At one point tracking the entire.org domain was done with PG as well
Still is; even more than that actually,.info too. See Afilias Global Registry for a full list. Afilias sponsors quite a bit of PostgreSQL related work, including employing some of the major contributors to Slony, one of the most popular PostgreSQL replication solutions.
At which point NTT would promptly hire/promote someone else into the same position with the same level of resources necessary to be effective. For example, in addition to committer Itagaki Takahiro, they also have some major work on replication being led by Masao Fujii. The point of having a big company like NTT involved is that you can't just make their need for PostgreSQL to be successful go away so easily. There's not just "that one NTT employee"--he's one of a whole team there doing PostgreSQL related work.
And those are just the public record. Because of its BSD license, PostgreSQL also gets used in plenty of places that don't talk about what they're doing with it. For example, I've worked with financial companies that are cutting loose Oracle for PostgreSQL whenever feasible, and with some US defense companies that use PostGIS for geographic databases. (looks out window) I may have already said too much.
I just spent a few hours playing the NES version of MULE with a few friends while snowed in a few weeks ago. It was just as great as ever. Some of the games you mention really haven't aged well; this is not the case for MULE at all.
I was a little disappointed by some of the gameplay changes in this new version. They were so close to having a perfect remake, then changed some things that I shift the gameplay balance in a way I didn't think was an improvement. If you're going to try and remake a game this carefully balanced, you should do as good of a job as possible first, then consider adding optional changes. The same issue plagued the earlier Space HORSE remake too.
Quoting the reference I linked to and highlighting the important bit:
"You can change the SQL mode at runtime by using a SET [GLOBAL|SESSION] sql_mode='modes' statement to set the sql_mode system value. Setting the GLOBAL variable requires the SUPER privilege and affects the operation of all clients that connect from that time on. Setting the SESSION variable affects only the current client. Any client can change its own session sql_mode value at any time."
Yes, you have to be the superuser to change the global mode. And the fact that it says "Server SQL Modes", but they're not really enforced by the server even if you change the defaults should a client decide to go sloppy, is exactly why this behavior is so shady. It's an example of exactly why the upthread idea of PostgreSQL trying to be as correct as possible by design from day one is so important--you can't just bolt this stuff on later and expect there to not be a hole left behind.
You'll find more references for what you're talking about Why PostgreSQL Instead of MySQL, and the associated Transactional DDL page goes over how it compares with Oracle--which has recently added a feature to address this area, so picking on it in this area is a bit out of date now.
Re:There is already a perfectly good free DBMS
on
Monty Wants To Save MySQL
·
· Score: 2, Informative
No problems with invalid data types if you bother to RTFM and setup the config properly. (It's not hard. Just turn strict mode on.)
Strict mode is a client setting, which means that you're always at the mercy of applications turning it off and inserting garbage. And there's the "IGNORE" keyword to allow that too. There is no way to make a MySQL server reject all incoming bad data.
The documentation is not GPL, a move defended quite weakly by Sun. That article does address one of the real reasons for what Monty is doing though: "MySQL forks should have their own, specific, documentation". Monty doesn't want to recreate all that from scratch, so forcing that into more open licensing would make his life easier.
Hint: supplement "infomercials" don't come with links to suggested reading material, including lists of relevant clinical studies, from sources like the NIH. I've read several of those, selected a product based on the best information available, and gave a quick summary of why I thought it was a good one. Your [citation needed] suggestion was to eat dirt for "the stuff you would have gotten from the food supply"--now that's unsupported infomercial talk.
I look forward to the future clinical results of your research, showing the positive impact of repopulating internal bacteria through dirt consumption. Until then I'll stay with the probiotics.
And what makes you so sure that your particular dirt will actually have useful bacteria in it? The problem here isn't just that people aren't consuming from the same sources they used to, it's that sometimes those things just don't have what they used to in them. There is a serious lack of evidence to your medical theory "modern dirt has everything you need". Whereas even conservative sources like the NIH have been compelled to note the growing evidence that probiotics are useful for digestion issues.
