The challenge to port an Oracle application to PostgreSQL is much less than to port an Oracle application to MySQL. Particularly in the training department, since MySQL is the most unlike Oracle.
Also, the application matters a lot. MySQL is very effective as a cache to hold a relation. It would not surprise me if many of those companies use Oracle/DB2/MSSQL/PostgreSQL as a backend database, and then use MySQL to cache some of the data for fast access. If you list the companies using PostgreSQL extensively, they are likely to be using PostgreSQL as a replacement for the likes of Oracle. If you list the companies using MySQL, that's probably not the case, it's more likely that they're using it as a complement to Oracle or for a purpose that you normally shouldn't use Oracle for.
I would say out of the relational databases, MySQL is the most different in terms of application domain, functionality, performance, and behavior.
Your first example is bad, as the determination of what constitutes cruel and unusual punishment is a Constitutional issue to be decided by the Supreme Court.
Well, they decided wrong. It makes no sense that something cruel and unusual magically turns into humane punishment when someone is one day older at the time the crime is committed.
The SC didn't turn to the text of the Constitution for answers, instead they cited "an emerging consensus" and referenced the policies of foreign nations -- HUH?!?. Shouldn't a "consensus" be decided by the legislature? Also, out of the states that permit the death penalty, a majority previously allowed minors to be executed.
This is a clear example of the federal government trampling states' rights. The SC is the worst offender, IMO. They often "interpret" the consitution to mean directly the opposite of what it actually says. Abortion should have been a legislative issue, and the 10th Amendment says it should be at the state level.
The feds take a large amount of taxes from everyone already, so there's no hope of the states supporting their own road system.
Reduce it to the following situation and then recosider your statement:
The feds take the states' citizens' money. The feds offer to give it back if they say "how high" when the feds say jump.
It's not like the state can say "OK, we'll pay you that much less in taxes then.". If Cali opts out, all the other states are basically just confiscating Californians' money.
Is Scott Peterson "guilty beyond a reasonable doubt"? Hell no,...
I don't mean to retry the case on/., but here's your "reasonable doubt" scenario:
Let's assume he was innocent. He is the last person to see her alive aside from the murderer(s). He had recently been making concrete anchors or something for his boat. He goes boating in the SF bay on Christmas eve. She disappears. Scott is asking the police about grief counseling before the search has even begun. He's having an affair and Amber doesn't know he's married. He tells Amber that his wife is dead before she actually dies. He tells Amber something about how he didn't do it, but he knows who did (or some such weird, ominous thing like that). Scott rents a car and drives 80 miles to look at the SF bay. Then Laci turns up in the SF bay. Then Scott goes down to San Diego and has tools and $10k in cash with him.
That's less evidence than the jury heard, I'm sure, but already that's not a very reasonable doubt to me. I just don't buy that story, and neither did the jury. A series of super-coincidences is not a "reasonable doubt".
That's a good point. I think he was probably just trying to generate discussion, but he should have asked a real question that generates more than a "visit this link" respone.
People like to throw numbers like that around but they just don't make sense.
Computers were useful when companies started using them, otherwise they wouldn't have started. It may have taken a while to make a secratary more productive, but there were real business applications before that time. Some may have been an investment in the future, but I doubt it took 20 years for anyone to be more productive.
Replication is a complicated issue. I would highly recommend that you actually test the replication to make sure it's doing exactly what you think it is.
Also, what about PgPool? That is a popular form of master-master replication for postgresql. It's not the be-all-end-all, but no single replication system is right for all situations. That's why PostgreSQL has so many replication options.
I would be very interested to read a case study of your master-master replication usage in MySQL. I understand that the master-master replication is only RAM-based and doesn't support many of the features in other table types. Perhaps you could set up a similar system using PostgreSQL/PgPool with fsync off and compare the results?
The following is a rant about replication, and not necessarily directed at you: The thing that you have to have in mind is that replication is inherently imperfect. If two systems are up at two different sites, but the WAN is having problems, the two databases cannot be kept in perfect sync. A simple example of a failure of two-phase-commit (2PC) is when both servers agree that they're ready to commit, and the first one does, and the second one never responds. It's just imperfect. So, when setting up a replication system it's largely about what your configuration is, and how you plan to handle the problems that you can't eliminate. ORAC does this by using sophisticated conflict resolution algorithms, which are also imperfect, but help their clients resolve the inconsistancies that happen. You really have to have a series of replication systems in place, and you have to have a plan for all of the problems that can occur, because there's no way to eliminate the possibility.
