Re: 9.5 Release press coverage

On Wed, Jan 13, 2016 at 7:40 AM, Magnus Hagander <magnus(at)hagander(dot)net> wrote:
> We have typically not mentioned the individual developers either in the
> press release, and I think that's a good practice. They are credited in the
> release notes where we have space to credit *everybody*, and do so in a fair
> way. So I think we should go back to not doing that.

+1.

> As for the EDB press release, the *biggest* problem was that it was sent to
> -announce and approved there. Those were two mistakes in a row. Mistakes
> happen, let's just try not to do that again.

+1. And hopefully next time EnterpriseDB won't send it there, either,
but if they do, well, that's what the reject button is for.

> The wording itself wasn't very good, with the way the "announced" verb was
> used. We certainly can't prevent them from doing that, but we should forward
> that feedback and ask that they be more careful about that wording the next
> time.

I'd really like to understand what wording would be found acceptable
to the community. I think it's natural for a press release put out by
a company to begin with the name of that company. The press release
says in the first sentence that the release was made by the PostgreSQL
community. Simon seems to think that's unacceptable, but I can say
in all honesty that if 2ndQuadrant put out a similar press release
with EnterpriseDB replaced by 2ndQuadrant, the quotes from
EnterpriseDB employees replaced by quotes from 2ndQuadrant employees,
and the discussion of features EnterpriseDB cares most about replaced
by features 2ndQuadrant cares most about, I'd have no issue with that.
To be honest, I'd be sort of surprised if 2ndQuadrant ever put out ANY
public statement that made as much mention an EnterpriseDB-developed
feature as that press release made of BRIN. In that sense, I thought
that press release was remarkably fair. If somebody had asked me
about that press release before it had been put out, I would have
recommended against sending it to pgsql-announce, but I would have had
no concerns about using it anywhere else. I still don't really
understand what the problem is. Can you be more specific?

If the community's position is that the only acceptable thing for
EnterpriseDB to promote is the community press release that doesn't
mention EnterpriseDB or any EnterpriseDB staff or any
EnterpriseDB-contributed feature, and that it must promote that only
without using the word EnterpriseDB, I think that's, well, I guess I
think that's ridiculous. It's reasonable to expect that EnterpriseDB
won't say that we are the one company behind PostgreSQL, and the ONE
article that said that has now been quite thoroughly corrected. It's
not reasonable to say that EnterpriseDB won't talk about EnterpriseDB.
I fully expect other people to talk about their own companies.

Generally, I'd say that EnterpriseDB generates a regular stream of
press releases, and the community doesn't get veto power over those.
The community, of course, has every right to decide which of those
announcements it will promote using its own channels (pgsql-announce,
for example). But it has no right to control EDB's access to the
media. If EnterpriseDB is making false statements, then it is
entirely right for people to be upset about that, but our press
release does not do that. I have read that press release several
times from top to bottom and I do not see a single statement in there
that is false or claims credit for anybody else's work. Period.

> If EDB alone generated 60% of the readers (I can't comment on whether that
> is true or not), then I think we can say that the rest of the community PR
> effort failed. The fact that they are the ones that got into papers that
> wanted to interview someone in person and called them makes a round of sense
> - the community doesn't really offer that up (other than a list of phone
> numbers -- but that's pull rather than push). But if their written press
> release hit that many more target than *our* written press release did, then
> our written press release failed.

I think this is the right way to look at it. Personally, I don't see
much evidence that the 60% figure is in any way accurate. Consider
the six links in the blog entry I put up last night:

http://www.infoworld.com/article/3020020/database/postgresql-95-bolsters-sql-and-nosql-features.html
- Doesn't even mention EnterpriseDB.
http://www.v3.co.uk/v3-uk/news/2441110/enterprisedb-claims-impressive-performance-gains-in-postgresql-95-database
- Talks a lot about EnterpriseDB, but seems pretty clear about the
distinction between EnterpriseDB and PostgreSQL.
http://www.fiercecio.com/story/postgresql-95-eyes-mysql-users-new-web-mobile-friendly-features/2016-01-08
- Doesn't even mention EnterpriseDB
http://thenewstack.io/postgresql-9-5-geared-liberate-enterprises-data-warehouse/
- Beginning is only about PostgreSQL, EnterpriseDB mentioned down a
few paragraphs, distinction seems clear
http://sdtimes.com/postgresql-9-5-finally-clears-migration-path-from-mysql/
- Doesn't even mention EnterpriseDB.
http://www.theregister.co.uk/2016/01/07/postgresql_95_lands/ - Doesn't
even mention EnterpriseDB

So, here we have two-thirds of the press not even mentioning
EnterpriseDB, not 60% of the press mentioning only EnterpriseDB. Now
maybe that sample isn't representative, but I picked those links
mostly at random from a list given to me by Renee. I hope she doesn't
get in trouble for getting EnterpriseDB's name into only 30% of our PR
at the same time we're getting based on this mailing list for
dominating the PR. I don't believe EnterpriseDB needs to apologize
for wanting to be mentioned in the PostgreSQL 9.5 press.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company