Обсуждение: Stability, PR
I've been using PostgreSQL for several years now, and have been extremely happy during that entire time period. Over the last 4-6 weeks, however, several potential clients questioning my choice of PostgreSQL, saying that it has a reputation for instability. These comments didn't come from people who are against using open-source software; my choice of Linux, Apache, mod_perl, Zope, or OpenACS (depending on the project) was generally applauded. I've managed to convince each of these clients that PostgreSQL is indeed a reliable database. But I I still have to wonder: Is there any truth to these statements? Is it simply FUD, or was PostgreSQL once prone to data corruption? Given the frequency with which I've heard such claims from otherwise well-informed people, I'm wondering if there's anything we users can (or should) do to counter this misinformation. Reuven
"Reuven M. Lerner" <reuven@lerner.co.il> writes:
> I've managed to convince each of these clients that PostgreSQL is
> indeed a reliable database. But I I still have to wonder: Is there
> any truth to these statements? Is it simply FUD, or was PostgreSQL
> once prone to data corruption?
Before about 7.0, I'd have said it was a reasonable complaint.
We've done an awful lot to improve reliability over the last few
releases.
regards, tom lane
> Over the last 4-6 weeks, however, several potential clients > questioning my choice of PostgreSQL, saying that it has a reputation > for instability. These comments didn't come from people who are > against using open-source software; my choice of Linux, Apache, > mod_perl, Zope, or OpenACS (depending on the project) was generally > applauded. From what I've read, I've inferred that long, long ago, very early versions of PG weren't all that great - but I have no personal experience with them, so I can neither confirm nor deny that. What I CAN tell you, though, is this: The amount of idiotic statements that I've heard about PostgreSQL is amazing. Nearly all of it has come from "non-technical techies". You know, the people who sell computers, deal with computer-related investments, manage technically-competant people, but aren't actually capable of doing anything useful with a computer themselves. I've had to deal with a (seemingly) innumerable host of these kinds of people. For some, a simple review of the performance, scalability, and rock-solid reliability of our PG servers is enough. For others, it's not, and they really seem bent on the idea of us using something different. My response to them is "Sure, we'll do it. And you'll only have to fork over a few hundred thousand dollars, maybe half a million, to cover the licensing fees and hardware upgrades to get us to the level we're at right now." Somehow, that seems to help them understand a little more clearly... steve
I love the "non technical techies", i know one who spent three days writing a program to time the "connect time", "retrieve time" and "delete time" on an access datafile and a postgres server. He did this to prove to me that "access would be quicker *sigh*. Needless to say, when we got down to select *'s and working with recordsets, PG came out on top. anywho, i did an impromptu survey not long ago on who's using Postgresql, it i posted the responses to the group, it should be in the archives. Not once did i see any references to data integrity. BTW, thanks to everyone for their reponses again. We're going full steam ahead on migrating the Access app to Postgres. It saved our company at least 60k USD not going to MS-SQL. ~corey "Why Postgres? Why not?" -----Original Message----- From: Steve Wolfe [SMTP:steve@iboats.com] Sent: Wednesday, May 22, 2002 3:24 PM To: pgsql-general@postgresql.org Subject: Re: [GENERAL] Stability, PR > Over the last 4-6 weeks, however, several potential clients > questioning my choice of PostgreSQL, saying that it has a reputation > for instability. These comments didn't come from people who are > against using open-source software; my choice of Linux, Apache, > mod_perl, Zope, or OpenACS (depending on the project) was generally > applauded. From what I've read, I've inferred that long, long ago, very early versions of PG weren't all that great - but I have no personal experience with them, so I can neither confirm nor deny that. What I CAN tell you, though, is this: The amount of idiotic statements that I've heard about PostgreSQL is amazing. Nearly all of it has come from "non-technical techies". You know, the people who sell computers, deal with computer-related investments, manage technically-competant people, but aren't actually capable of doing anything useful with a computer themselves. I've had to deal with a (seemingly) innumerable host of these kinds of people. For some, a simple review of the performance, scalability, and rock-solid reliability of our PG servers is enough. For others, it's not, and they really seem bent on the idea of us using something different. My response to them is "Sure, we'll do it. And you'll only have to fork over a few hundred thousand dollars, maybe half a million, to cover the licensing fees and hardware upgrades to get us to the level we're at right now." Somehow, that seems to help them understand a little more clearly... steve ---------------------------(end of broadcast)--------------------------- TIP 6: Have you searched our list archives? http://archives.postgresql.org
(For those who don't run across Linux Journal, Reuven has been giving
PostgreSQL some nice exposure on a regular basis.)
