Goal: Portable C++ Library call from Herb Sutter

[Tavi]

Herb Sutter made yesterday and the end of his keynote of the GoingNative event, a very cool call, about what C++ really needs: more standardized (or defacto) libraries (http://channel9.msdn.com/Events/GoingNa ... and-Beyond)

Portable C++ Library (PCL)
Goals:
- Large set of useful and current libraries
- Available on all major platforms
- Shipped with and supported by C++ implementations
- And composable, using consistent types.

- Minimum: de facto availability as part of all major compiler products
- Ideal: De jure inclusion in Standard C++

Poco Libraries was specified there few times. Would be really cool if Poco would propose and have success in this direction!

[rmoffer]

Thanks for the pointer - it was an interesting talk.

What I took away was that its expensive (Boost, Google, Microsoft and Adobe all have a vested interest in getting their libraries accepted), which means if we want Poco to be part of that standardization process, then we (the Poco Community) need to build a plan for resources (money, time) to be spent on it.

We shouldn't rely on Guenther and team to bear the costs themselves since our whole community will benefit.

[guenter]

It will definitely need a community effort to bring POCO into the standardization process. I'm not sure yet how to approach this. It will definitely require significant work to write proposals and defend them at WG21 meetings. Any input on this is very welcome.

[Tavi]

I think standardization is just a distant goal. The model would be probably similar with what boost libraries achieved with tr1. The portable c++ library should evolve in a coordinated way (by Herb Sutter & Co), because of the composability and type consistency.

Here is also a relevant quote from the last day panel of the conference:

Question:

Advice for people who are willing to help to close the libraries gap and put food on the table?

Herb Sutter:

First, please email. Let me know. If you are a zlib, boost, adobe, apache, poco, apple, and so forth developer who has access to in-house libraries that could be contributed, or just want to help, please let me know because that's what we try to put together. We have a spreadsheet with a list of the features we want, only half of them are filled in with existing practice and smaller that that with people who are able to contribute work on them. So if you have time, it is not expected to be a full time job, but it does require help, and that's what we need most, expert help.

The 'food on the table' part was addressed: it is volunteer work (with a funny comment from Bjarne regarding this )

If Poco has on the TODO list an overhaul in order to better use and not overlap with new C++11 features, this may be the way: the changes could be coordinated as part of this larger effort.

[alex]

It will definitely need a community effort to bring POCO into the standardization process. I'm not sure yet how to approach this. It will definitely require significant work to write proposals and defend them at WG21 meetings. Any input on this is very welcome.

Tough effort, no doubt. First couple of things that come in mind are: naming convention and coupling. All POCO libraries depend on Foundation which essentially means - Foundation should go into the standard; given that we did not reinvent the wheel (like some other candidates ), this may not be as far-fetched as one would think, but we'd have to put a significant effort into C++11. The naming convention still remains as a concern - this would essentially mean having a parallel development branch for standard candidate. I'll touch base with Herb and see whether this makes sense at all. The shameless plugs in the past have yielded some results, after all . Let's se how far this can go. I'll keep you posted.

[alex]

I'll keep you posted.

So, I'll keep my word: we've got a reply and it is a positive one - Poco is a serious candidate for standardization, Microsoft is also interested (not clear yet to what extent). We will need volunteers. This effort is in very, very early stage; if you are willing to help, please make it known here.

[alex]

Here's the most recent news on this topic.