I lost faith that this guy knows what he's talking about when I read this:
> The PostgreSQL mailinglist doesn't have "don't kill -9 the postmaster!" as a standard signature to list messages for nothing.
Indeed that is standard advice, but it has NOTHING WHATSOEVER to do with power failure recoverability. The reason you're not supposed to do it is that hard-killing the postmaster doesn't get rid of its subprocesses or shared memory segment, which could make a subsequent attempt to restart the postmaster hazardous. But those things won't survive a system crash due to power loss (or any other reason).
I'm not really qualified to evaluate all the other statements in the article, but the fact that the one statement I do know about is hogwash doesn't make me feel good about the others.
Let's see... Shannon estimated human performance as being between 0.6 and 1.3 bits per character. How many significant digits are there in those numbers, do you think? Barely 1, else he'd have given a tighter range. To claim that 1.319 bits/char is "outside" the range of human performance while 1.299 bits/char will be "inside" it betrays an absolutely stupefying lack of understanding of statistics and experimental measurement.
This thread's various criticisms of the purpose of the Hutter Prize seem valid to me, but even if you accept the premise as sound, the original post is the silliest sort of meaningless hype-creation.
FC, like the rest of Colorado, is a great place to live as long as your definition of fun involves being up-close-n-personal with a lot of snow. If you don't like to ski, there's not so much to do.
Personally, I worked for HP there twenty-plus years ago, and I'm quite happy to have moved back to Pittsburgh.
(Not but what Pgh. got dissed in this survey too... most comic book stores per capita??? Not our claim to fame, I think.)
What I find kinda incredible in this discussion is the notion that Oracle... or any other big company... can "buy" Red Hat.
Yeah, they could buy up all the shares, and what would they have? A dead corporate shell. The people who make Red Hat what it is would certainly leave.
I'm afraid that Larry Ellison does not get this simple clue, and thinks that he can finagle the stock price and buy out Red Hat and have something worth owning. What would really happen is that the talent would scatter to the four winds, and probably eventually regroup, but still the general cause of FOSS would be set back several years while that happens. The only winner from that scenario is Redmond. Poor Larry, whose only ambition is to be richer than Bill, and yet he doesn't understand that his ambition is itself making sure that that won't happen.
> How much new SW has RedHat actually contributed to the community (not just support)?
You are kidding, no?
Red Hat employs quite a number of the core hackers for the Linux kernel, gcc, glibc, and any number of other critical bits of the FOSS infrastructure. Now certainly RH did not originate any of the components I just mentioned, but they have contributed a huge part of what's in those components today.
ObDisclaimer: I work for Red Hat... they pay me to work on PostgreSQL.
> RH was already struggling with a 34% drop in profit in their last quarter.
Just for the record: there was no "34% drop in profit". Red Hat's operating profit increased. The reason for the reported drop is that Red Hat started to account for stock options as current expense, as required by recent changes in federal law. Every other publicly traded company in the country either has done this or will do this in the next year or so, and every one of them that gives any significant amount of options to their employees will be taking a huge numerical hit when they change --- unless perhaps their stock has been so anemic that the options are worthless.
Red Hat's last couple of quarterly reports show the numbers both with and without this accounting change, and the numbers without are quite pleasant thank you...
(ObDisclaimer: I made a tidy profit off Red Hat options this year, so some of that discrepancy is in my pocket.)
Apparently you've never actually *USED* Postgres, because if you had, you'd know it prefers all-lower-case. You also seem to be quite unaware of the SQL standard requirements in this area. Mixed-case table names are not easy in any DB that comes close to honoring the letter of the SQL spec --- so claiming that this is a problem for conversion just shows that you don't know what you're talking about.
That's an utterly stupid argument, especially if you think it's a reason to contribute to mysql instead.
There isn't any way for someone to "take control" of a BSD-licensed project. Sure, someone could use a BSD project as the base for a proprietary project, but that isn't going to discourage anybody else from working on the open original. We have in fact watched several proprietary "improvements" of postgres quietly tank over the years.
On the other hand, MySQL AB own mysql lock stock and barrel, and only release GPL versions because they choose to. They could announce tomorrow that all their future versions will be high-price closed-source shrink-wrap, and no one could say boo to them about it. The difference from the postgres situation is that MySQL AB could take with them the vast majority of the existing development expertise for the code base. Postgres will continue as an open-source project no matter what any one company thinks about it --- you cannot say the same about mysql.
