I always thought that GPU + physics engine would be a perfect combination. Ultimately, the AGEIA card is just a DSP + software driver for calculating physics. A GPU is... also a DSP + software driver for calculating graphics. It wouldn't be too hard to write a driver that does both: some of the pipelines could be allocated to graphics, and some to physics. Might even make a software-configurable to dedicate more/less units to physics.
SQL already allows you to drop to READ_UNCOMMITTED if you really really want to -- though the DB actually under no obligation to drop to that level, you're just specifying that you don't care. That removes A, C, and I all at once. Why not make the D part optional too?
False. SQL standard explicitly specifies that writing to the database under READ UNCOMMITTED isolation is not allowed. You can only do read-only queries. Further, PostgreSQL doesn't even support READ UNCOMMITTED. There is no need for it. PostgreSQL implements MVCC such that each transaction gets a private snapshot of the database. With that you get READ COMMITTED for free.
I'm with the original poster here. Asynchronous transactions seem like a bad idea. But then it's not PostgreSQL's responsibility to enforce good software design. And maybe in some corner cases people can find use for them.
Why do they use sodium to cool reactors? It reacts violently when exposed to water producing highly corrosive NaOH. Why would you want to use that stuff? Seems like a really bad idea.
Excellent comments about MySQL. You've summarized my thoughts exactly. However, you've really dropped the ball by not moving to PostgreSQL. It's really a shame that such a gem gets left in the dust because people simply don't know any better. It's either MySQL on low end or some proprietary DB on high-end. But the truth is that PostgreSQL can do anything that MSSQL can, and it actually does a few things better. For example:
* If you start a transaction in PostgreSQL and one of the SQL statements causes an error, the transaction is aborted immediately, and all resources are immediately released. The server still expects the client to send "ROLLBACK", but that is merely an acknowledgment. Not so with MSSQL. It allows you to continue writing to the database, AND issue a COMMIT statement at the end, thus committing a partially-failed transaction! This is from the WTF??? department.
* For whatever reason, the default collation is not case-sensitive. This caused some interesting bugs when we least expected.
* Client libraries are just pathetic, especially on non-windows platforms. JDBC driver is usable, but C client library is essentially reverse-engineered and not supported by microsoft (http://www.freetds.org/). It is a small miracle we can access MSSQL at all from Linux.
Anyway, I've used PostgreSQL in my past few projects and currently using MSSQL (not my choice). It is a decent database (other than the annoyances I mentioned), but I still prefer PostgreSQL. Feature-wise, I can't think of anything that MSSQL has and PostgreSQL doesn't. It's really a shame it doesn't get used more.
There is no doubt that right now you have a much better product. However, Microsoft can afford to give away their product for free [or rather, subsidize it with their monopoly profits from other software]. You can't. And ultimately, they don't need to build a better product. They just need a product that is good enough so that they can leverage their monopoly to "cut off your air supply". For instance, does "lower price for Windows Vista used on virtualized computers" apply to microsoft VPC only or to all hypervisors? In your position, I would start worrying. It's not the first time microsoft killed off a superior product through underhanded tactics.
Which is simultaneously a great strength and great weakness. Why will people end up using the Microsoft stack to host Linux VMs? Because everything has a windows driver.
Does microsoft hypervisor even support non-microsoft OSes? I seem to recall that they specifically disabled that functionality when they bought VPC from Conntectix.
Probably the same people who think vaccination is eeevil. But then, if you get your health information from youtube, you deserve the darwin award. So I see no problem.
Does anyone still doubt that Miguel de Icaza is a microsoft shill? I think it's time for him to join microsoft and work on their "open source" strategy. Hey, it worked out for Bill Hilf...
This license governs use of the accompanying software. If you use the software, you accept this license. If you do not accept the license, do not use the software.
This is contrary to any Open Source license I know of. The whole point of Open Source is that you can use the software in any way you want. You have to agree to the license only when you distribute. Microsoft is attempting to subvert OSI, just like it has already subverted ISO.
SCOX has just filed this motion. Novell gets 15 days to reply, which puts the deadline in the second week of september. Then SCOX gets 7 days to reply to Novell. But the trial is scheduled to begin on september 17. The easiest thing for the judge to do is deny SCOX's motion as moot and say "don't worry, you'll get your final judgment soon enough".
