I was wondering if we couldn't use it as a starting point of our modularization activity. Some questions: * Is the file-based dependency manager what Boost is looking for its own module dependencies detection?
Well, it doesn't handle cyclic dependencies, however, being at the file level it can handle better the dependencies that have "bridge" header files. So say you include serialization, but don't use any of the date time serialization, then it will just pull serialization(instead of pulling in date time as well).
* Could an author state explicitly its dependencies on other libraries (versions)?
Yes, with versions.
* It Biicode open source?
Not yet. Their ultimate goal is to release it as open source, however, their investors are requiring 10000 users before they can release it as open source. I have heard they are trying to convince the investors to release it open source before then. Especially, since biicode can benefit from community support. Furthermore, if boost were to decide to use biicode, perhaps boost could use its weight to influence the decision to go open source.
* How its CMake autogeneration and bjam interact?
Well currently it just calls out to bjam from cmake I believe. You can see the source code for it here: https://github.com/Manu343726/boost-biicode Ideally, I think it would be for boost to switch to cmake as their is more community support for the tool, and it can better integrate into other tools as well. Paul -- View this message in context: http://boost.2283326.n4.nabble.com/Re-Boost-interest-Biicode-dependendy-mana... Sent from the Boost - Dev mailing list archive at Nabble.com.