On 11/5/23 4:39 PM, Peter Dimov via Boost wrote:
If we at Boost don't test our own libraries, who is supposed to?
Why do I have to test and chase down bugs in libraries I don't even use? That you have to ask the question illustrates the problem. Testing as it's currently constituted isn't working. It needs to be rethought. Maybe we should have another thread for this. FWIW - I did propose and implement and alternative for the Boost Library Incubator - but I failed to convince enough people to consider the issue. Perhaps the time is ripe for serious discussion. BTW - this is only one of several issues based in common C++ software practices and misconceptions that are hindering progress in Boost. Again - another thread. And of course they apply to C++ and the standards process in general. But they're more obvious in Boost because it's much higher visibility than other software. Maybe we agree for now that all software released by Boost should be tested? And that this is not being done now? Robert Ramey
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost