Re: The naming question (Postgres vs PostgreSQL)

-----BEGIN PGP SIGNED MESSAGE-----
Hash: RIPEMD160

Apologies, Andrew, but I think this is a very important topic, and
one that should be addressed.

Josh Berkus wrote:
> My perspective on the naming question is that if we wanted to do it, we needed
> to do it for 8.0 which was when PostgreSQL became much higher profile. We
> did seriously evaluate it at that time, but the amount of effort involved
> would have held up the release and we decided not to.

That's not my recollection. I recall battling to make 7.5 into 8.0, but there was
no real debate about a name change at that time. Since then, however, I've talked to
lots of people about changing the name and found no real objection, only
support, including from most of Core. Even Tom Lane admits it was a mistake:
http://tinyurl.com/2v4fdl

> It is still the case that changing the name would involve a lot of tedious
> effort and hunting down references.

Nobody says we have to go crazy and hunt down and change every single reference.
"Postgres" would simply be the preferred and official name, while "PostgreSQL"
would be the older, alternative spelling.

> Overall, I'd say that the advantages of changing the name don't balance the
> effort involved in doing so. That's a very qualitative judgement though, and
> if there was overwhelming support in the community, *including* several
> volunteers to help do the work, I'd certainly re-consider my stance.

I think the support in the community is there, and I don't see what all the
effort involved is. We simply make it the official name, and migrate things
over as we can, and don't sweat the older stuff. Count me as one volunteer,
at any rate.

> Also note that I was chatting with Kaj Arno at OSCON, and he was bitching
> about people mispronouncing "MySQL". So don't assume that shortening our
> name would necessarily help ...

I'm pretty sure that "Postgres" would have little to none of the pronunciation
problems that PostgreSQL has. Also, MySQL is a compound, while PostgreSQL
is a portmanteau, from which springs many of our problems.

Ron Peterson notes:
> It's google'able. Think of all of the documentation out there which
> refers to PostgreSQL. When people start searching for 'postgres'
> they'll get virtually nothing.

Sure they do: 5.7 million results, including postgresql.org as the first hit.
"Postgres" is the name the great majority of people use anyway. It's the name
of the OS user and the default database we create alongside template1. And the
nice thing is that it is still a unique word, so it won't get swallowed up in
lots of other words (e.g. "Oracle" and "Firebird")

> Why 'Postgres'? Any marketing effort this large should start from
> scratch, and consider all possibilities, not just default to something
> barely different for the sake of easing the terrible pronunciation
> challenge.

It's not so much a marketing effort as realizing that our current name
is unwieldy, prone to errors in spelling, translation, and pronunciation,
and is quickly changed to "Postgres" by most anyway. Being "barely different"
is a strength, not a weakness. One thing is for sure is that Postgres is really
the only option here, we're not going to rebrand the entire project like
firefox did, if that's your implied suggestion. :)

> Maybe some people don't like PostgreSQL, but why assume everyone likes
> 'Postgres' better. I don't.

What's your alternative?

-----BEGIN PGP SIGNATURE-----

iD8DBQFG00PevJuQZxSWSsgRA8aTAJsFXIpwajz5kqg3iTf0LNPjm7AtuQCggEJF
oYO2hW3I5lQNYkrlPWG3VNc=
=W+Xy
-----END PGP SIGNATURE-----