The responses on this thread seem to imagine that the NASDAQ-100 is based on some kind of nefarious scheme. Pay attention folks: this index is very simply defined, it's the top 100 NASDAQ stocks by market capitalization, membership refigured once a year in December. AFAICT there is no choice or strategy involved on the part of the index makers. You can argue all you like about whether an index defined this way is actually interesting or not --- but the fact that (eg) Red Hat is now "in" has nothing to do with any embrace of Linux, or "hot tech stocks", or anything except where the most investment dollars are going.
MaxDB is what MySQL AB calls the SAPdb codebase they bought from SAP. It has essentially nothing to do with MySQL 5.0 though. As for your implication that SAP does or ever will run on MySQL 5.0, yeah right.
No kidding. My recent business trips have mostly been Pittsburgh-to-and-from-Toronto. Door to door is about seven hours if I drive, and six hours if I fly (compared to about four hours before 9/11). Any more BS added onto the airport security check, and they lose this passenger permanently.
Of course it's not the first time. Check the credits for Internet Explorer sometime --- last I heard they were still using libjpeg, same as they have been since about 1994.
Microsoft loves open source, as long as it's the kind they can use without giving anything back to the community. So, BSD = good, GPL = bad, in their eyes.
Don't ignore the possibility that it's your own equipment malfunctioning. I was getting the weirdest junk on caller ID a couple months ago, until I figured out that the batteries in my caller ID display had died...
Unfortunately, you don't know what you're talking about. Afilias runs all of their registries on Postgres, and has done so since day one. (I know, I was there helping them go live with.info.)
That's a fairly silly stat, actually. The reason it shows up is not so much lines-of-source-code as everyone-on-the-planet-uses-libjpeg. The survey you're thinking of counted references, not net lines of code.
Not that I'm not proud of libjpeg. But UCB and FSF and MIT have certainly done a ton more open code than that.
Which part of "recovery tool" didn't you understand? pgfsck is a useful low-level dump tool... not my favorite, but okay... but it makes no attempt to repair broken PG files. I don't know of any code that does so attempt, and (to repeat myself) I wouldn't trust it if it did.
Eh? FSF won't accept contributions if you don't assign rights. Feel free to argue your interpretation of the GPL with RMS;-)
I think that MySQL AB are misusing the assignment privilege, but your concept of "force" seems misplaced. MySQL AB control that code base and they certainly can make their own decisions about what code they accept into it. No one is "forcing" anyone to make a contribution to MySQL, and certainly no one can or should force MySQL AB to accept contributions that they do not like the licensing of.
To put the shoe on the other foot: PostgreSQL is BSD-license, and I am on record as wanting to reject any contributions from people who want their contributions to be GPL'd. (We have accepted GPL'd contributions in past moments of thoughtlessness, and we need to do something to fix that...)
I'm not sure why I'm bothering to reply, when you're so obviously a troll who can't do basic arithmetic, but the various possible approaches for backup/restore in Postgres are described in http://developer.postgresql.org/docs/postgres/back up.html. The PITR-related options are new as of 8.0, but we have always had some form of hot backup.
> No bug fixes from anyone outside AB? No feature additions from anyone outside AB?
MySQL AB's approach to this is that they own your contributions, thank you very much --- see for example the intro to http://dev.mysql.com/doc/mysql/en/Contributors.htm l. I have not been through this personally but I think they require contributors of nontrivial work to sign an assignment of rights. Kinda like the FSF does, but with less benign intentions.
Slashdot Mirror
Yup, my reaction exactly. Whoever wrote this tool completely failed to get email. It's not IM, and that is not a bug.
I lost faith that this guy knows what he's talking about when I read this:
> The PostgreSQL mailinglist doesn't have "don't kill -9 the postmaster!" as a standard signature to list messages for nothing.
Indeed that is standard advice, but it has NOTHING WHATSOEVER to do with power failure recoverability. The reason you're not supposed to do it is that hard-killing the postmaster doesn't get rid of its subprocesses or shared memory segment, which could make a subsequent attempt to restart the postmaster hazardous. But those things won't survive a system crash due to power loss (or any other reason).
