Sorry for intervention guys but Vinnie just mentioned a thing that was in
my mind for a while:

> for example, keeping it on GitHub


I'd say that's true for the whole graph library. With all my respect to its
authors, it's definitely not a general-purpose functionality so I don't
understand why put so much effort instead of simply implementing it on
GitHub and distributing via Conan/vcpkg and improving much more frequently
than once in 3 years (at best).

On Thu, Jun 18, 2026 at 9:55 PM Phil Ratzloff via SG19 <
sg19_at_[hidden]> wrote:

> Michael has been the one to promote a graph library as part of the
> standard as a part of the functionality used in machine learning, so I'll
> let him provide the details.
>
>
> ------------------------------
> *From:* Vinnie Falco <vinnie.falco_at_[hidden]>
> *Sent:* Thursday, June 18, 2026 2:27 PM
> *To:* Phil Ratzloff <Phil.Ratzloff_at_[hidden]>
> *Cc:* sg19_at_[hidden] <sg19_at_[hidden]>
> *Subject:* Re: [isocpp-sg19] Standardizing Graph concepts
>
> *EXTERNAL*
> On Thu, Jun 18, 2026 at 11:24 AM Phil Ratzloff <Phil.Ratzloff_at_[hidden]>
> wrote:
>
> They're used to guarantee that the functions needed by an algorithm are
> correctly defined. For instance, index_adjency_list requires vertices(g),
> out_edges(g,u) and others be defined, the underlying vertices(g) range is a
> random_access_range, the underlying out_edges(g,u) is a forward_range, etc.
>
>
> Thank you so much, and I agree that graph concepts are useful. Yet
> evidence of utility is not the same as evidence of standardization
> need. What problem does standardization of graph concepts solve that can't
> be more cheaply solved by keeping it in the ecosystem (for example, keeping
> it on GitHub).
>
> Thanks
>
> --
> SG19 mailing list
> SG19_at_[hidden]
> https://lists.isocpp.org/mailman/listinfo.cgi/sg19
>

-- 
Regards,
Oleksandr Koval.