And when a marketroid starts talking as if their solution is the be-all-end-all without examining your details specifically, you know they're BSing you and you have no reason to think that you're safe.
If you're running tables that big and don't reading the MySQL manual or don't know how to google "myisam file size limit", you have bigger problems.
I had a legitimate question; that kind of response was unecessary. I am a PostgreSQL user, but I try to keep knowledge of other databases handy so I can avoid pitfalls if I need to use another DB.
2GB isn't a big table really. What is that, like $2 worth of disk? If I had set up an archive or log or something a year ago, it would probably have bitten me by now if I used MySQL and there was indeed a default limit of 2GB.
Also, software manuals are, unfortunately, not a great place to learn about pitfalls. Users are a much better place.
OSes generally support files much larger than 2GB. It's usually the filesystem that's a limiting factor. However, if you use Reiserfs v3/4 or XFS or JFS, it is no problem at all.
His question was about big name companies using PostgreSQL so that his boss could sleep at night. For anyone to make anything close to a reasonable suggestion, he would have to give us a huge amount of information, which is unreasonable for the ask-/. format. In short, his question was not a technical one.
You're right though, different situations drastically change the requirements. However, it looks like he's already made that decision himself, otherwise he would be trying to convince himself and not his boss.
Just because your company can spend hundreds of thousands of dollars (or millions for a large installation) on something that's really orthogonal to the actual business that your company is in, doesn't mean you should.
If I was a PHB type for an online retailer and I looked at the costs and noticed that 50% of our profits are going to Oracle rather than to our pockets, I'd have some questions for the IT guys like:
(1) Are we a retailer or a data warehousing company? (2) What is Oracle and why is it so expensive? (3) Can you get the same job done with less money? If so, what costs, benefits, and risks might we see? (4) My friend's IT guys use this thing called Post-whatever-SQL, and it costs $0. Is Oracle kinda like that?
By configuration or recompilation? Do you need to dump/reload or will it work with the old data files? What OSes and filesystems have what limitations?
I want to draw attention to this interview. This is a great interview that has good answers to the original question.
In particular, Josh talks about Fujitsu's involvement (a $43 billion company).
A good quote:
Josh Berkus: Basically, the head of their open source business applications division, Mr. Takayuki Nakazawa, said, "We are committed to helping make PostgreSQL the leading database management system."
That's what I thought at first, but after I learned how to do it I though to myself: "Can I come up with an easier way to administer PITR?" and I couldn't think of anything. PITR is a complicated concept (time warping, multiple timelines; it starts to get a little weird), and I'm impressed that they are supporting the feature. If you can think of an easier way to administer it, let the lists know, and I wouldn't be surprised if some tools appeared.
First off, I think he's less worried about PostgreSQL's ability and more worried about making his boss comfortable with the idea. Others have already pointed out some case studies, so that's a good start.
To say that you're jumping the gun in your post would be an understatement. Read only databases aren't hard; there are many ways to accomplish that and with MySQL you would still have to convince your boss to use F/OSS. I hardly see how that answers his original question, it seems more like you're trying to convince someone that PostgreSQL is a bad choice (which seems strange considering that you don't know anything about his setup).
Also, the one thing that he did say is that he has a preexisting database in Oracle. It's much easier to go Oracle->PostgreSQL that Oracle->MySQL, but that may not be his primary concern.
Internet communities can be bad as well as good. The last thing you want is to support a vile, disgusting community like that.
Even if no money is being exchanged, many people will no doubt be contributing to the community, and that's why you don't want this community at all.
I feel bad for the creators of freenet. They wanted to create a free network, but it only caught on among the worst kinds of people. I had no idea that this bad stuff was so prevalent on the freenet network. Because of that, I will not run a freenet node.
Actually "juvenile diabetes" is an accurate name. I was diabetic for at least 10 years before I was diagnosed at 12-13.
Just because 90% of doctors don't care about patients at ALL and are so dense they can't diagnose a common condition with unique, easily testable symptoms before the patient is an adult doesn't mean you should change the name of the disease.
<rant> If you couldn't tell, I'm bitter. My mom thought I was diabetic the whole time and I went to many many doctors (including a special childrens' hospital in Oakland when I was 4) and I wasn't diagnosed until I was 12-13. If you don't have a doctor in your family (I don't), then you are really screwed. Really, really screwed. No matter how much you pay them, doctors just don't care unless you're going to die before the appointment is up.