In any case, the situation around one of the specific bacteria mentioned in the article is much more controversial: h. pylori disappearing is not necessarily a bad thing. And given that the common transmission vectors are oral-oral or fecal-oral, if you wanted more, unless you're using dirt that people shit in regularly you'd be better off with some kissing instead.
Note that the GP's specific suggestion, the Jarrow Labs EPS, is crap. I'm taking Nature's Way Primadophilus Optima probiotics, which has almost an order of magnitude more bacteria when new and includes the food the little buggers live off of in the tablet too. >10 bacterial stains, >25M CFUs when new, and bundled FOS are the minimums I look for in a good quality probiotic. Nobody is quite sure which of the bacteria are responsible for what yet, so a shotgun approach works better than a focused one here.
The idea of using a USB drive for any sort of large storage gives me the willies for the reason you list, which is extremely important, as well as some others:
-USB bridges rarely pass any SMART data back to the host, which means that if it starts to fail you won't know until it's already lost data. How much data? You'll have no idea. Well, you could if you used ZFS or other checksumming, but even then you're not going to know what wrong until data is gone.
-Drives in the 1TB range are going to be regular internal drives with a USB bridge on them. Ever notice how external drives normally have a much shorter warranty on them than the internal versions of the same basic disk? That's because moving a desktop drive around is really bad for it. They just don't last very well in that situation, thus the derated warranty and lower expected lifetime.
-Drive failure rates are so bad, and the time it takes to make a full copy of a drive this big so long over USB, that you're way too exposed to losing two drives. If you care about your data at all, ideally you'd want a RAID-1 pair of internal drives in a server system. It's reasonable in that situation to make an off-site backup you sync periodically, but you don't want a portable drive to be your only primary copy.
Note that the first couple of these issues, the ones related to the crappy USB chipsets, can be avoided if you get an external drive that connects via eSATA. Then you get SMART, no write cache fiasco, much better situation. But it's still a desktop drive that's very likely to fail when used this way.
The first phone I had from Blackberry eventually had the scrollwheel jam rendering it useless.
The scroll ball on my Blackberry 8800 (through T-Mobile) broke in the on-position, rendering the phone completely useless, 5 days before the 1-year warranty wore out. Felt quite pleased, got a free upgrade to the current model instead (8830). That was 11 months ago. The scroll ball on the new one is starting to stick in the same way. My 2-year contract will be up in another month, and as much as I like the Blackberry service and the phones in general, it's hard to get too attached to the phone when it's got such an obvious design flaw to it.
In addition, Sun got MySQL's sales channel too. I've heard speculation that one thing Sun was hoping to learn from them was how to sell open-source software to people. Sun would love to have a better idea how to keep the free OpenSolaris going while still making money selling Solaris, but it's not so clear that they know how to do that well. That's something MySQL has done fairly well--the commercial versions are just different enough that people buy them, while still getting plenty of new users flowing in through the free stuff. Insights into that area are not as useful to Oracle though.
Account Settings/Offline & Disk Space controls this feature now. "Available for offline use" defaults to off, you have to toggle it on for all folders you want that behavior for.
The point is that had the variables been given useful names, what the idiom accomplished would have been obvious even to someone seeing it for the first time. For example, if we take that code:
for(ss = s->ss; ss; ss = ss->ss)
And rename its variables to not be retarded:
for(node = head->next; node; node = node->next)
Then not only would a newbie coder likely recognize what the code was doing, they'd have learned a useful C idiom through code reading even if they're new enough to that language to not have known it already.
Vinyl is the only format of the three that contains very high and very low frequency data that you cannot hear. You can't hear this data, but you can feel it, physically with your body.
It's silliness like this that gives high-end audio a bad name. I have a little gag I like to play on people who spew the whole "vinyl plays stuff that CD just can't" crowd. I have some records I meticulously transferred from vinyl to CD using a decent quality sound card. When people stop by to listen, I'll then fire that record up, then ask the vinyl fans what they think. The usual raves about how awesome vinyl is come out: "notice how you can feel it with your body" etc, the stuff you're spewing. Then comes the fun part: I lift the tonearm and the music keeps on going, because I was playing the CD version the whole time!
