Re: Change the name

From: Christian Voelker <C(dot)Voelker(at)gmx(dot)net>
To: pgsql-advocacy(at)postgresql(dot)org
Subject: Re: Change the name
Date: 2007-09-05 14:32:05
Message-ID: F1ACDEEC-8475-4CC8-A003-BA1934C98EAB@gmx.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-advocacy pgsql-de-allgemein

Hello,

new to this list but long term postgres user.

David Fetter mentioned the discussion in his weekly update that gets
posted to the german list and he called it civilized. That made me
reading certainly some hundred posts from the archive, maybe half
of the thread and finally joining the list to participate in this
discussion. I came to the conclusion that it is not friendly at all.

Can there really be people who search for misspelled semicolons in
their source files as a day job, blaming others to make a lot of hot
air out of nothing when arguing about pronounciation of the project
name? I can hardly believe.

Pronounciation might be only one single argument compared to so
many arguments against a name change. And it is a soft point com-
pared to a missing semicolon that inevitably stops your query.
But that doesnt make it be a weak argument. For me, there can
hardly be anything more striking.

Working code is of utmost importance, but all the bruhaha around
doesnt count less I believe. And if marketing oriented people have
a strong opinion about that issue, constantly neglecting the im-
portance of their input feels rather destructive. Some day, they
will leave you coders alone and the project might decline.

I personally like to pronounce the RDBMS as Postgres-Qu-El but am
too lazy to write the whole thing, so actually I do just the opposite
of what seems to be common use. But when reminded of the problems with
the two names I can recall the hurdles that they exposed in the be-
ginning. And that was long before I asked anybody about it in a face
to face talk on a fair.
It was really disturbing when googling for a solution on an issue
I had, never to know exactly whether the results referring to postgres
would really apply to the postgresql thing I was using as well.

Name confusion is not a minor thing and having postfix and postgres
running on the same machine (postfix is the default MTA under MacOS X)
caused confusion with the postmaster process showing in top and ps
when I expected it not to run, only to find out that it was not what
it seemed to be. And finally, you took action, great.

The strangest solution to the problem I found in this thread was
renaming to PostgresQL. Sorry, but there is no such thing as a
separate Postgres Query Language, incompatible with Standard SQL
as implied by this name. This would certainly bring confusion to
a new level. And please, do not install a committee to decide on
that - these are kind of solutions that come from such circles.

For me the discussion seems to be the type of thing where you keep
to say "no" until the day where you finally say yes. Its in the air.
I dont like postgres much better or better at all, but it is obvious
that it will be the official name sooner or later. It needs to happen
some day. The sooner the better.

From my point of view, the marketing name move should not be tied to
the next major release. Doing so puts pressure on the decision what
might be worth a major number and when it should happen (ASAP). This
is not appropriate for a project, where quality standards should
drive the decision for final release dates.

On the other hand, I share the doubts with package renaming. I use
the macports project for updating my system and in this package
management system the port to be installed is called postgresql8.
Renaming this to postgres8 would break all ports depending on it.
Most ports will continue to use an old version that does not receive
security fixes any more. But with the advent of postgres 9, there
will be a separate package named postgersql9 or postgres9 anyway,
because some other package maintainer might decide to stay with
the older release. Nobody will mind the name change then.
I guess this is the same with freebsd ports, debian apt-get,
redhat rpm, gentoo and myriad of other package management repos.

Under these circumstances, a big bang solution does not seem feasable.
The big bang type of name change usually happens only with companies
that have the bang to do so. But even they sometimes decide for a
incremental change. Rebranding the german mobile carrier Mannesmann D2
to vodafone was an example for this strategy and renaming the electric
power company HEW to Vattenfall also took more then a year and several
big sports events to make sure the transition works smoothly and people
dont refuse to pay to a company they never contracted with.

But, this does not mean necessarily that we need to switch silently.
I tend to use the elephants trunk and trumpet it out, but I have to
admit that it does not fit perfectly with an incremental strategy.

As with the all the work that needs to be done on the marketing side,
I dont understand these objections completely, because they should be
mentioned by those who would have to deal with it and not by the coders
who are opposed to a change. I bet, that people who do care for a good
name will take care for a good visual appearance as well. I guess that
the artwork stuff will be done in no time by people who are happy to
contribute. FreeBSD ran such a thing for their website relaunch as
their first Google Summer of Code project some years ago IIRC - just
a proposal.

Regarding printed paper, it was already mentioned that it usually
grows old before a year is over. T-Shirts are certainly different,
but they get a pale stripe across the breast when exposed to the sun
and dont look good after a while as well. Everybody who buys marketing
stuff for more then a year in advance will regret this decision some
day, whether the name changes or not, because of the cash drain, the
waste of storage and all this. And if somebody comes moaning to a
booth on a fair, complaining that the name on the booth and the website
are different, what better time to ask for some money to change this?

The only thing that I saw that is a real obstacle besides dullness were
the legal issues raised by the fact that a company holds rights to the
name but is not willing to explain what they intend to do with it. This
could become a complete showstopper. But I am not a lawyer and there
will be others who might find a way to deal with that.

This was more then two cents I admit, but spending hours reading,
I could not resist and keep my mouth shut.

Bye, Christian

In response to

Responses

Browse pgsql-advocacy by date

  From Date Subject
Next Message Robert Treat 2007-09-05 14:44:25 Re: Theme of this release: Performance?
Previous Message Ron Mayer 2007-09-05 14:18:13 Re: Theme of this release: Performance?

Browse pgsql-de-allgemein by date

  From Date Subject
Next Message Bernd Helmle 2007-09-05 15:24:15 Re: Out of Memory Probleme bei einem bytea Feld
Previous Message Albe Laurenz 2007-09-05 14:16:20 Re: Permanentbackup, alles ohne Unterbrechung des Betriebs