I'm not really qualified to evaluate all the other statements in the article, but the fact that the one statement I do know about is hogwash doesn't make me feel good about the others.
Let's see ... Shannon estimated human performance as being between 0.6 and 1.3 bits per character. How many significant digits are there in those numbers, do you think? Barely 1, else he'd have given a tighter range. To claim that 1.319 bits/char is "outside" the range of human performance while 1.299 bits/char will be "inside" it betrays an absolutely stupefying lack of understanding of statistics and experimental measurement.
This thread's various criticisms of the purpose of the Hutter Prize seem valid to me, but even if you accept the premise as sound, the original post is the silliest sort of meaningless hype-creation.
"Cygwin is required"? Apparently you haven't actually looked at Postgres in a few years. There's been a native port since PG 8.0.
FC, like the rest of Colorado, is a great place to live as long as your definition of fun involves being up-close-n-personal with a lot of snow. If you don't like to ski, there's not so much to do.
... most comic book stores per capita??? Not our claim to fame, I think.)
Personally, I worked for HP there twenty-plus years ago, and I'm quite happy to have moved back to Pittsburgh.
(Not but what Pgh. got dissed in this survey too
Um, did anyone actually bother to *READ* the linked-to wikipedia article?
...
The smiley as we know it was invented by Scott Fahlman at CMU, in late 1982. I remember, I was there
What I find kinda incredible in this discussion is the notion that Oracle ... or any other big company ... can "buy" Red Hat.
Yeah, they could buy up all the shares, and what would they have? A dead corporate shell. The people who make Red Hat what it is would certainly leave.
I'm afraid that Larry Ellison does not get this simple clue, and thinks that he can finagle the stock price and buy out Red Hat and have something worth owning. What would really happen is that the talent would scatter to the four winds, and probably eventually regroup, but still the general cause of FOSS would be set back several years while that happens. The only winner from that scenario is Redmond. Poor Larry, whose only ambition is to be richer than Bill, and yet he doesn't understand that his ambition is itself making sure that that won't happen.
> How much new SW has RedHat actually contributed to the community (not just support)?
... they pay me to work on PostgreSQL.
You are kidding, no?
Red Hat employs quite a number of the core hackers for the Linux kernel, gcc, glibc, and any number of other critical bits of the FOSS infrastructure. Now certainly RH did not originate any of the components I just mentioned, but they have contributed a huge part of what's in those components today.
ObDisclaimer: I work for Red Hat
> RH was already struggling with a 34% drop in profit in their last quarter.
...
Just for the record: there was no "34% drop in profit". Red Hat's operating profit increased. The reason for the reported drop is that Red Hat started to account for stock options as current expense, as required by recent changes in federal law. Every other publicly traded company in the country either has done this or will do this in the next year or so, and every one of them that gives any significant amount of options to their employees will be taking a huge numerical hit when they change --- unless perhaps their stock has been so anemic that the options are worthless.
Red Hat's last couple of quarterly reports show the numbers both with and without this accounting change, and the numbers without are quite pleasant thank you
(ObDisclaimer: I made a tidy profit off Red Hat options this year, so some of that discrepancy is in my pocket.)
Apparently you've never actually *USED* Postgres, because if you had, you'd know it prefers all-lower-case. You also seem to be quite unaware of the SQL standard requirements in this area. Mixed-case table names are not easy in any DB that comes close to honoring the letter of the SQL spec --- so claiming that this is a problem for conversion just shows that you don't know what you're talking about.
That's an utterly stupid argument, especially if you think it's a reason to contribute to mysql instead.
There isn't any way for someone to "take control" of a BSD-licensed project. Sure, someone could use a BSD project as the base for a proprietary project, but that isn't going to discourage anybody else from working on the open original. We have in fact watched several proprietary "improvements" of postgres quietly tank over the years.
On the other hand, MySQL AB own mysql lock stock and barrel, and only release GPL versions because they choose to. They could announce tomorrow that all their future versions will be high-price closed-source shrink-wrap, and no one could say boo to them about it. The difference from the postgres situation is that MySQL AB could take with them the vast majority of the existing development expertise for the code base. Postgres will continue as an open-source project no matter what any one company thinks about it --- you cannot say the same about mysql.