There's something about the sound of vinyl that is fun to listen to. Do not confuse this with technical superiority in any dimension; it's just distortion people seem to like. The vinyl record is incapable of even matching the frequency extremes possible with CD audio, and the idea that goes beyond them is indefensible. Whatever form of vinyl distortion you prefer is quite easy to capture onto digital, fully preserved. Shoot, the last time I ran into Michael Fremer at a show, he had some CD-Rs he'd made from a turntable that costs more than my car, and was demonstrating how close to the original they sounded. People were returning to the room trying to figure out when the turntable got installed it was so obviously great vinyl sound--easily stored at 16 bits/44.1Khz
Most of my serious listening is done on SACD or DVD-A; now that's real fidelity beyond what a CD is capable of, approaching the quality of the original analog master tape. I once sat in a room comparing such a master tape against a real-time 24/96K and 16/44.1K converted versions, single-blind. I had no problem identifying which was which. Guess what? Most of the people in the room preferred the CD version and presumed it was the original master tape as a result, even though to me it was easy to pick out as the worst of the lot. People like the kind of distortion they grew up with; used to be vinyl, then it was crappy CD mastering artifacts, and now we've moved onto people liking MP3 distortion.
The standard way to remove setuid requirements from ping is to implement the capabilities API, which was finally done in 2.6.26 even though the basic idea goes back to the 2.1 kernel in 1998. A good intro is available from IBM.
I ran into the same class of sound issues, in my case the primary one was trouble getting all of the output in the right mode. The sound card was convinced that the regular audio output was actually the coax one, and it was hard to figure out how to tell it otherwise (even though the problem and the solution were quite obvious). Just like old times, when everything broke after merging PulseAudio.
I was eventually able to dig into the sound issues using tools like alsamixer and manually tweaking what driver I was using to get things working again on the sacrificial test system. The contortions required made the new setup seemed really fragile, and I'm not sure exactly what fixed the issue. That means I might have to do this again after some future system update. While the Jaunty Sound Preferences panel was never elegant, it did at least work most of the time. As you point out, it looks like all of the GUI-based tools you used to be able to do troubleshooting and easily try alternative configs with are either gone or not in an obvious place anymore in Karmic. Given how problematic Linux sound has been over the years, it takes a very peculiar form of arrogance to presume it's finally fixed now all of the sudden, and therefore it's fine to seriously deprecate alternatives that (while not the preferred approach) were sometimes the only thing that did work in earlier releases.
Since there were some other really annoying bits in this release (the awful and so ubiquitous it's difficult to turn off new Notify OSD comes to mind), so far it looks like I'll be skipping this release. I skipped 7.10, 8.04, and the first few months of 8.10 due to quality control issues too, so this isn't that surprising. Ubuntu may put out a new release every six months, but I only seem to find one worth upgrading to every two years anyway. Seems pretty clear to me the 6-month release cycle is faster than Canonical and the community can really deliver stable software in. And that's regardless of LTS tagging, 8.04 was the worst of the bunch and its backported bug fixes were minimal for the problems I ran into; all the awful bugs were marked "fixed in Intrepid" and that was the end of it. I feel lucky that the LTS 9.04 release is the good one now, am hoping things work out similarly to how 7.04 kept me going for a long time before I needed to update. Of course, audio problems with Skype are still looming...
The way most application interactions happen, it's impossible for a standard filesystem to distinguish in advance between a series of writes that will update a small enough subset of the file that it should work, and ones where eventually enough blocks will be updated that it will run out of space. If you reject all possible writes that *could* run out of space eventually if the user keeps going, right from the start, that's not a good answer either. You'll still be in a position where you're rejecting possibly reasonable behavior, in a way that won't easily make sense to the user.
Slashdot Mirror
At one point tracking the entire .org domain was done with PG as well
Still is; even more than that actually, .info too. See Afilias Global Registry for a full list. Afilias sponsors quite a bit of PostgreSQL related work, including employing some of the major contributors to Slony, one of the most popular PostgreSQL replication solutions.
