Re: Usability, MySQL, Postgresql.org, gborg, contrib, etc.

> When all is said and done, I think the PostgreSQL project lacks a "Product
> Management" group which steers the public perception and defines
> usability. This is something *all* other systems have, including MySQL.

Well, NO, not a chance.

As one of the de-facto heads of our Advocacy group, let me say "you've got to
be kidding."

MySQL is a private closed-shop software manufacturer who uses the GPL as one
of their methods of distribution. They are not a "real" open source project
-- they are a private, commercial, for-profit software company.

We are not. We are a open, 100% voluntary community of contributing
developers and support volunteers. Nobody is going to take orders from a
"Product Manager", that's a position for paid software products departments.
It would be horribly inappropriate for PostgreSQL, and would destroy
everything that has made us successful to date. You can't give orders to
volunteers.

You're statement that "all other systems" have a Product Manager is also
wildly inaccurate. I think you're thinking of MySQL and Mozilla only.
Heck, even OpenOffice.org doesn't have a Product Manager, and that is a
Sun-sponsored project. Let alone, say, emacs. Or Linux.

--
Josh Berkus
Aglio Database Solutions
San Francisco

>
>> When all is said and done, I think the PostgreSQL project lacks a
>> "Product
>> Management" group which steers the public perception and defines
>> usability. This is something *all* other systems have, including MySQL.
>
> Well, NO, not a chance.
>
> As one of the de-facto heads of our Advocacy group, let me say "you've got
> to
> be kidding."

I don't think so, let me explain below.

>
> MySQL is a private closed-shop software manufacturer who uses the GPL as
> one
> of their methods of distribution. They are not a "real" open source
> project
> -- they are a private, commercial, for-profit software company.

I agree, I don't like MySQL all that much.

>
> We are not. We are a open, 100% voluntary community of contributing
> developers and support volunteers. Nobody is going to take orders from a
> "Product Manager", that's a position for paid software products
> departments.
> It would be horribly inappropriate for PostgreSQL, and would destroy
> everything that has made us successful to date. You can't give orders to
> volunteers.

It depends on the volunteers. Some are useless at taking orders, this is
true, others, however, are very welcoming to direction. It depends in the
individual. Lastly, Bruce, Tom, Peter, and others are very didicated to
PostgreSQL. If a real case can be made for a feature, I'm sure they are
reasonable enough to see that and grudgingly implement it. Someone,
however, has to keep an eye on that ball.

>
> You're statement that "all other systems" have a Product Manager is also
> wildly inaccurate. I think you're thinking of MySQL and Mozilla only.
> Heck, even OpenOffice.org doesn't have a Product Manager, and that is a
> Sun-sponsored project. Let alone, say, emacs. Or Linux.

Linux has Linus, he has a very good eye in the market forces. Emacs, is a
non-entity outside UNIX hard core UNIX guys. OpenOffice is very much
managed by Sun.

Hey,

First of all, who is this? I don't recognize the e-mail, and you haven't
been signing any of your posts.

> true, others, however, are very welcoming to direction.

AFAIK, this includes none of our major code contributors. So all you're
really talking about is manipulating the TODO list. You can't tell
programmers what to code unless you're paying them.

> It depends in the
> individual. Lastly, Bruce, Tom, Peter, and others are very didicated to
> PostgreSQL. If a real case can be made for a feature, I'm sure they are
> reasonable enough to see that and grudgingly implement it. Someone,
> however, has to keep an eye on that ball.

Yes, but they don't need a title to do so. Nor is there any reason for this
to be one person. In fact, you've just described one of the reason for the
Core's existance -- and even the Core defers to the consensus of decision on
this forum about which features to implement and how.

Now, if you're arguing that we could use a more cohesive, readable roadmap?
Sure! Want to prepare one? I can even help you find out what's under
development and what's not likely any time soon.

> Linux has Linus, he has a very good eye in the market forces.

Uh-huh. So? That still doesn't make him a "product manager".

> OpenOffice is very much
> managed by Sun.

I used to be a Project Lead for OpenOffice.org. I think the amount of
consensus and compromise, and the extent to which the Community Council and
the Project Leads govern the project, would surprise you.

Overall, I've not seen you present any coherent arguments as to:
1) why we need a new person with a title for marketing stuff;
2) what this person would be doing that's not already covered by existing
groups;
3) how this person would be able to accomplish their "job"; and
4) who this person would be.

As far as I'm concerned, we need use titles here only if it lends the entitled
some kind of authority with the outside world that helps them on their
volunteer projects (Robert Bernier, "Business Intelligence Analyst", is a
good example of a good use of titles -- that one convinces companies that he
approaches about case studies that he's for real). Titles are not at all
useful *inside* the community, we don't need them.

--
-Josh Berkus
Aglio Database Solutions
San Francisco

> Hey,
>
> First of all, who is this? I don't recognize the e-mail, and you haven't
> been signing any of your posts.

I've been posting on hackers on and off for a few years. My name is Mark.
>
>> true, others, however, are very welcoming to direction.
>
> AFAIK, this includes none of our major code contributors. So all you're
> really talking about is manipulating the TODO list. You can't tell
> programmers what to code unless you're paying them.