> Given the frequency with which I've heard such claims from otherwise
> well-informed people, I'm wondering if there's anything we users can
> (or should) do to counter this misinformation.
One of the nicest things about the PostgreSQL team is the lack of FUD
and marketing fluff. To a fault, and to an extent not seen in other
database products, including other open source database products.
How does one prove something? By developing a history of success, not by
claiming it. And folks like yourself help a lot by showing success.
- Thomas
Thomas Lockhart <lockhart@fourpalms.org> writes: > (For those who don't run across Linux Journal, Reuven has been giving > PostgreSQL some nice exposure on a regular basis.) > > > Given the frequency with which I've heard such claims from otherwise > > well-informed people, I'm wondering if there's anything we users can > > (or should) do to counter this misinformation. > > One of the nicest things about the PostgreSQL team is the lack of FUD > and marketing fluff. To a fault, and to an extent not seen in other > database products, including other open source database products. > > How does one prove something? By developing a history of success, not by > claiming it. And folks like yourself help a lot by showing success. > > - Thomas I couldn't agree more. The PostgreSQL team is honest to a fault about the true state of the codebase. They will happily tell you which parts of the code are still somewhat scary. On the other hand, when they say that something is ready for prime time, you can be assured that it *is* ready for production use.
On Thu, May 23, 2002 at 09:35:12AM -0600, Jason Earl wrote:
> Thomas Lockhart <lockhart@fourpalms.org> writes:
> > One of the nicest things about the PostgreSQL team is the lack of FUD
> > and marketing fluff.
> > How does one prove something? By developing a history of success, not by
> > claiming it. And folks like yourself help a lot by showing success.
> >
> > - Thomas
>
> I couldn't agree more. The PostgreSQL team is honest to a fault about
> the true state of the codebase. They will happily tell you which
> parts of the code are still somewhat scary. On the other hand, when
> they say that something is ready for prime time, you can be assured
> that it *is* ready for production use.
While I agree completely with these sentiments, it does strike me
that one of the (many) sad bits about the demise of Great Bridge was
the death of the marketing efforts that they were able to undertake.
I had a recent experience which reminded me that the full-blow
bull-pucky claims of certain "unbreakable" systems stick in the minds
of those "non-technical techies" someone was recently bemoaning.
And, like it or not, those guys make a lot of decisions.
Worse, there are a lot of _bona fide_ Old Fart techies who are just
not as plugged into the current technology as they once were.
They've been promoted (?) to mangle^H^H^Hagement, and they are asked
to make decisions that they are really no longer informed enough to
be making. In such cases, they are likely to reach for the "safe"
bet, even if there's nothing especially safe about it.
So, a little PR goes a long way. But a _lot_ of PR goes a much
longer way, and may make the little PR unheard.
It struck me earlier today that maybe what free RDBMS need is
something like USENIX and SAGE. That way, there'd be a consistent
voice focussing just on the RDBMS world, pointing out the advantages
of systems like PostgreSQL. It also struck me that there might
already be such a thing. Is there? And if not, is there any reason
to hope that one could get it going without having a lot of pointless
name-calling about (for instance) whether MySQL really belongs?
A
--
----
Andrew Sullivan 87 Mowat Avenue
Liberty RMS Toronto, Ontario Canada
<andrew@libertyrms.info> M6K 3E3
+1 416 646 3304 x110
Andrew Sullivan <andrew@libertyrms.info> writes:
> It struck me earlier today that maybe what free RDBMS need is
> something like USENIX and SAGE. That way, there'd be a consistent
> voice focussing just on the RDBMS world, pointing out the advantages
> of systems like PostgreSQL. It also struck me that there might
> already be such a thing. Is there?
I had thought that OSDN's "Open Source Database Summit" conferences
might evolve into something like that --- but I fear they got
discouraged by last year's low turnout. (Never mind that a lot of
people were afraid to travel last September...) I don't see any
indication on their website that they're planning another one.
Too bad.
regards, tom lane
Tom Lane wrote: > Andrew Sullivan <andrew@libertyrms.info> writes: > > It struck me earlier today that maybe what free RDBMS need is > > something like USENIX and SAGE. That way, there'd be a consistent > > voice focussing just on the RDBMS world, pointing out the advantages > > of systems like PostgreSQL. It also struck me that there might > > already be such a thing. Is there? Agreed. We need another big marketing push. > I had thought that OSDN's "Open Source Database Summit" conferences > might evolve into something like that --- but I fear they got > discouraged by last year's low turnout. (Never mind that a lot of > people were afraid to travel last September...) I don't see any > indication on their website that they're planning another one. > Too bad. Yes, I noticed that too. We usually would have heard about it by now. -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania 19026