At which point NTT would promptly hire/promote someone else into the same position with the same level of resources necessary to be effective. For example, in addition to committer Itagaki Takahiro, they also have some major work on replication being led by Masao Fujii. The point of having a big company like NTT involved is that you can't just make their need for PostgreSQL to be successful go away so easily. There's not just "that one NTT employee"--he's one of a whole team there doing PostgreSQL related work.
I would like to have a list of serious companies using PostgreSQL for serious stuff
PostgreSQL Featured Users; Quotes has additional detail about the scope of some of those. Most people are probably familiar with names like Skype and Cisco on there, but less well known companies like NTT are huge too--and they even sponsor a good chunk of PostgreSQL development because it's so heavily used there.
And those are just the public record. Because of its BSD license, PostgreSQL also gets used in plenty of places that don't talk about what they're doing with it. For example, I've worked with financial companies that are cutting loose Oracle for PostgreSQL whenever feasible, and with some US defense companies that use PostGIS for geographic databases. (looks out window) I may have already said too much.
I just spent a few hours playing the NES version of MULE with a few friends while snowed in a few weeks ago. It was just as great as ever. Some of the games you mention really haven't aged well; this is not the case for MULE at all.
I was a little disappointed by some of the gameplay changes in this new version. They were so close to having a perfect remake, then changed some things that I shift the gameplay balance in a way I didn't think was an improvement. If you're going to try and remake a game this carefully balanced, you should do as good of a job as possible first, then consider adding optional changes. The same issue plagued the earlier Space HORSE remake too.
Quoting the reference I linked to and highlighting the important bit:
"You can change the SQL mode at runtime by using a SET [GLOBAL|SESSION] sql_mode='modes' statement to set the sql_mode system value. Setting the GLOBAL variable requires the SUPER privilege and affects the operation of all clients that connect from that time on. Setting the SESSION variable affects only the current client. Any client can change its own session sql_mode value at any time."
Yes, you have to be the superuser to change the global mode. And the fact that it says "Server SQL Modes", but they're not really enforced by the server even if you change the defaults should a client decide to go sloppy, is exactly why this behavior is so shady. It's an example of exactly why the upthread idea of PostgreSQL trying to be as correct as possible by design from day one is so important--you can't just bolt this stuff on later and expect there to not be a hole left behind.
You'll find more references for what you're talking about Why PostgreSQL Instead of MySQL, and the associated Transactional DDL page goes over how it compares with Oracle--which has recently added a feature to address this area, so picking on it in this area is a bit out of date now.
No problems with invalid data types if you bother to RTFM and setup the config properly. (It's not hard. Just turn strict mode on.)
Strict mode is a client setting, which means that you're always at the mercy of applications turning it off and inserting garbage. And there's the "IGNORE" keyword to allow that too. There is no way to make a MySQL server reject all incoming bad data.
The documentation is not GPL, a move defended quite weakly by Sun. That article does address one of the real reasons for what Monty is doing though: "MySQL forks should have their own, specific, documentation". Monty doesn't want to recreate all that from scratch, so forcing that into more open licensing would make his life easier.
There's a list of PostgreSQL hosting companies available. Last time I was picking one A2 and hub.org made my short list; there are plenty of others too.
Both U2 and the Stones evade their respective taxes through the Netherlands, making Bono quite the hypocrite. Details on the tax changes that motivated the switch out of Ireland.
Hint: supplement "infomercials" don't come with links to suggested reading material, including lists of relevant clinical studies, from sources like the NIH. I've read several of those, selected a product based on the best information available, and gave a quick summary of why I thought it was a good one. Your [citation needed] suggestion was to eat dirt for "the stuff you would have gotten from the food supply"--now that's unsupported infomercial talk.
I look forward to the future clinical results of your research, showing the positive impact of repopulating internal bacteria through dirt consumption. Until then I'll stay with the probiotics.
And what makes you so sure that your particular dirt will actually have useful bacteria in it? The problem here isn't just that people aren't consuming from the same sources they used to, it's that sometimes those things just don't have what they used to in them. There is a serious lack of evidence to your medical theory "modern dirt has everything you need". Whereas even conservative sources like the NIH have been compelled to note the growing evidence that probiotics are useful for digestion issues.
