That reminds me of my roommate in college who wanted to join the water polo team and decided he had to put on about 10 lbs of fat. That was about when he started eating "nacho" cheese out of the gallon can from Costco. Yechh.
"[OS X] hit its high point of 6.46% in May, but it slipped back to 6% in June."
What are they implying? That OSX users suddenly abandoned their Macs and switched to Vista or other?
Well, the reality is that all of the hard-core Mac users went to the secret Apple retreat sponsored by Steve Jobs in the Bohemian Grove to drink some yummy Apple koolaid so they weren't surfing the web in June.
For the reasons Declan in particular would be excluded, see my blog. Declan wasn't the reporter excluded. Anne Broache was the reporter excluded. Declan is reporting on her exclusion.
Why would Anne Broache, in particular, be excluded?
The fuel tax combines both transportation cost and an incentive for fuel economy and that's where it goes a little off.
A fair tax for road usage would require you to write in your mileage every year and pay a tax. It's easier to hide it in the price of the fuel and it also gives an incentive for fuel efficiency as long as everyone is using the common fuel
However, we're now seeing an expansion of alternate fuels and this is starting to make the fuel tax look silly. This man was fulfilling one of the goals of the fuel tax - decreased consumption of petroleum but was not paying for his road usage. If you have an electric car that you plug into the grid at night there is no fuel tax. Is it fair to let electrics use the roads for free?
Nope. Whacky but true. A common way to get out of a speeding ticket is to inquire if a traffic survey has been performed recently (by law, I think it needs to be done every 5 years) and if it hasn't been done within the required time then the speed limit is null and void (this does not apply to violating the absolute maximum).
Use of the 85th percentile speed acknowledges that 15% of the drivers are traveling above a speed that is reasonable and proper. It is to this 15% that enforcement action is directed. Studies have shown that these are the drivers who cause many of the crashes and have the worst driving records.
There may be valid reasons for the police to keep actions secret *temporarily*. There should, however, be a record kept by the police of what happened that is available for public scrutiny afterwards.
There's also a difference between the police keep something secret and the police requiring citizens to keep things secret. Traditionally, even "classified" information is only something that employees of the government are required to keep secret. If a non-government employee gets the information they are largely free to do what they want with it. (This is in the United States. The UK, for example, has the "Official Secrets" act which makes disclosure of classified information by anybody illegal).
You *are* the government. When you start thinking about the government as being an entity that you are not responsible for you've abdicated your responsibilities as a citizen.
This wiretapping statue was presumably passed to protect the rights of citizens as it makes no special mention of the government. Its use by public officials to circumvent oversight is a travesty.
There's a big difference between their rights as citizens and their rights while acting as agents of the government.
Off-duty police officers have the rights as anyone else.
On-duty, those rights are restricted. Police officers and other public servants do not have an expectation of privacy while carrying out their duties. Why do police officers have name tags and numbered badges? To identify them both so that they can carry out their duties and so that they can be identified while carrying out their duties.
The police have a lot more "power" than the average citizen. I can't tell a random person on the street to move along. I can't ask people for their ID's. I can't hit or shoot people because they fail to comply with my instructions.
I'm sorry that you think that restricting the power of public servants is somehow restricting their human rights. It is not and it is a necessary and fundamental principle for a free society.
Yes! The government is not like your nosy neighbor. We can and should put additional restrictions on public servants while they are performing the jobs that we ask them to. Don't treat the government like a peer - it is both a useful and dangerous servant that needs to be watched and restricted so that those who would abuse the power we give the government cannot.
Even when you were clearly going faster than the posted speed limit.
Speed limits are funny things. In California (not sure about other states), speed limits are supposed to be set to the 85th percentile of speeds as determined by a traffic survey (provided that speed is under the maximum speed limit which is 65 mph these days, I think). So, on your average city street, 15% of people are *supposed* to be speeding.
Continuous gearing would be better, but that has never been made effective.
I've been driving a Honda HRV with a CVT transmission for 5 years now with no problems. I don't think a CVT will handle a muscle car yet but for a daily driver it's great. I've always hated automatics but the CVT is really nice to drive. Doesn't go hunting around for the right gear, feels like the gas pedal is connected to the drive train. I highly recommend it.