I'm surprised it has not been covered on slashdot, but similar things have occured in Germany, Switzerland, Norway, Portugal, Australia, etc. Microsoft is determined to push its proprietary "open" format through by any means neccessary:
I think the silly MS license has the same sort of logical error in it. It has boilerplate language that says it applies to use, but it places no restrictions on use. If it's not free, then GTK+ is not free.
There is a difference. You get GPL/LGPL "EULA" because of brain-dead installers that assume there must be EULA, and/or people who write the install scripts. However, the license itself explicitly states that you do *not* need to accept it merely to use the software. Microsoft's "license" explicitly states exactly the opposite. And while MS-PL does not actually restrict use, MS-LPL absolutely does. Therefore, MS-PL is a trojan horse: it's purpose is to make people accept the idea that controlling how the supposedly "open source" software is used is ok. I do not believe this is a logic error, as you say. I believe it's intentional.
Microsoft "permissive license" attempts to control the mere use of the software:
This license governs use of the accompanying software. If you use the software, you accept this license. If you do not accept the license, do not use the software.
So it is neither a "license" nor "permissive". It is unilateral contract, no different than click-through EULA.
In contrast, Free software licenses (BSD, MIT, GPL, etc.) cover only the distribution of the software. You do not need to accept any "license" just to use the software. For example, here the relevant paragraph from GPL:
You are not required to accept this License in order to receive or
run a copy of the Program. Ancillary propagation of a covered work
occurring solely as a consequence of using peer-to-peer transmission
to receive a copy likewise does not require acceptance. However,
nothing other than this License grants you permission to propagate or
modify any covered work. These actions infringe copyright if you do
not accept this License. Therefore, by modifying or propagating a
covered work, you indicate your acceptance of this License to do so.
So Free software licenses are indeed licenses: i.e. they grant you more rights than what you get by default under copyright law. EULAs, including microsoft's "permissive license" attempt to restrict your rights by controlling how you can use the software.
So it is difficult to see microsoft's "permissive license" as anything but a trojan horse. Especially since it has an uglier brother, the "limited permissive license", which sounds confusingly similar to "permissive license", but adds a completely ridiculous restriction: you can only run the software on windows.
How is it "net neutrality" if you have to pay extra lest your packets be lost? Sounds more like extortion to me. (or precisely the big telco version of "net neutrality").
I also fail to see how the ISP can "guarantee" net neutrality. They can do nothing if their upstream provider decides to throttle some sites.
Slashdot Mirror
I always thought that GPU + physics engine would be a perfect combination. Ultimately, the AGEIA card is just a DSP + software driver for calculating physics. A GPU is... also a DSP + software driver for calculating graphics. It wouldn't be too hard to write a driver that does both: some of the pipelines could be allocated to graphics, and some to physics. Might even make a software-configurable to dedicate more/less units to physics.
I'm with the original poster here. Asynchronous transactions seem like a bad idea. But then it's not PostgreSQL's responsibility to enforce good software design. And maybe in some corner cases people can find use for them.
BSA is not the government. What gives them the right to bust into your office and demand fines?
Why do they use sodium to cool reactors? It reacts violently when exposed to water producing highly corrosive NaOH. Why would you want to use that stuff? Seems like a really bad idea.
Excellent comments about MySQL. You've summarized my thoughts exactly. However, you've really dropped the ball by not moving to PostgreSQL. It's really a shame that such a gem gets left in the dust because people simply don't know any better. It's either MySQL on low end or some proprietary DB on high-end. But the truth is that PostgreSQL can do anything that MSSQL can, and it actually does a few things better. For example:
* If you start a transaction in PostgreSQL and one of the SQL statements causes an error, the transaction is aborted immediately, and all resources are immediately released. The server still expects the client to send "ROLLBACK", but that is merely an acknowledgment. Not so with MSSQL. It allows you to continue writing to the database, AND issue a COMMIT statement at the end, thus committing a partially-failed transaction! This is from the WTF??? department.
* For whatever reason, the default collation is not case-sensitive. This caused some interesting bugs when we least expected.
* Client libraries are just pathetic, especially on non-windows platforms. JDBC driver is usable, but C client library is essentially reverse-engineered and not supported by microsoft (http://www.freetds.org/). It is a small miracle we can access MSSQL at all from Linux.