In any case, the situation around one of the specific bacteria mentioned in the article is much more controversial: h. pylori disappearing is not necessarily a bad thing. And given that the common transmission vectors are oral-oral or fecal-oral, if you wanted more, unless you're using dirt that people shit in regularly you'd be better off with some kissing instead.
Note that the GP's specific suggestion, the Jarrow Labs EPS, is crap. I'm taking Nature's Way Primadophilus Optima probiotics, which has almost an order of magnitude more bacteria when new and includes the food the little buggers live off of in the tablet too. >10 bacterial stains, >25M CFUs when new, and bundled FOS are the minimums I look for in a good quality probiotic. Nobody is quite sure which of the bacteria are responsible for what yet, so a shotgun approach works better than a focused one here.
The idea of using a USB drive for any sort of large storage gives me the willies for the reason you list, which is extremely important, as well as some others:
-USB bridges rarely pass any SMART data back to the host, which means that if it starts to fail you won't know until it's already lost data. How much data? You'll have no idea. Well, you could if you used ZFS or other checksumming, but even then you're not going to know what wrong until data is gone.
-Drives in the 1TB range are going to be regular internal drives with a USB bridge on them. Ever notice how external drives normally have a much shorter warranty on them than the internal versions of the same basic disk? That's because moving a desktop drive around is really bad for it. They just don't last very well in that situation, thus the derated warranty and lower expected lifetime.
-Drive failure rates are so bad, and the time it takes to make a full copy of a drive this big so long over USB, that you're way too exposed to losing two drives. If you care about your data at all, ideally you'd want a RAID-1 pair of internal drives in a server system. It's reasonable in that situation to make an off-site backup you sync periodically, but you don't want a portable drive to be your only primary copy.
Note that the first couple of these issues, the ones related to the crappy USB chipsets, can be avoided if you get an external drive that connects via eSATA. Then you get SMART, no write cache fiasco, much better situation. But it's still a desktop drive that's very likely to fail when used this way.
The first phone I had from Blackberry eventually had the scrollwheel jam rendering it useless.
The scroll ball on my Blackberry 8800 (through T-Mobile) broke in the on-position, rendering the phone completely useless, 5 days before the 1-year warranty wore out. Felt quite pleased, got a free upgrade to the current model instead (8830). That was 11 months ago. The scroll ball on the new one is starting to stick in the same way. My 2-year contract will be up in another month, and as much as I like the Blackberry service and the phones in general, it's hard to get too attached to the phone when it's got such an obvious design flaw to it.
In addition, Sun got MySQL's sales channel too. I've heard speculation that one thing Sun was hoping to learn from them was how to sell open-source software to people. Sun would love to have a better idea how to keep the free OpenSolaris going while still making money selling Solaris, but it's not so clear that they know how to do that well. That's something MySQL has done fairly well--the commercial versions are just different enough that people buy them, while still getting plenty of new users flowing in through the free stuff. Insights into that area are not as useful to Oracle though.
Or you could have ZFS where you don't even need to resize.. it just happens.
Right, except if you want to do something crazy like reduce pool capacity, which is impossible.
If they have a decent transactional database, they'd be keeping a record of everything that happened.
They don't
Account Settings/Offline & Disk Space controls this feature now. "Available for offline use" defaults to off, you have to toggle it on for all folders you want that behavior for.
OMG! I saw the same thing once! Only it turned out the prohibited item not allowed in the cabin was actually Megatron.
The point is that had the variables been given useful names, what the idiom accomplished would have been obvious even to someone seeing it for the first time. For example, if we take that code:
for(ss = s->ss; ss; ss = ss->ss)
And rename its variables to not be retarded:
for(node = head->next; node; node = node->next)
Then not only would a newbie coder likely recognize what the code was doing, they'd have learned a useful C idiom through code reading even if they're new enough to that language to not have known it already.
I'm not sure I can follow an analogy in this form; can you retell it in a way that revolves around the delivery guy's car?
Vinyl is the only format of the three that contains very high and very low frequency data that you cannot hear. You can't hear this data, but you can feel it, physically with your body.
