On 24 Nov 2014 at 21:53, Hartmut Kaiser wrote:
Many things are a lot easier with C++11/14, we all know that. But what is the point of creating a list of libraries _requiring_ a C++11 compiler? Code which does not use C++11/14 features is still compilable with modern compilers just fine. Also, it's still possible to use older code together with newer libraries. The only purpose I would see is to tell people which libraries they can't use if they are stuck with older compilers for some reason...
Sorry I miss the point of this discussion and will shut up.
You may remember my C++ Now 2014 talk and white paper was exceedingly gloomy about Boost - lots of empirical trend graphs pointing at two year long trend declines and a fair bit of hyperbole from me about poisonous anti-change anti-innovation anti-process cultures etc. Six months later I'm feeling that something is changing, but I don't know what exactly or how profound it might be. In thinking about measurable proxies of change I could go examine, I can see that studying C++ 11 only Boost libraries could be considered a proxy for "green shoots" of new growth in Boost. Hence my hypothesis, and investigating that might yield an answer to whether Boost is going to resurge in a big way or something completely else. I appreciate it's a bit abstract. Hopefully when you're evaluating the C++ Now submissions it'll seem worthy to you then. Niall -- ned Productions Limited Consulting http://www.nedproductions.biz/ http://ie.linkedin.com/in/nialldouglas/