Why is it that everytime I see MMORPG I start thinking Might Morphin' Power Rangers?
Re:How about a more rational debate, Linus?
on
Linus on GIT and SCM
·
· Score: 1
I'm glad you guys haven't sunk to Linus' level. I thought his comments about SVN and its developers were unnecessarily rude when in reality all he is saying is that SVN doesn't work for his needs. I'm using SVN and it's a big help every day so THANK YOU!
His idea of having thousands of different branches and then picking and choosing among them is great for him but not very helpful when you have to PAY for the development of all those branches.
Well, supposedly SHA-1 collisions just won't happen in our lifetimes - heh. I would say that as long as no one is actively trying to create collisions and the number of files is small this will be true. Linus mentioned that GIT is working with on the order of 22000 files which, in the SHA-1 hash space is a tiny number. When you start having to deal with multiple billions of objects then I would start to worry although mathematically the chances of a collision are still tiny.
Re:Distributed version control gaining ground in F
on
Linus on GIT and SCM
·
· Score: 2, Insightful
What you're citing as advantages for decentralized version control are not the result of decentralization.
Cheap branches? Subversion has cheap branches. Better merging? This is a result of algorithms has nothing to do with whether the system is centralized or not.
If you're on a fast net with the server you can commit as often as you like. If you can branch/merge easily it's no problem.
If you want to cite advantages for decentralized version control it might be more like:
If you have to talk to a server over slow links, decentralized is much better
There's a big difference between saying "This idea is stupid" and "Those people are stupid". I watched the talk (it's actually the first time I've ever bothered to watch Linus talk) and he came across from start to finish as a big ass. All he talked about was how wonderful his system was but there was no real explanation of how it was so wonderful. All I got was that he was able to merge things faster which seems like a good thing for him but not how it did it or what the tradeoffs might be. He trashed the Subversion system and people repeatedly for no good reason. He trashed centralized version control for things like not being able to create tags which aren't numbers which makes no sense - that certainly isn't a result of things being centralized. CVS has weird rules on tag names but SVN has none that I'm aware of.
Just because you say up front that you're an ass and kind of laugh at yourself about it doesn't make it acceptable to be an ass.
You back up data you want protect content for roll-back. You archive data you don't expect to access for long term storage. You replicate data for which you need high availability.
These are not bad definitions and I'm sure that they are valid for your site. However, there is more to the world than what you do and these terms are overloaded.
We called this replication but it's not replication in the way the you define it. When users asked we explained it as being a form of backup because that's what they understood.
You are lying to them them. Which might be fine if they are end users. But if they are in the storage business, they need to understand these concepts. Frankly your end users really should know too. In the situation you describe, you can't restore deleted data because you have no backup, only replicas.
No, we weren't. Your definition of replication seems to be real time remote replication, probably between disk arrays over a SAN underneath a file system or a database. That's fine in your environment.
In our environment, replication referred to making multiple copies onto media like tape or optical disks. Now, I was doing replication of this form before remote replication was a twinkle in a disk array vendors eye (RAID was still a pretty new concept back then). The system doesn't let you truly delete data, you can always roll back your metadata (that is, the directory structure) to a previous version and the data is out there on stable storage someplace. So, is there a backup in the house?
You want to define a replica in a certain way and you want to define backup in a certain way that not everyone else would agree with. Given your definitions of backup and replica, sure, you can say that replication is not backup. I've certainly spent plenty of time explaining to people that having a RAID-1 mirror doesn't protect them against a lot of threats and that maybe they should consider making backup copies. I've also spent plenty of time explaining to people why they should have offsite copies of their backups which was where we came in in this discussion.
Now, the original poster said:
In IT, a backup kept on the same site as the original is hardly a backup at all.
Obviously what they meant was that having all of the copies of your data in a single site is a bad idea. You wanted to pull out your narrow definitions and show off what a hot shot you are and started off about how you do replication instead of backup to a second site and why would you want to offsite your backups. To which I said Replication is a subcategory of backup, and in the context we were talking it *is*.
