Re: New and investigating - Mailing list pgsql-novice

Hi,
    Just thought I might drop my two cents worth in, so take this as my opinion
only ;-)

One thing that came across very clearly to me reading 'The cathedral and the
Bazaar' (read it if you haven't) is the difference in attitude between a
closed source company such as oracle and the open source hackers who write
products such as Postgres. Closed source companies tend to spend a lot of
time and money generated marketting hype, test cases and "big" clients in
order to make you want to buy their products. This makes sense as they need
to sell the product to make money. From the buyers point of view it gives
them that warm fuzzy feeling about spending all that money. I.e. if that
government or corporation is using this product then it must be good.

Open source people on the other hard tend to spend more time working on the
product and allow it to speak for them. This means that the sort of
documentation you might be looking for will either not be there or be a lot
harder to find. This may sound silly but on the other hand it also makes a
lot of sense. It allows the developers of the product to devote all of their
time to making the product good. This is best illustrated by the speed at
which open source products adds new features and fix bugs. I've yet to see a
closed source company that can respond as fast.

So is Postgres good ? I think you can only judge this by the quantity and
strength of the activity around it. Any open source project which does not
have a good core product will quickly die as people get frustrated and move
to other products. The fact that Postgres is doing well with a healthy and
active user base, plus plenty of development going on speaks well for it.

Should you try it ? I don't know. This is a decision that only you can make. I
would suggest that you do spend some of your time to install it on a test
machine and 'play' with it for a while to get the feel of the software and
how it works. It's not going to cost you anything bar some of your time. From
this you will then be able to decide whether Postgres is for you.

If theres one thing I've learned over the years it's that to never believe
what sales people tell me. I don't care how big the software company is or
how fabulous the product appears. I'll play with before I decide whether to
buy it. basically because there are always things they never tell you. Little
"gotcha's!" that aren't in the doco. Of course every product has them, even
Postgres. but whether a gotcha is a bug, design issue or just because you are
used to doing things differently, is different for each user.

Anyway, thats my little speach. Please don't shoot me if you don't like it.
It's just the way I think of things at the moment. I hope it helps in some
way.

cio
Derek.