That's what you get when you bestow titles on people I guess. Screw the AMA, let non-AMA people practice legally. </rant>
The new technology they seem to be promising does have advantages for diabetics. All the problems with the site at which the pump injects the insulin basically go away. No longer does it have to be moved every few days, and no longer will it leave you screwed when the site accidently gets ripped out or clogged and you can't fix it for a couple hours. Also, if the site ever does get "saturated" (where the insulin becomes decreasingly effective due to so much of it coming in at that one site), it's really easy to move. In fact, it could have many different sites, increasing the absorption rate to near-nondiabetic speed.
Also, it no longer requires the insulin to be forced through tubes, which could mean a lot less insulin waste, and a lot more practical thing to carry around.
Others have already pointed out how valuable research is, so I'll ignore that for now.
Hurd is also attempting to solve very real practical problems. Consider a typical UNIX network daemon:
(1) Must be started as root to listen on a privileged port (2) Upon an incoming connection, must accept and then "drop privileges"
This causes mnay, MANY problems, that are very practical. If you took all the remote root security exploits ever in UNIX, and you subtracted those that involved a network daemon that needed to be run as root to listen on a privileged port, you'd be left with a rather secure system.
I just can't imagine why nobody recognizes a problem like that. You don't inherently need to run Apache/BIND/Sendmail with the privilege to overwrite the boot sector, but people ignore it as if "Oh well, it's a network daemon, of course it needs to be able to rewrite the boot sector. We'll just hope there are no bugs.".
Not only is this a security nightmare (which is only mitigated by the fact that UNIX is compared to windows rather than an ideal), but it's also got many performance implications. If you're measuring raw performance of already-written applications, a monolithic kernel will never be worse than a microkernel architecture. However, on a linux system, a lot of resources are traded in order to jump through hoops that don't need to be there, particularly for security. Maybe you don't really need to start that new process, and you can do everything you need in the current process. Sounds like a serious performance win to me, and not "my kernel avoids that 0.1% performance penalty you have to take on operation X".
For example, what about Apache, CGI and suexec? You really don't need all that when you could just be getting/releasing authentication tokens and using an apache module rather than starting a new process just so you can change privileges.
You can't separate the details of performance characteristics from capabilities. Capabilities may cost overhead, but may reduce algorithmic requirements.
* an employer * recently dealt with an employee who wrote derivative code from IBM Devloper Works * intended to patent the code the employee wrote * intended to distribute the code the employee wrote under a non-GPL license * intended to claim ownership over all kinds of things that the employee did on time that wasn't directly compensated for
I would be pretty damn suspicious that this story was about me. Each of these things individually doesn't mean a whole lot probably, but all of them combined, along with the proper time frame, would pretty much narrow it down to one person in the world. Maybe it wouldn't, but it would be highly coincidental. Then just think to yourself: "Might my employee use an alias like 'Daimaou'?". It couldn't be that hard to figure out. Just look at some http requests when the employee visits/.
Can't someone do something dumb without people mulling over it many years after they're dead? Not only that, friends don't walk friends into open manholes.
because of that government oversight, very little sale of human organs happen
First off, the moral issue (for me anyway) is not whether someone sells something, the moral issue is whether there is an incentive to make organs available where they wouldn't have been available before.
Even though there is huge demand for organs, people in the United States aren't murdering eachother to sell organs, nor are they raising new humans for the purpose of organ harvesting. It's not realistic to expect that will happen soon, either, in part because the people with the organs are likely to complain before they are harvested.
However, it is realistic to think that human embryos will be created for the purpose of harvesting. An embryo can't complain.
And, no, with government funded research, often no one makes any money.
Then why need funding? The government funding goes to someone. People want to be funded for research if they are not already, and people want more funding if they are already funded. "Profit" is mostly just an accounting term to recognize when something is a good investment and when it's not. It's not equivalent to making money. Lots of people make money from government funds, and it doesn't matter that it's not called "profit".
What amazes me is that the argument for government-funded embryonic stem cell research seems to be that it is not morally questionable. But the mere fact that we're in disagreement means that it is morally questionable. So, how about we not take people's money if they are morally opposed to the research anyway? Obviously, a significant portion of the U.S. is against it.
Slashdot Mirror
The challenge to port an Oracle application to PostgreSQL is much less than to port an Oracle application to MySQL. Particularly in the training department, since MySQL is the most unlike Oracle.