Actually, I develop large scale storage systems and was doing 100 terabyte systems back in the early nineties. You can narrowly define backup the way you did if you like but it's not terribly useful. People understand the term "backup" as something that will keep them safe so when you start telling them they don't want backups you confuse them. Maybe you enjoy doing that.
The terms are fluid and not well defined. For example, in your RAID-1 example, what happens if you cleanly dismount the disk and break the mirror and put one of the disks on the shelf and replace it? Do you now have a backup or a replica?
A lot of my work has been with storage systems so large that you can't replicate them in a reasonable amount of time (these were in the 100 terabyte range back in the early 90's and are now in the multi-petabyte range). The only solution is to make multiple copies as the data is loaded into the system. We called this replication but it's not replication in the way the you define it. When users asked we explained it as being a form of backup because that's what they understood.
It sounds to me like you took some training and got a certificate - good for you. Try not to be so snotty to people who are using terms differently than you, though. And some of us around here are the ones who define the terms.
I do most of my work in Java these days. There are still a lot of gotchas available as you get deeper into parallelism. How many locks is your code holding right now? If it's holding more than 1 you have a potential for deadlock. Are all of the classes that you're using thread-safe? Oh, you're using an ArrayList and you didn't synchronize it? Oh, and BTW, the UI classes are not thread safe. Did you make sure that all of your UI events are executing on the UI thread?
If you think that synchronizing instruction streams is easy it's just because you're not doing anything very serious. Java's support is useful but it just basically lets you skip the easy problems and move straight to the hard ones.
I've been doing true parallel programming for the better part of 20 years now. I started off writing kernel code on multi-processors and have moved on to writing distributed systems.
Multi-threaded code is hard. Keeping track of locks, race conditions and possible deadlocks is a bitch. Working on projects with multiple programmers passing data across threads is hard (I remember one problem that took days to track down where a programmer passed a pointer to something on his stack across threads. Every now and then by the time the other thread went to read the data it was not what was expected. But most of the time it worked).
At the same time we are passing comments back and forth here on Slashdot between thousands of different processors using a system written in Perl. Why does this work when parallel programming is so hard?
Traditional multi-threaded code places way too much emphasis on synchronization of INSTRUCTION streams, rather than synchronization of data flow. It's like having a bunch of blind cooks in a kitchen and trying to work it so that you can give them instructions so that if they follow the instructions each cook will be in exactly the right place at the right time. They're passing knives and pots of boiling hot soup between them. One misstep and, ouch, that was a carving knife in the ribs.
In contrast, distributed programming typically puts each blind cook in his own area with well defined spots to use his knives that no one else enters and well defined places to put that pot of boiling soup. Often there are queues between cooks so that one cook can work a little faster for a while without messing everything up.
As we move into this era of cheap, ubiquitous parallel chips we're going to have to give up synchronizing instruction streams and start moving to programming models based on data flow. It may be a bit less efficient but it's much easier to code for and much more forgiving of errors.
he got a small fine and some community service - it's not like they're putting him to death over it. a good outcome imo, and hopefully next time he'll just pay out a couple of dollars for some coffee.
He was charged with a felony. The article doesn't state if the charge was reduced or not - it's quite possible the judge reduced the penalties but left the charge as a felony. I don't think leeching WI-FI is worth a felony conviction and I don't even think it's something the police should be involved with. WI-FI is using public airwaves and has simple ways to control access. It's a big difference between actively breaking into someone's network to cause damage and simply hooking up to someone's network. It's analagous to walking onto unsigned, unfenced private property, though not even that, since WI-FI is running over public spectrum.
Well, one problem with WI-FI is that you can be on someone else's network without even knowing.
I installed WI-FI back in 2000 in my house in Tokyo. I didn't bother setting up WEP. One day I was checking my server logs and I saw that someone was ssh'ing in, as me, from an unfamiliar IP address. After a bit of frantic security work I looked at the IP numbers a little more closely and then checked my laptop. The IP number was me! One of my neighbors had set up WI-FI themselves and from certain areas of the house their network was picked up in preference to my own.
So, by your analogy, this is like me wandering into my neighbor's house by accident, sitting down and watching their TV and having NO CLUE I'm in the wrong house and then getting arrested for trespassing.