On Thursday 30 January 2003 09:56, you wrote:
> Thanks for your "litle humour and rumour" Aarni.
>
> You evidently are and advocate of the "free" camp, and of course, how else
> could it be on this list! ;-)
>
> I apreciate your comments.
>
> About the question on "industry strength", you say that it "truly is" and
> should be an industry standard. Hopefuly, and because the truth always
> imposes itself (yeah right), it will happen.
>
> Now, how do i go then beyond personal opinions? Like you said yourself, it
> "might sound like a bit of propaganda". Nothing bad about it, but one needs
> to have hard evidence sometime... Where are the materials, links,
> companies, etc... that can back this up?? It is easy for you to say your
> opinion, because well... it is yours (!) and you have your own experience
> about it. But how about me? I don't have any of these...
>
> Can i possibly start a 'real' project just based on comments? No
> Can i invest my rare free time in investigating all this then? No.
> (Don't have enough of it. I would probably finish by 2010!)
> Do i have peers around it with the experience? Nope.
> Can i find enough material to get suficiently confident in PostgreSQL to
> just take the risk? Well, not yet...
> So, am i just doomed to look at PostgreSQL from far away? (and sigh... :-]
> )
>
> That can't be so.
>
> I of course understand that PostgreSQL is another kind of effort than
> Oracle et al, and doesn't count with its huge salesforce that pushes all
> kind of stuff down your throat to show you how good they are, but
> nevertheless, as someone who would like to know more (not even yet a
> 'newcomer'...) i still expect to find some kind of info that suficiently
> backs these things. Where do i find it? (or, how are people suposed to take
> decisions?)
>
> Or is this illusory? Should i just drop it, and must start with my own
> tests to sort this out? That sounds like a lot of work to just begin!
> Someone please tell me.
>
> (Tom Lane, how about a shot at this?)
>
> Alain Gougeon
> La Paz, Bolivia
>
>
> ----- Original Message -----
> From: "Aarni Ruuhimäki" <aarni.ruuhimaki@kymi.com>
> To: "Alain Gougeon" <agougeon@sigma.gov.bo>
> Cc: <pgsql-novice@postgresql.org>
> Sent: Tuesday, January 28, 2003 9:30 PM
> Subject: Re: [NOVICE] New and investigating
>
> > On Tuesday 28 January 2003 10:14 pm, you wrote:
> > > there
> > >
> > > My name is Alain, i have like 14 years behind me working with Oracle,
>
> and
>
> > > nothing else basically.
> >
> > It's about time ?
> >
> > > Recently i came across PostgreSQL and found it to be a very atractive
> > > thing, in concept at least. I would like to know more about it, but not
> > > technically yet, rather on the side of the capabilities it has from a
> > > "managerial" point of view. I was wondering if i could find some
>
> pointers
>
> > > here. Thanks everybody for they contributions.
> > >
> > > (i already saw the oficial PostgreSQL "advocacy" page)
> > >
> > > The kind of questions that i ask myself are like:
> > >
> > > . Is it really industrial strength? Where can i find articles/white
> > > papers/etc that document it? (many people in the lists mention MySQL
> > > and stuff like that, that's another ballgame alltogether)
> >
> > It truely is, in my opinion it should be the platform-independent
> > industry standard. PostgreSQL www-site is a good starting point for the
> > docs. With regard to MySQL, it is fairly easy to learn either one, once
> > you know the basics of the other one.
> >
> > > . How does it compare to Oracle and the other main players in regards
> > > of solidity, availability, trustability, dificulty of learning,
>
> installing,
>
> > > mantaining?
> >
> > I'am not familiar with Oracle, by choice I'am afraid, not too keen on
>
> these
>
> > 'official main proprietary players' like MS-SQL (you mean to say? I have
> > enough experience with MS-SQL not to play around with it anymore) etc.,
>
> but
>
> > as I see it, there are two standards: unix/linux and the rest of the
>
> world.
>
> > With two years' experience in the real world, and with dos /
>
> win-background,
>
> > I would not even consider setting up a serious db-system on an other
>
> platform
>
> > than linux + pgsql or such like. Learning curve may be steep if you do
> > not know your way around the basic command-line environment, but it is
> > really worth the 'trouble'. Right to the core, it is actually and
> > amazingly
>
> simple
>
> > and powerful. Ok, it gets complex with complex scenarios and if you want
>
> it
>
> > to ..., but help is readily available, just ask the guys ! One thing you
>
> will
>
> > have perhaps to do is to learn some perl, C, php, python or tlc/tk etc.
> > in order to make your backend available to your clients. I'd prefer some
> > GUI
>
> via
>
> > 'www', then you have an interface that is platform-independent, all your
