Re: Training approval policy on pg.org

Le 10/01/2013 19:14, Magnus Hagander a écrit :
> On Thu, Jan 10, 2013 at 6:45 PM, Josh Berkus <josh(at)agliodbs(dot)com> wrote:
>>
>>> It is likely because we have in the past had lots of people advertise
>>> training in bulk and then end up canceling most of them.
>>
>> We did. The policy was specifically to address a couple of companies
>> who were listing a training event every week, in order to upstage other
>> training companies.
>
> Yes.
>
> An honest question to Damien though - do you actually expect to *run*
> all these training sessions, or are you basicaly doing the same thing
> - settings up lots of options and then plan to run the most popular
> ones?
>

Short answer is "Yes I expect to run most of these training sessions".

Here's why :

First of all, every session I've submitted is "real" in the sense that
you can find it on our website (http://www.dalibo.com/formations) and on
our resellers catalogs... We don't invent fake trainings just for fun.
To be honest, we don't really care about other training companies
because we're operating on a specific market (French-speaking Europe)
where we face little competition. So basically we're not submitting
trainings on pg.org to upstage anyone, we just want to let people know
what we are doing.

About training cancellation : last year we planned approx. 30 public
sessions and actually run approx. 20 sessions. That's a 33% cancellation
rate, which is not so bad for training company in fact. For 2013, we've
made a lot of strategic changes to lower that rate and we're planning to
have a 20% cancellation rate. (I can't be more specific about what we
did to achieve that sorry)

In the meantime, we're facing an average growth of 30% on the Postgres
training market in France every year since 2009 and I don't see that
changing in 2013. Once again I can't post more details on this list.

That alone makes me think that we will run 33 sessions among the 42 we
have planned.

That being said, I don't see why submitting "unlikely sessions" should
be a problem at all. For instance, we trying new things this year such
as some PostGIS trainings (with Oslandia) and a couple sessions in
Brussels (with Open DB Team). I can't really say if this is gonna work
or fail, because it's new for us... It's a test and it's exactly in
cases like this that we need to publish the sessions on pg.org. It's a
basic chicken-egg situation : you need a minimum number of attendees to
run a training session. Noboby will register if you don't plan at least
a few sessions. When you try new trainings, you have a high cancellation
rate.

I understand there might have been a problem before with a couple of
trolls posting too many unlikely sessions... But this is not what we are
doing here. We don't believe in the Google pagerank religion. We suck at
SEO. We don't need to upstage anyone.

We just want to let people know what we plan to do. If that's not
possible on postgresql.org, well nevermind. We'll find something else to
do with our time :-)