These days I often see multiple WI-FI networks available anywhere I go. If I go to a coffee shop that has free WI-FI access I might wind up connected not to their network but to their next door neighbor's network, making me, technically, a felon.
The judge should have thrown the court out of case, given genius boy a lecture on ethics (like, go buy some coffee the next time you want to use the free WI-FI) and the prosecutor a long lecture on wasting the court's time.
Slashdot Mirror
That reminds me of my roommate in college who wanted to join the water polo team and decided he had to put on about 10 lbs of fat. That was about when he started eating "nacho" cheese out of the gallon can from Costco. Yechh.
E-World lives!
"[OS X] hit its high point of 6.46% in May, but it slipped back to 6% in June."
What are they implying? That OSX users suddenly abandoned their Macs and switched to Vista or other?
Well, the reality is that all of the hard-core Mac users went to the secret Apple retreat sponsored by Steve Jobs in the Bohemian Grove to drink some yummy Apple koolaid so they weren't surfing the web in June.
For the reasons Declan in particular would be excluded, see my blog.
Declan wasn't the reporter excluded. Anne Broache was the reporter excluded. Declan is reporting on her exclusion.
Why would Anne Broache, in particular, be excluded?
The fuel tax combines both transportation cost and an incentive for fuel economy and that's where it goes a little off.
A fair tax for road usage would require you to write in your mileage every year and pay a tax. It's easier to hide it in the price of the fuel and it also gives an incentive for fuel efficiency as long as everyone is using the common fuel
However, we're now seeing an expansion of alternate fuels and this is starting to make the fuel tax look silly. This man was fulfilling one of the goals of the fuel tax - decreased consumption of petroleum but was not paying for his road usage. If you have an electric car that you plug into the grid at night there is no fuel tax. Is it fair to let electrics use the roads for free?
This is a model for the law: http://www.ibiblio.org/rdu/nma-zone.html
This is a decent explanation: http://www.kentcountyroads.net/policies-speedlimi
From the Kent County information:
There may be valid reasons for the police to keep actions secret *temporarily*. There should, however, be a record kept by the police of what happened that is available for public scrutiny afterwards.
There's also a difference between the police keep something secret and the police requiring citizens to keep things secret. Traditionally, even "classified" information is only something that employees of the government are required to keep secret. If a non-government employee gets the information they are largely free to do what they want with it. (This is in the United States. The UK, for example, has the "Official Secrets" act which makes disclosure of classified information by anybody illegal).
You *are* the government. When you start thinking about the government as being an entity that you are not responsible for you've abdicated your responsibilities as a citizen.
This wiretapping statue was presumably passed to protect the rights of citizens as it makes no special mention of the government. Its use by public officials to circumvent oversight is a travesty.
There's a big difference between their rights as citizens and their rights while acting as agents of the government.
Off-duty police officers have the rights as anyone else.
On-duty, those rights are restricted. Police officers and other public servants do not have an expectation of privacy while carrying out their duties. Why do police officers have name tags and numbered badges? To identify them both so that they can carry out their duties and so that they can be identified while carrying out their duties.
The police have a lot more "power" than the average citizen. I can't tell a random person on the street to move along. I can't ask people for their ID's. I can't hit or shoot people because they fail to comply with my instructions.
I'm sorry that you think that restricting the power of public servants is somehow restricting their human rights. It is not and it is a necessary and fundamental principle for a free society.
Yes! The government is not like your nosy neighbor. We can and should put additional restrictions on public servants while they are performing the jobs that we ask them to. Don't treat the government like a peer - it is both a useful and dangerous servant that needs to be watched and restricted so that those who would abuse the power we give the government cannot.
Speed limits are funny things. In California (not sure about other states), speed limits are supposed to be set to the 85th percentile of speeds as determined by a traffic survey (provided that speed is under the maximum speed limit which is 65 mph these days, I think). So, on your average city street, 15% of people are *supposed* to be speeding.