> > clients need is a connection and a browser. And how fast can you move
> > your entire db-system to another machine ? With a couple of commands and
> > file transfers ? Yes.
> >
> > > . How powerful is its internal language (we're quite spoiled here
>
> working
>
> > > with Oracle's PL/SQL which is very good...)
> >
> > Hmm, very powerful I'd say, and just by reading this list, you learn a
>
> lot.
>
> > > . Is there any article or study done of PostgreSQL from an Oracle
> > > background: ups and downs, surprises, diferences, important
> > > considerations, etc...
> >
> > Don't know, sorry, but the question has been asked now ?
> >
> > > . In what cases would it be recomendable and relatively riskless to
> > > consider using PostgreSQL instead of Oracle, and, much more
> > > importantly,
>
> in
>
> > > which cases NOT.
> >
> > Any and can't think of any. In a broader perspective, if you are willing
>
> to
>
> > take the responsibility to design and further develop a system that fits
>
> your
>
> > needs now and tomorrow you'll go PG and open source and if it fails
>
> everyone
>
> > laughs at your decision and you. You can also spend a small or bigger
>
> fortune
>
> > on a commercial system that does not really fit your needs or probably
> > over-fits it by far and you do not have much control over, and if that
>
> fails,
>
> > it is not your responsibilty, but the big international vendor's and
>
> nobody
>
> > dares to laugh ... ( but really is laughing at both ? )
> >
> > > . It seems that documentation might be sparse. How confident can one be
>
> of
>
> > > obtaining technical knowledge of PostgreSQL without a very long trial
>
> and
>
> > > fail path? Are there books on it?
> >
> > On-line docs, mailing lists, etc. for free. And good books, Bruce ? Trial
>
> and
>
> > fail is how one really learns. But again, with the help of others further
>
> on
>
> > the road you don't have to follow all the wrong paths.
> >
> > > . In the few mails that i have been checking, i saw a lot of thing with
> > > which
> > > i am not familiar, but... PostgreSQL is compliant with ANSI SQL, right?
> >
> > And more.
> >
> > > . I've read an interview of Tom Lane where it comes out that PostgreSQL
> > > came out to be faster than the other databases in some test. Anybody
> > > got
>
> a
>
> > > link to
> > > that article? tests?
> > >
> > > . How efficient is PostgreSQL in terms of memory usage, queueing and
> > > all these elaborated but useful (usually) things that show up
> > > inevitably in high end products (like Oracle). (that could be rephrased
> > > as, "how not
>
> so
>
> > > basic is PostgreSQL?"
> > >
> > > etc...
> >
> > Tom's territory ? "Hardware is the limit".
> >
> > > I guess you see the picture.
> >
> > Yes, have no fear !
> >
> > > At present all the development i am in charge of relies heavily on
>
> Oracle's
>
> > > PL/SQL, but with the new ages coming up of having several tiers, and
> > > the evergrowing needs for computing resources in front of the shrinking
>
> budgets
>
> > > of struggling economies, this is something i would reconsider.
>
> PostgreSQL
>
> > > places itself like a potential alternative to the oficial great actual
> > > players, but it is really hard to trust anybody's speech right out. The
> > > truth normally lies in many more details, known to the experienced
> > > practicioners. I hope to be able to gather some impresions here
> >
> > Shrinking budgets and resources, yes. So why waste money on a huge and
>
> clumsy
>
> > system that messes-up things all by itself and eats up all resources just
>
> to
>
> > keep itself running ? Repairing calls for out-of-town consultants, who
>
> come
>
> > in, do their thing and vanish before you can say do-re-mi in swahili.
> > Something like 600 USD per hour ... And initial cost ? Tons of money for
> > a pile of installation cds and nothing else. Maybe a manual and contact
>
> numbers
>
> > for the consultants.
> >
> > > Thanks everybody.
> >
> > This might sound like a bit of propaganda here, but having seen 'the
>
> light', I
>
> > am quite enthusiastic about it.
> >
> > And please see:
> >
> > http://www.mslinux.org
> >
> > A little humour and rumour ...
> >
> > Best regards,
> >
> > aarni
> >
> > > Alain Gougeon
> > > La Paz, Bolivia.
> > >
> > >
> > >
> > >
> > > ---------------------------(end of
> > > broadcast)--------------------------- TIP 3: if posting/reading through
> > > Usenet, please send an appropriate subscribe-nomail command to
> > > majordomo@postgresql.org so that your message can get through to the
> > > mailing list cleanly
> >
> > ---------------------------(end of broadcast)---------------------------
> > TIP 6: Have you searched our list archives?
> >
> > http://archives.postgresql.org
>
> ---------------------------(end of broadcast)---------------------------
> TIP 4: Don't 'kill -9' the postmaster

--
Derek Clarkson
Web: http://users/bigpond/net/au/drekka