Yes and no. People can and do what's needed when it is clearly
articulated. What is lacking is a clear direction WRT marketing.

>
>> It depends in the
>> individual. Lastly, Bruce, Tom, Peter, and others are very didicated to
>> PostgreSQL. If a real case can be made for a feature, I'm sure they are
>> reasonable enough to see that and grudgingly implement it. Someone,
>> however, has to keep an eye on that ball.
>
> Yes, but they don't need a title to do so. Nor is there any reason for
> this
> to be one person. In fact, you've just described one of the reason for
> the
> Core's existance -- and even the Core defers to the consensus of decision
> on
> this forum about which features to implement and how.

I think I am talking about something different. In a company, the core
team would be the CTO. I think some entity, one or more people, needs to
define the product. Typically this is marketing and product management.

>
> Now, if you're arguing that we could use a more cohesive, readable
> roadmap?
> Sure! Want to prepare one? I can even help you find out what's under
> development and what's not likely any time soon.

Absolutely, but it would be meaningless if no body listens.

>
>> Linux has Linus, he has a very good eye in the market forces.
>
> Uh-huh. So? That still doesn't make him a "product manager".

Maybe I've overstated my case, by management I mean the small 'm' not the
big 'M'

>
>> OpenOffice is very much
>> managed by Sun.
>
> I used to be a Project Lead for OpenOffice.org.

Very cool. It is a great project/product.

> I think the amount of
> consensus and compromise, and the extent to which the Community Council
> and
> the Project Leads govern the project, would surprise you.

No it wouldn't.

>
> Overall, I've not seen you present any coherent arguments as to:
> 1) why we need a new person with a title for marketing stuff;

The why is that there is no real entity doing so.

> 2) what this person would be doing that's not already covered by existing
> groups;

All the groups, with the exception of advocacy, are "here's what we are
building" and "here's a bug" groups. There is planning on hackers, but it
is almost purely technical. Marketing features do no often get a
reasonable hearing.

> 3) how this person would be able to accomplish their "job";
I think that a talented manager could make the case for certain features.

> 4) who this person would be.
We recrute like a company does.

>
> As far as I'm concerned, we need use titles here only if it lends the
> entitled
> some kind of authority with the outside world that helps them on their
> volunteer projects (Robert Bernier, "Business Intelligence Analyst", is a
> good example of a good use of titles -- that one convinces companies that
> he
> approaches about case studies that he's for real). Titles are not at all
> useful *inside* the community, we don't need them.

I'm not trying to change the dynamic significantly, but I think, again if
increasing usership is important, that some market driven lessons need to
be learned.

Hi, Mark,

Yes, I've seen your e-mails around. You should use a sig, though, they're
easy to create.

> I think I am talking about something different. In a company, the core
> team would be the CTO. I think some entity, one or more people, needs to
> define the product. Typically this is marketing and product management.

But, as Peter reminds me all the time, we're not a company. ;-)

> The why is that there is no real entity doing so.
>
> > 2) what this person would be doing that's not already covered by existing
> > groups;
>
> All the groups, with the exception of advocacy, are "here's what we are
> building" and "here's a bug" groups. There is planning on hackers, but it
> is almost purely technical. Marketing features do no often get a
> reasonable hearing.

That's not an argument for not using existing apparatus. What you've
persuaded me is that you should:
a) join the Advocacy group;
b) galvanize people around developing a coherent marketing plan;
c) Lead a crew of volunteers and follow through on that process until the plan
is ready for comments by Core and Hackers,
d) stick around for the arguments and revisions

That's what we *need*. We don't need a volunteer with a title who might or
might not do any of the above.

I'm particulary struck by the fact that you chose to inaugurate this
discussion on Hackers, instead of Advocacy where it would have been more
appropriate and where more of the *existing* marketing volunteers would have
participated. At this point, I'd have to forward the whole thing to transfer
it ...

> I think that a talented manager could make the case for certain features.

So? So could any community member with a good grasp of database engineering
and an ability to write persuasive e-mails.

> > 4) who this person would be.
> We recrute like a company does.

Um, and pay them with what? Cowrie shells?

--
-Josh Berkus
Aglio Database Solutions
San Francisco

On Mon, 26 Apr 2004, Josh Berkus wrote:

> > I think that a talented manager could make the case for certain features.
>
> So? So could any community member with a good grasp of database engineering
> and an ability to write persuasive e-mails.

I'd like to inject here that I was the one who started the whole argument
over date style enforcement about a year ago. It was a bit of a slog at
first, but I felt my input was good, and the cause was just :-).

Because of that discussion, PostgreSQL accpeting dates like

2004-31-03 and turning them into 2004-03-31 on the fly stopped. as did
postgresql accepting a date like 03/31/2004 and turning it into 31/03/2004
when it was set to accept Euro style dates.

I am not a PostgreSQL hacker, just a user. My input made the difference.
No title needed. What PostgreSQL needs more of is caring users who can
spot bugs / misbehaviours, make a decent argument about it, and maybe even
a patch or two. We're way more short on indians than chiefs in my
opinion.