling the percentage of current Boost libraries that get into the TRs? Or doubling the number of libraries in Boost? I ask because the former might be possible, although not easy.

The context was a discussion of an effort by Herb Sutter to get some large corporations to start contributing libraries. Because the discussion was focused on that effort, someone erroneously jumped to the conclusion that the committee no long cared about other source. Bjarne jumped in to say we still very much cared about traditional sources, mentioning Boost and one or two others. I doubt he had anything specific in mind.

While the latter seems almost impossible given the low throughput reviews (not because of the process but the lack of managers).

Yeah, we need to fix that problem. But that's a different discussion.

...

Here are some of the process changes:

* Library proposals will be taken on as "work Items". A decision as to whether a work item ends up in a technical report, in the standard itself, or even becomes a stand-alone international standard, will be deferred until technical work is complete (I.E. full standardese complete, polished, and ready to ship).

* Domain specific "Study Groups" will replace the LWG's current sub-groups. Study groups have official ISO standing, so can get more work done between meetings than the old sub-groups, which were unofficial.

Trying to understand this new structure.. Does it mean that a Study Group would handle multiple proposals within a domain? Or making a Study Group per proposal?

A study group per domain, although some domains may only have one proposal active a given time.

I don't see a "Data Structures" group. Was that considered? Will it be considered?

So far, groups are only being formed if they have specific proposals to work on, and the LWG or EWG wants to pass the work off to a sub-Group. So nothing on Data Structures, yet. Man and woman power is an issue too. There were new faces and a lot of energy at the meeting; let's hope that continues and wasn't just because the meeting was in Hawaii. Thanks, --Beman

Cory Nelson wrote:

As someone who does not have the means to attend these meetings, I hope someone will be able to champion variant, optional, and intrusive.

The "variant" functionality is something I have often seen in "real" code, even so it never used boost::variant. Most often it was homegrown, but sometimes also QVariant (from qt). But QVariant is actually more related to boost::any than boost::variant. So what about boost::any? I have seen boost::optional in "real" code, but that code wasn't very convincing (to say the least). But I see that it has real use cases for class members, containers, and for optionally returning a value. The free "get_pointer" function is especially useful for efficiency and generic code (assuming it is specialized for all "pointer like" objects, including real pointers), but I don't fully understand why the equivalent member function "get_ptr" has a different name. I though more than once about actually using "intrusive" in my code, and I guess it would really have a positive impact on performance at the places where I considered using it. What is also intriguing about "intrusive" is the huge collection of containers offered with it. The only drawback is that I could imagine even more container types that would make sense for "intrusive", but probably already the existing collection of container types will be "too large" for the standard. Regards, Thomas