I've been driving a Honda HRV with a CVT transmission for 5 years now with no problems. I don't think a CVT will handle a muscle car yet but for a daily driver it's great. I've always hated automatics but the CVT is really nice to drive. Doesn't go hunting around for the right gear, feels like the gas pedal is connected to the drive train. I highly recommend it.
Why is it that everytime I see MMORPG I start thinking Might Morphin' Power Rangers?
I'm glad you guys haven't sunk to Linus' level. I thought his comments about SVN and its developers were unnecessarily rude when in reality all he is saying is that SVN doesn't work for his needs. I'm using SVN and it's a big help every day so THANK YOU!
His idea of having thousands of different branches and then picking and choosing among them is great for him but not very helpful when you have to PAY for the development of all those branches.
Well, supposedly SHA-1 collisions just won't happen in our lifetimes - heh. I would say that as long as no one is actively trying to create collisions and the number of files is small this will be true. Linus mentioned that GIT is working with on the order of 22000 files which, in the SHA-1 hash space is a tiny number. When you start having to deal with multiple billions of objects then I would start to worry although mathematically the chances of a collision are still tiny.
What you're citing as advantages for decentralized version control are not the result of decentralization.
Cheap branches? Subversion has cheap branches.
Better merging? This is a result of algorithms has nothing to do with whether the system is centralized or not.
If you're on a fast net with the server you can commit as often as you like. If you can branch/merge easily it's no problem.
If you want to cite advantages for decentralized version control it might be more like:
If you have to talk to a server over slow links, decentralized is much better
There's a big difference between saying "This idea is stupid" and "Those people are stupid". I watched the talk (it's actually the first time I've ever bothered to watch Linus talk) and he came across from start to finish as a big ass. All he talked about was how wonderful his system was but there was no real explanation of how it was so wonderful. All I got was that he was able to merge things faster which seems like a good thing for him but not how it did it or what the tradeoffs might be. He trashed the Subversion system and people repeatedly for no good reason. He trashed centralized version control for things like not being able to create tags which aren't numbers which makes no sense - that certainly isn't a result of things being centralized. CVS has weird rules on tag names but SVN has none that I'm aware of.
Just because you say up front that you're an ass and kind of laugh at yourself about it doesn't make it acceptable to be an ass.
These are not bad definitions and I'm sure that they are valid for your site. However, there is more to the world than what you do and these terms are overloaded.
No, we weren't. Your definition of replication seems to be real time remote replication, probably between disk arrays over a SAN underneath a file system or a database. That's fine in your environment.
In our environment, replication referred to making multiple copies onto media like tape or optical disks. Now, I was doing replication of this form before remote replication was a twinkle in a disk array vendors eye (RAID was still a pretty new concept back then). The system doesn't let you truly delete data, you can always roll back your metadata (that is, the directory structure) to a previous version and the data is out there on stable storage someplace. So, is there a backup in the house?
You want to define a replica in a certain way and you want to define backup in a certain way that not everyone else would agree with. Given your definitions of backup and replica, sure, you can say that replication is not backup. I've certainly spent plenty of time explaining to people that having a RAID-1 mirror doesn't protect them against a lot of threats and that maybe they should consider making backup copies. I've also spent plenty of time explaining to people why they should have offsite copies of their backups which was where we came in in this discussion.
Now, the original poster said:
Obviously what they meant was that having all of the copies of your data in a single site is a bad idea. You wanted to pull out your narrow definitions and show off what a hot shot you are and started off about how you do replication instead of backup to a second site and why would you want to offsite your backups. To which I said Replication is a subcategory of backup, and in the context we were talking it *is*.
Actually, I develop large scale storage systems and was doing 100 terabyte systems back in the early nineties. You can narrowly define backup the way you did if you like but it's not terribly useful. People understand the term "backup" as something that will keep them safe so when you start telling them they don't want backups you confuse them. Maybe you enjoy doing that.
The terms are fluid and not well defined. For example, in your RAID-1 example, what happens if you cleanly dismount the disk and break the mirror and put one of the disks on the shelf and replace it? Do you now have a backup or a replica?
A lot of my work has been with storage systems so large that you can't replicate them in a reasonable amount of time (these were in the 100 terabyte range back in the early 90's and are now in the multi-petabyte range). The only solution is to make multiple copies as the data is loaded into the system. We called this replication but it's not replication in the way the you define it. When users asked we explained it as being a form of backup because that's what they understood.