Also, the application matters a lot. MySQL is very effective as a cache to hold a relation. It would not surprise me if many of those companies use Oracle/DB2/MSSQL/PostgreSQL as a backend database, and then use MySQL to cache some of the data for fast access. If you list the companies using PostgreSQL extensively, they are likely to be using PostgreSQL as a replacement for the likes of Oracle. If you list the companies using MySQL, that's probably not the case, it's more likely that they're using it as a complement to Oracle or for a purpose that you normally shouldn't use Oracle for.
I would say out of the relational databases, MySQL is the most different in terms of application domain, functionality, performance, and behavior.
Your first example is bad, as the determination of what constitutes cruel and unusual punishment is a Constitutional issue to be decided by the Supreme Court.
Well, they decided wrong. It makes no sense that something cruel and unusual magically turns into humane punishment when someone is one day older at the time the crime is committed.
The SC didn't turn to the text of the Constitution for answers, instead they cited "an emerging consensus" and referenced the policies of foreign nations -- HUH?!? . Shouldn't a "consensus" be decided by the legislature? Also, out of the states that permit the death penalty, a majority previously allowed minors to be executed.
This is a clear example of the federal government trampling states' rights. The SC is the worst offender, IMO. They often "interpret" the consitution to mean directly the opposite of what it actually says. Abortion should have been a legislative issue, and the 10th Amendment says it should be at the state level.
I think that was his point...
The feds take a large amount of taxes from everyone already, so there's no hope of the states supporting their own road system.
Reduce it to the following situation and then recosider your statement:
The feds take the states' citizens' money.
The feds offer to give it back if they say "how high" when the feds say jump.
It's not like the state can say "OK, we'll pay you that much less in taxes then.". If Cali opts out, all the other states are basically just confiscating Californians' money.
Is Scott Peterson "guilty beyond a reasonable doubt"? Hell no, ...
/., but here's your "reasonable doubt" scenario:
I don't mean to retry the case on
Let's assume he was innocent. He is the last person to see her alive aside from the murderer(s). He had recently been making concrete anchors or something for his boat. He goes boating in the SF bay on Christmas eve. She disappears. Scott is asking the police about grief counseling before the search has even begun. He's having an affair and Amber doesn't know he's married. He tells Amber that his wife is dead before she actually dies. He tells Amber something about how he didn't do it, but he knows who did (or some such weird, ominous thing like that). Scott rents a car and drives 80 miles to look at the SF bay. Then Laci turns up in the SF bay. Then Scott goes down to San Diego and has tools and $10k in cash with him.
That's less evidence than the jury heard, I'm sure, but already that's not a very reasonable doubt to me. I just don't buy that story, and neither did the jury. A series of super-coincidences is not a "reasonable doubt".
Yeah, I realize that Fujitsu is way more than a software company. I worded my previous post poorly.
That's a good point. I think he was probably just trying to generate discussion, but he should have asked a real question that generates more than a "visit this link" respone.
Interesting interpretation.
I thought that Oracle was the second biggest software company out there. I didn't know whether that put them in the $40 billion range or not.
I'd be interested to read a case study if you're willing to write one.
Just to make sure, you didn't leave the postgresql shared_buffers setting default did you?
Which 20 years are Hemos referring to?
Yeah, as soon as I read that claim I called BS.
People like to throw numbers like that around but they just don't make sense.
Computers were useful when companies started using them, otherwise they wouldn't have started. It may have taken a while to make a secratary more productive, but there were real business applications before that time. Some may have been an investment in the future, but I doubt it took 20 years for anyone to be more productive.
Replication is a complicated issue. I would highly recommend that you actually test the replication to make sure it's doing exactly what you think it is.
Also, what about PgPool? That is a popular form of master-master replication for postgresql. It's not the be-all-end-all, but no single replication system is right for all situations. That's why PostgreSQL has so many replication options.
I would be very interested to read a case study of your master-master replication usage in MySQL. I understand that the master-master replication is only RAM-based and doesn't support many of the features in other table types. Perhaps you could set up a similar system using PostgreSQL/PgPool with fsync off and compare the results?
The following is a rant about replication, and not necessarily directed at you:
The thing that you have to have in mind is that replication is inherently imperfect. If two systems are up at two different sites, but the WAN is having problems, the two databases cannot be kept in perfect sync. A simple example of a failure of two-phase-commit (2PC) is when both servers agree that they're ready to commit, and the first one does, and the second one never responds. It's just imperfect. So, when setting up a replication system it's largely about what your configuration is, and how you plan to handle the problems that you can't eliminate. ORAC does this by using sophisticated conflict resolution algorithms, which are also imperfect, but help their clients resolve the inconsistancies that happen. You really have to have a series of replication systems in place, and you have to have a plan for all of the problems that can occur, because there's no way to eliminate the possibility.