The responses on this thread seem to imagine that the NASDAQ-100 is based on some kind of nefarious scheme. Pay attention folks: this index is very simply defined, it's the top 100 NASDAQ stocks by market capitalization, membership refigured once a year in December. AFAICT there is no choice or strategy involved on the part of the index makers. You can argue all you like about whether an index defined this way is actually interesting or not --- but the fact that (eg) Red Hat is now "in" has nothing to do with any embrace of Linux, or "hot tech stocks", or anything except where the most investment dollars are going.
Er ... aren't you confusing MaxDB with MySQL?
MaxDB is what MySQL AB calls the SAPdb codebase they bought from SAP. It has essentially nothing to do with MySQL 5.0 though. As for your implication that SAP does or ever will run on MySQL 5.0, yeah right.
Um ... on what basis do you say that MySQL has the edge on development community size?
Sure they've got more *users*, but developers I don't think so.
No kidding. My recent business trips have mostly been Pittsburgh-to-and-from-Toronto. Door to door is about seven hours if I drive, and six hours if I fly (compared to about four hours before 9/11). Any more BS added onto the airport security check, and they lose this passenger permanently.
Of course it's not the first time. Check the credits for Internet Explorer sometime --- last I heard they were still using libjpeg, same as they have been since about 1994.
Microsoft loves open source, as long as it's the kind they can use without giving anything back to the community. So, BSD = good, GPL = bad, in their eyes.
The unicode upper/lower stuff works fine in PG 8.0 (except on Windows, where it will work fine in 8.1). I think your information is obsolete.
It's pronounced post-gres-cue-ell. That wasn't so hard was it?
... particularly on days when the code wasn't doing what I wanted ...
Not but what I've been heard to pronounce PGSQL as pig-squeal
Don't ignore the possibility that it's your own equipment malfunctioning. I was getting the weirdest junk on caller ID a couple months ago, until I figured out that the batteries in my caller ID display had died...
Unfortunately, you don't know what you're talking about. Afilias runs all of their registries on Postgres, and has done so since day one. (I know, I was there helping them go live with .info.)
a nts-13.htm#Response13TheInternetSocietyISOC ... but the competitors seem to have figured out that that argument isn't winning.
If you want independent confirmation, check out their bids for the registries, such as
http://www.icann.org/tlds/org/questions-to-applic
They got considerable flak for this the first couple times (orchestrated none too subtly by Oracle)
That's a fairly silly stat, actually. The reason it shows up is not so much lines-of-source-code as everyone-on-the-planet-uses-libjpeg. The survey you're thinking of counted references, not net lines of code.
Not that I'm not proud of libjpeg. But UCB and FSF and MIT have certainly done a ton more open code than that.
Which part of "recovery tool" didn't you understand? pgfsck is a useful low-level dump tool ... not my favorite, but okay ... but it makes no attempt to repair broken PG files. I don't know of any code that does so attempt, and (to repeat myself) I wouldn't trust it if it did.
Eh? FSF won't accept contributions if you don't assign rights. Feel free to argue your interpretation of the GPL with RMS ;-)
I think that MySQL AB are misusing the assignment privilege, but your concept of "force" seems misplaced. MySQL AB control that code base and they certainly can make their own decisions about what code they accept into it. No one is "forcing" anyone to make a contribution to MySQL, and certainly no one can or should force MySQL AB to accept contributions that they do not like the licensing of.
To put the shoe on the other foot: PostgreSQL is BSD-license, and I am on record as wanting to reject any contributions from people who want their contributions to be GPL'd. (We have accepted GPL'd contributions in past moments of thoughtlessness, and we need to do something to fix that...)
I'm not sure why I'm bothering to reply, when you're so obviously a troll who can't do basic arithmetic, but the various possible approaches for backup/restore in Postgres are described in http://developer.postgresql.org/docs/postgres/back up.html. The PITR-related options are new as of 8.0, but we have always had some form of hot backup.
> No bug fixes from anyone outside AB? No feature additions from anyone outside AB?
m l.
MySQL AB's approach to this is that they own your contributions, thank you very much --- see for example the intro to
http://dev.mysql.com/doc/mysql/en/Contributors.ht
I have not been through this personally but I think they require contributors of nontrivial work to sign an assignment of rights. Kinda like the FSF does, but with less benign intentions.