It's silliness like this that gives high-end audio a bad name. I have a little gag I like to play on people who spew the whole "vinyl plays stuff that CD just can't" crowd. I have some records I meticulously transferred from vinyl to CD using a decent quality sound card. When people stop by to listen, I'll then fire that record up, then ask the vinyl fans what they think. The usual raves about how awesome vinyl is come out: "notice how you can feel it with your body" etc, the stuff you're spewing. Then comes the fun part: I lift the tonearm and the music keeps on going, because I was playing the CD version the whole time!
There's something about the sound of vinyl that is fun to listen to. Do not confuse this with technical superiority in any dimension; it's just distortion people seem to like. The vinyl record is incapable of even matching the frequency extremes possible with CD audio, and the idea that goes beyond them is indefensible. Whatever form of vinyl distortion you prefer is quite easy to capture onto digital, fully preserved. Shoot, the last time I ran into Michael Fremer at a show, he had some CD-Rs he'd made from a turntable that costs more than my car, and was demonstrating how close to the original they sounded. People were returning to the room trying to figure out when the turntable got installed it was so obviously great vinyl sound--easily stored at 16 bits/44.1Khz
Most of my serious listening is done on SACD or DVD-A; now that's real fidelity beyond what a CD is capable of, approaching the quality of the original analog master tape. I once sat in a room comparing such a master tape against a real-time 24/96K and 16/44.1K converted versions, single-blind. I had no problem identifying which was which. Guess what? Most of the people in the room preferred the CD version and presumed it was the original master tape as a result, even though to me it was easy to pick out as the worst of the lot. People like the kind of distortion they grew up with; used to be vinyl, then it was crappy CD mastering artifacts, and now we've moved onto people liking MP3 distortion.
The standard way to remove setuid requirements from ping is to implement the capabilities API, which was finally done in 2.6.26 even though the basic idea goes back to the 2.1 kernel in 1998. A good intro is available from IBM.
I ran into the same class of sound issues, in my case the primary one was trouble getting all of the output in the right mode. The sound card was convinced that the regular audio output was actually the coax one, and it was hard to figure out how to tell it otherwise (even though the problem and the solution were quite obvious). Just like old times, when everything broke after merging PulseAudio.
I was eventually able to dig into the sound issues using tools like alsamixer and manually tweaking what driver I was using to get things working again on the sacrificial test system. The contortions required made the new setup seemed really fragile, and I'm not sure exactly what fixed the issue. That means I might have to do this again after some future system update. While the Jaunty Sound Preferences panel was never elegant, it did at least work most of the time. As you point out, it looks like all of the GUI-based tools you used to be able to do troubleshooting and easily try alternative configs with are either gone or not in an obvious place anymore in Karmic. Given how problematic Linux sound has been over the years, it takes a very peculiar form of arrogance to presume it's finally fixed now all of the sudden, and therefore it's fine to seriously deprecate alternatives that (while not the preferred approach) were sometimes the only thing that did work in earlier releases.
Since there were some other really annoying bits in this release (the awful and so ubiquitous it's difficult to turn off new Notify OSD comes to mind), so far it looks like I'll be skipping this release. I skipped 7.10, 8.04, and the first few months of 8.10 due to quality control issues too, so this isn't that surprising. Ubuntu may put out a new release every six months, but I only seem to find one worth upgrading to every two years anyway. Seems pretty clear to me the 6-month release cycle is faster than Canonical and the community can really deliver stable software in. And that's regardless of LTS tagging, 8.04 was the worst of the bunch and its backported bug fixes were minimal for the problems I ran into; all the awful bugs were marked "fixed in Intrepid" and that was the end of it. I feel lucky that the LTS 9.04 release is the good one now, am hoping things work out similarly to how 7.04 kept me going for a long time before I needed to update. Of course, audio problems with Skype are still looming...
The way most application interactions happen, it's impossible for a standard filesystem to distinguish in advance between a series of writes that will update a small enough subset of the file that it should work, and ones where eventually enough blocks will be updated that it will run out of space. If you reject all possible writes that *could* run out of space eventually if the user keeps going, right from the start, that's not a good answer either. You'll still be in a position where you're rejecting possibly reasonable behavior, in a way that won't easily make sense to the user.