From: Andrey Semashev (andrey.semashev_at_[hidden])
Date: 2024-03-28 14:08:40
On 3/28/24 17:01, Vinnie Falco via Boost wrote:
> On Wed, Mar 27, 2024 at 2:34 PM Robert Ramey via Boost <
> boost_at_[hidden]> wrote:
>
>> I would like to see suggestions as to how we could get usage statistics
>> on boost libraries.
>>
>
> Yes, this is an important area of analysis for which no perfect solution
> exists. However, researchers at The C++ Alliance have developed an
> experimental technique which offers hope for newly submitted libraries. It
> works like this:
>
> * We maintain an inventory of unique names that are not found on the web
> * New libraries will be assigned a unique name from our inventory
> * A C++ Alliance cloud application will trawl the web for references to the
> name
>
> For example, the next library which is accepted may be given the name
> Boost.Zissifrak, a term which nets zero search results using Google. Using
> this method we can indirectly measure the popularity of the library over
> time by observing the evolution of the number of matching search results.
Are you being serious? You might as well generate UUIDs for library names.
Please don't. Preferably, a library name should reflect its purpose,
and, if possible, use recognizable terminology in its domain.