It sounds to me like you took some training and got a certificate - good for you. Try not to be so snotty to people who are using terms differently than you, though. And some of us around here are the ones who define the terms.
Replication is a subcategory of "backups"
I do most of my work in Java these days. There are still a lot of gotchas available as you get deeper into parallelism. How many locks is your code holding right now? If it's holding more than 1 you have a potential for deadlock. Are all of the classes that you're using thread-safe? Oh, you're using an ArrayList and you didn't synchronize it? Oh, and BTW, the UI classes are not thread safe. Did you make sure that all of your UI events are executing on the UI thread?
If you think that synchronizing instruction streams is easy it's just because you're not doing anything very serious. Java's support is useful but it just basically lets you skip the easy problems and move straight to the hard ones.
I've been doing true parallel programming for the better part of 20 years now. I started off writing kernel code on multi-processors and have moved on to writing distributed systems.
Multi-threaded code is hard. Keeping track of locks, race conditions and possible deadlocks is a bitch. Working on projects with multiple programmers passing data across threads is hard (I remember one problem that took days to track down where a programmer passed a pointer to something on his stack across threads. Every now and then by the time the other thread went to read the data it was not what was expected. But most of the time it worked).
At the same time we are passing comments back and forth here on Slashdot between thousands of different processors using a system written in Perl. Why does this work when parallel programming is so hard?
Traditional multi-threaded code places way too much emphasis on synchronization of INSTRUCTION streams, rather than synchronization of data flow. It's like having a bunch of blind cooks in a kitchen and trying to work it so that you can give them instructions so that if they follow the instructions each cook will be in exactly the right place at the right time. They're passing knives and pots of boiling hot soup between them. One misstep and, ouch, that was a carving knife in the ribs.
In contrast, distributed programming typically puts each blind cook in his own area with well defined spots to use his knives that no one else enters and well defined places to put that pot of boiling soup. Often there are queues between cooks so that one cook can work a little faster for a while without messing everything up.
As we move into this era of cheap, ubiquitous parallel chips we're going to have to give up synchronizing instruction streams and start moving to programming models based on data flow. It may be a bit less efficient but it's much easier to code for and much more forgiving of errors.
he got a small fine and some community service - it's not like they're putting him to death over it. a good outcome imo, and hopefully next time he'll just pay out a couple of dollars for some coffee.
He was charged with a felony. The article doesn't state if the charge was reduced or not - it's quite possible the judge reduced the penalties but left the charge as a felony. I don't think leeching WI-FI is worth a felony conviction and I don't even think it's something the police should be involved with. WI-FI is using public airwaves and has simple ways to control access. It's a big difference between actively breaking into someone's network to cause damage and simply hooking up to someone's network. It's analagous to walking onto unsigned, unfenced private property, though not even that, since WI-FI is running over public spectrum.
Well, one problem with WI-FI is that you can be on someone else's network without even knowing.
I installed WI-FI back in 2000 in my house in Tokyo. I didn't bother setting up WEP. One day I was checking my server logs and I saw that someone was ssh'ing in, as me, from an unfamiliar IP address. After a bit of frantic security work I looked at the IP numbers a little more closely and then checked my laptop. The IP number was me! One of my neighbors had set up WI-FI themselves and from certain areas of the house their network was picked up in preference to my own.
So, by your analogy, this is like me wandering into my neighbor's house by accident, sitting down and watching their TV and having NO CLUE I'm in the wrong house and then getting arrested for trespassing.
These days I often see multiple WI-FI networks available anywhere I go. If I go to a coffee shop that has free WI-FI access I might wind up connected not to their network but to their next door neighbor's network, making me, technically, a felon.
The judge should have thrown the court out of case, given genius boy a lecture on ethics (like, go buy some coffee the next time you want to use the free WI-FI) and the prosecutor a long lecture on wasting the court's time.
None of them foresaw cell phones, let alone cell phones with cameras in them.
Hey, don't forget Dick Tracy. He had all that stuff!