Anyway, I've used PostgreSQL in my past few projects and currently using MSSQL (not my choice). It is a decent database (other than the annoyances I mentioned), but I still prefer PostgreSQL. Feature-wise, I can't think of anything that MSSQL has and PostgreSQL doesn't. It's really a shame it doesn't get used more.
Natales,
There is no doubt that right now you have a much better product. However, Microsoft can afford to give away their product for free [or rather, subsidize it with their monopoly profits from other software]. You can't. And ultimately, they don't need to build a better product. They just need a product that is good enough so that they can leverage their monopoly to "cut off your air supply". For instance, does "lower price for Windows Vista used on virtualized computers" apply to microsoft VPC only or to all hypervisors? In your position, I would start worrying. It's not the first time microsoft killed off a superior product through underhanded tactics.
Cause I can summarize the current administration with one word: honest.
Probably the same people who think vaccination is eeevil. But then, if you get your health information from youtube, you deserve the darwin award. So I see no problem.
Our slap-on-the-wrist "punishment" has done jack-shit for 5 years. I know! Let's extended for another 5 years! That ought to do it!
"assus maximus" = "huge ass".
Does anyone still doubt that Miguel de Icaza is a microsoft shill? I think it's time for him to join microsoft and work on their "open source" strategy. Hey, it worked out for Bill Hilf...
Let me guess: people are civilized?
This is contrary to any Open Source license I know of. The whole point of Open Source is that you can use the software in any way you want. You have to agree to the license only when you distribute. Microsoft is attempting to subvert OSI, just like it has already subverted ISO.
They'll just add 5 more dimensions.
SCOX has just filed this motion. Novell gets 15 days to reply, which puts the deadline in the second week of september. Then SCOX gets 7 days to reply to Novell. But the trial is scheduled to begin on september 17. The easiest thing for the judge to do is deny SCOX's motion as moot and say "don't worry, you'll get your final judgment soon enough".
I'm surprised it has not been covered on slashdot, but similar things have occured in Germany, Switzerland, Norway, Portugal, Australia, etc. Microsoft is determined to push its proprietary "open" format through by any means neccessary:
1 23112581
1 25524759
2 35113424
http://www.groklaw.net/article.php?story=20070824
http://www.groklaw.net/article.php?story=20070815
http://www.groklaw.net/article.php?story=20070723
There is a difference. You get GPL/LGPL "EULA" because of brain-dead installers that assume there must be EULA, and/or people who write the install scripts. However, the license itself explicitly states that you do *not* need to accept it merely to use the software. Microsoft's "license" explicitly states exactly the opposite. And while MS-PL does not actually restrict use, MS-LPL absolutely does. Therefore, MS-PL is a trojan horse: it's purpose is to make people accept the idea that controlling how the supposedly "open source" software is used is ok. I do not believe this is a logic error, as you say. I believe it's intentional.
In contrast, Free software licenses (BSD, MIT, GPL, etc.) cover only the distribution of the software. You do not need to accept any "license" just to use the software. For example, here the relevant paragraph from GPL:
So Free software licenses are indeed licenses: i.e. they grant you more rights than what you get by default under copyright law. EULAs, including microsoft's "permissive license" attempt to restrict your rights by controlling how you can use the software.So it is difficult to see microsoft's "permissive license" as anything but a trojan horse. Especially since it has an uglier brother, the "limited permissive license", which sounds confusingly similar to "permissive license", but adds a completely ridiculous restriction: you can only run the software on windows.
How is it "net neutrality" if you have to pay extra lest your packets be lost? Sounds more like extortion to me. (or precisely the big telco version of "net neutrality").
I also fail to see how the ISP can "guarantee" net neutrality. They can do nothing if their upstream provider decides to throttle some sites.
Yeah. Too bad you elected the guy who does just that.
slashdot stops posting dupes. No wait...
Never heard of this guy before. What I know now is:
... wait, I know! He is Bender!
- he bends metal
- he is annoying
- he is con-artist
Uhhhm what syntax? ANSI SQL? Can you actually provide any examples, cause what you're saying doesn't make any sense.
I guess you never actually used postgres. Or oracle.