And when a marketroid starts talking as if their solution is the be-all-end-all without examining your details specifically, you know they're BSing you and you have no reason to think that you're safe.
If you're running tables that big and don't reading the MySQL manual or don't know how to google "myisam file size limit", you have bigger problems.
I had a legitimate question; that kind of response was unecessary. I am a PostgreSQL user, but I try to keep knowledge of other databases handy so I can avoid pitfalls if I need to use another DB.
2GB isn't a big table really. What is that, like $2 worth of disk? If I had set up an archive or log or something a year ago, it would probably have bitten me by now if I used MySQL and there was indeed a default limit of 2GB.
Also, software manuals are, unfortunately, not a great place to learn about pitfalls. Users are a much better place.
OSes generally support files much larger than 2GB. It's usually the filesystem that's a limiting factor. However, if you use Reiserfs v3/4 or XFS or JFS, it is no problem at all.
His question was about big name companies using PostgreSQL so that his boss could sleep at night. For anyone to make anything close to a reasonable suggestion, he would have to give us a huge amount of information, which is unreasonable for the ask-/. format. In short, his question was not a technical one.
You're right though, different situations drastically change the requirements. However, it looks like he's already made that decision himself, otherwise he would be trying to convince himself and not his boss.
Just because your company can spend hundreds of thousands of dollars (or millions for a large installation) on something that's really orthogonal to the actual business that your company is in, doesn't mean you should.
If I was a PHB type for an online retailer and I looked at the costs and noticed that 50% of our profits are going to Oracle rather than to our pockets, I'd have some questions for the IT guys like:
(1) Are we a retailer or a data warehousing company?
(2) What is Oracle and why is it so expensive?
(3) Can you get the same job done with less money? If so, what costs, benefits, and risks might we see?
(4) My friend's IT guys use this thing called Post-whatever-SQL, and it costs $0. Is Oracle kinda like that?
it can be increased if needed.
By configuration or recompilation? Do you need to dump/reload or will it work with the old data files? What OSes and filesystems have what limitations?
In particular, Josh talks about Fujitsu's involvement (a $43 billion company).
A good quote:
Although PITR in pgsql is something of a PITA
That's what I thought at first, but after I learned how to do it I though to myself: "Can I come up with an easier way to administer PITR?" and I couldn't think of anything. PITR is a complicated concept (time warping, multiple timelines; it starts to get a little weird), and I'm impressed that they are supporting the feature. If you can think of an easier way to administer it, let the lists know, and I wouldn't be surprised if some tools appeared.
and the .info registry
Both are run by Afilias, which is a big user and big developer of PostgreSQL. They're the ones that did the work on the Slony-I replication server.
First off, I think he's less worried about PostgreSQL's ability and more worried about making his boss comfortable with the idea. Others have already pointed out some case studies, so that's a good start.
To say that you're jumping the gun in your post would be an understatement. Read only databases aren't hard; there are many ways to accomplish that and with MySQL you would still have to convince your boss to use F/OSS. I hardly see how that answers his original question, it seems more like you're trying to convince someone that PostgreSQL is a bad choice (which seems strange considering that you don't know anything about his setup).
Also, the one thing that he did say is that he has a preexisting database in Oracle. It's much easier to go Oracle->PostgreSQL that Oracle->MySQL, but that may not be his primary concern.
Internet communities can be bad as well as good. The last thing you want is to support a vile, disgusting community like that.
Even if no money is being exchanged, many people will no doubt be contributing to the community, and that's why you don't want this community at all.
I feel bad for the creators of freenet. They wanted to create a free network, but it only caught on among the worst kinds of people. I had no idea that this bad stuff was so prevalent on the freenet network. Because of that, I will not run a freenet node.
Actually "juvenile diabetes" is an accurate name. I was diabetic for at least 10 years before I was diagnosed at 12-13.
Just because 90% of doctors don't care about patients at ALL and are so dense they can't diagnose a common condition with unique, easily testable symptoms before the patient is an adult doesn't mean you should change the name of the disease.
<rant>
If you couldn't tell, I'm bitter. My mom thought I was diabetic the whole time and I went to many many doctors (including a special childrens' hospital in Oakland when I was 4) and I wasn't diagnosed until I was 12-13. If you don't have a doctor in your family (I don't), then you are really screwed. Really, really screwed. No matter how much you pay them, doctors just don't care unless you're going to die before the appointment is up.
That's what you get when you bestow titles on people I guess. Screw the AMA, let non-AMA people practice legally.
</rant>
The new technology they seem to be promising does have advantages for diabetics. All the problems with the site at which the pump injects the insulin basically go away. No longer does it have to be moved every few days, and no longer will it leave you screwed when the site accidently gets ripped out or clogged and you can't fix it for a couple hours. Also, if the site ever does get "saturated" (where the insulin becomes decreasingly effective due to so much of it coming in at that one site), it's really easy to move. In fact, it could have many different sites, increasing the absorption rate to near-nondiabetic speed.
Also, it no longer requires the insulin to be forced through tubes, which could mean a lot less insulin waste, and a lot more practical thing to carry around.
Others have already pointed out how valuable research is, so I'll ignore that for now.
Hurd is also attempting to solve very real practical problems. Consider a typical UNIX network daemon:
(1) Must be started as root to listen on a privileged port
(2) Upon an incoming connection, must accept and then "drop privileges"
This causes mnay, MANY problems, that are very practical. If you took all the remote root security exploits ever in UNIX, and you subtracted those that involved a network daemon that needed to be run as root to listen on a privileged port, you'd be left with a rather secure system.
I just can't imagine why nobody recognizes a problem like that. You don't inherently need to run Apache/BIND/Sendmail with the privilege to overwrite the boot sector, but people ignore it as if "Oh well, it's a network daemon, of course it needs to be able to rewrite the boot sector. We'll just hope there are no bugs.".
Not only is this a security nightmare (which is only mitigated by the fact that UNIX is compared to windows rather than an ideal), but it's also got many performance implications. If you're measuring raw performance of already-written applications, a monolithic kernel will never be worse than a microkernel architecture. However, on a linux system, a lot of resources are traded in order to jump through hoops that don't need to be there, particularly for security. Maybe you don't really need to start that new process, and you can do everything you need in the current process. Sounds like a serious performance win to me, and not "my kernel avoids that 0.1% performance penalty you have to take on operation X".
For example, what about Apache, CGI and suexec? You really don't need all that when you could just be getting/releasing authentication tokens and using an apache module rather than starting a new process just so you can change privileges.
You can't separate the details of performance characteristics from capabilities. Capabilities may cost overhead, but may reduce algorithmic requirements.
If I met the following criteria:
/.
* an employer
* recently dealt with an employee who wrote derivative code from IBM Devloper Works
* intended to patent the code the employee wrote
* intended to distribute the code the employee wrote under a non-GPL license
* intended to claim ownership over all kinds of things that the employee did on time that wasn't directly compensated for
I would be pretty damn suspicious that this story was about me. Each of these things individually doesn't mean a whole lot probably, but all of them combined, along with the proper time frame, would pretty much narrow it down to one person in the world. Maybe it wouldn't, but it would be highly coincidental. Then just think to yourself: "Might my employee use an alias like 'Daimaou'?". It couldn't be that hard to figure out. Just look at some http requests when the employee visits
... fell into an open manhole while walking ...
Can't someone do something dumb without people mulling over it many years after they're dead? Not only that, friends don't walk friends into open manholes.
because of that government oversight, very little sale of human organs happen
First off, the moral issue (for me anyway) is not whether someone sells something, the moral issue is whether there is an incentive to make organs available where they wouldn't have been available before.
Even though there is huge demand for organs, people in the United States aren't murdering eachother to sell organs, nor are they raising new humans for the purpose of organ harvesting. It's not realistic to expect that will happen soon, either, in part because the people with the organs are likely to complain before they are harvested.
However, it is realistic to think that human embryos will be created for the purpose of harvesting. An embryo can't complain.
And, no, with government funded research, often no one makes any money.
Then why need funding? The government funding goes to someone. People want to be funded for research if they are not already, and people want more funding if they are already funded. "Profit" is mostly just an accounting term to recognize when something is a good investment and when it's not. It's not equivalent to making money. Lots of people make money from government funds, and it doesn't matter that it's not called "profit".
What amazes me is that the argument for government-funded embryonic stem cell research seems to be that it is not morally questionable. But the mere fact that we're in disagreement means that it is morally questionable. So, how about we not take people's money if they are morally opposed to the research anyway? Obviously, a significant portion of the U.S. is against it.