Absolutely the colors on some seem to be out of balance with what we expect to see. Some of the plates, because of a slight time lapse, have rgb artifacts becuase of moving objects.
I choose to believe that some of the brilliant colors may stand out because the colors that were being recorded were absolutely amazing. Some of the blues in the architecture section are stunning, and the beauty of the scenes are touching and beyond words.
This has got to be the coolest thing I've seen posted on slashdot in months.
However procmail (and maybe other local delivery agents?) can be used with the maildir format, so you get all of the advantages without worrying about running qmail if you're not ready yet.
-Peter
Re:Do owners have good reasons to keep people out?
on
Infiltration
·
· Score: 2
I grew up (and still live on) Roosevelt Island in New York City. The abandoned ruins on the south side of the island were the playground for most of the male pre-adolescents on the island at one point or another.
It was fantastic. Huge former hospitals, former prisons. Stone structures partly laid bare, rotting, falling apart. A true adventure and novelty each time you visited. 6 or seven buildings in all, and never a day where exertion and discovery didn't overcome boredom.
A few years after I stopped playing in those ruins a kid died falling through a rotted floor, and security there finally became serious, and many of the buildings were torn down, for safety reasons.
I cherish my memories of exploring those buildings, finding iron lungs and other odd contraptions.
Based on some limited investigation of chiliware, I agree with bruce here. I don't think they're capable of writing 100 useable applications in the next year. I think they may be capable of packaging other people's software, though.
Note the slew of sites that chilliware wants to use to drive its name - none of which have anything on them. Chilliware looks like it's another internet play with linux as it's hole card.
If you want a good commercial application suite, I'd say it's not chilliware you'll be getting it from.
I've put linux at a lot of workplaces, and found that IS is very scared of it at first, but tend to forget about it until they *don't* have to worry about it - i.e. when the latest virus, os update, etc. comes through.
However, there are times when linux has presented valid problems - having samba set up and declare itself the master browser in such a way as to disrupt windows systems from connecting to shares on the domain, etc. (this may no longer be a problem, but it has happened). However this just points to the need to document how new elements must behave, and make it clear that new things must be tried in an environment that is not critical to production. This can teach real lessons and improve networks a lot.*
IMO an IT department that resists change above all else is one that will resist itself out of existance. It's the users that try new things for themselves, and from these evaluations learn new things, that make systems and networks useable.
-Peter
* Anyone else remember when doom first hit big novell networks? Networks were being crashed by the flood of ipx broadcast traffic. Even though this killed business in many places, in the end it fixed configs across the world to keep private networks private. Disruptive events like that, and like linux have to be evaluated and lessons learned. They can't be squashed.
Most journals are written asynchronously for performance reasons. The performance of doing otherwise is considered unacceptable.
So what make you feel that a journal necessarily has any more current and consistant information then a tree that's 2 phases out at the point in time of a disaster?
Linux can use swapfiles, but they're less effecient the a swap partition because the various VM and FS policies that deal with normal user files are very different from the optimal use of a swap file. I.e. you really don't want or need a cache layer between your memory and swap file.
Also, dynamicly creating and removing swap files (or extending and shrinking them) is going to cause your filesystem to become massivly fragmented very quickly, causing many multiples worse performance then the already horrendously nasty and unreasonable performance of having to page out 600MB, then do something, then page it back in.
What kind of glutton for punishment are you that you want to do this to yourself?
Actually, can you give an example of an application that really does require this kind of swap? When you run a large database you try to pin its memory (shm, cache, and table buffers) into active memory so it can't be paged out. Graphics rendering systems would be slowed down many orders of magnitude if this much data had to be swapped per frame.
The only reason I can think of for the need for 600 MB of swap in most systems is because of an application with a lot of leaking memory. Please let slashdot know if you've got another reason:)
I believe that the speedup is the result of gathering writes into groups. The more data you can throw on the disk in sequence, the better off you are in terms of speed. This is one of the ways that write-back caches speed up your life. All writes can be arranged so that data is written in a linear fashion to the disk.
Using a data and metadata log can make some operations faster, but you do suffer with sun's ufs+logging the problem of having to write to the log, then copy the log to the filesystem. This can be an issue in an environment where the filesystem is used for a lot of transactions.
IMO I believe that tux2 can do better in situations like this because of the elimination of an unneeded copy.
Also note that sun's ufs filesystem has long been horribly, horribly slow relative to it's cousin, the bsd FFS. I'm going to guess that when adding the logging feature, sun's developers decided it could be seen as an opportunity to add other speed enhancements besides just logging.
In the interview, they used buzz-terms like "programming by contract" which is a practice that can make a lot of problems go away, but doesn't address the question of how the system was designed. If it's still a monolithic binary with authoratitve and caching built into one, w/o a well defined seperation then it seems like the default is still going to have trouble with preveninting attacks that poison the cache. If the credability rules still exist, then it'll again be easy to start an attack through BIND.
I do hope BIND9 is better then 4 and 8, but I don't think I'm going to use it now that I've got everything I need in djbdns.
I'm not sure you've got all the facts WRT immigrants coming here to "be americans".
It was a common practise for immigrants to come to america, make money, and return to their home towns to settle down. I.e. the patriarch of the Trump family (i.e. the builder of New Yorks City's ugliest buildings) made their fortune, returned to europe (germany or austria I believe) with money, and then was chased out because he had returned just after he was over mandatory military service age.
There are mining towns in Pennsylvania where the entire town came from one village in italy and they came here to work, but brought over the culture etc. of their old world. Certianly not to become americans.
The current state of america's amazingly broad background of people seems to me to be at least as much the result of people who were determined they were never going to stay but were seduced by the better life here as it was people who up and moved.
Don't believe the happy crap in the history books. Most of it is lies. Lots of people moved to america because the alternative was to be put on a firing line or starving. I believe that if a person moves just to make money, in most cases they will believe they will return to their home. It's desperation that causes people to uproot themselves completely.
Also, AFAIK akamai doesn't do full peering in their colo facilities. They may get a full routing table, but they're not propegating their own routes, are they?
Internap just buys enough transit and sees enough routes to actually make informed decisions on their own routes. They are in a position to send any packet to it's destination via the best path as far as can be determined via BGP.
Whether they take full advantage of this or not is really the question. Obviously they're trying.
-Peter
(Former internap customer. Quit that job, but really liked their service. They were the most proactive ISP I've ever seen, bar none).
This URL points to an abstract of a system that seems to assume some of the prior art that you are asking for. However it's been done on an old Xerox LISP machine, not on a UNIX box using the X window system.
However the abstract does say that the database is running on a sun server with a relational database, and the article is from 1990.
I don't have an ACM account with which to access the complete article, and I'm not going to pay for the download. Perhaps your library has this on the shelf?
Yah. I had a high school teacher who used to work in power plants. He recounted engineers in places where hydrogen was used walking down aisles with wooden brooms waving in front of them when there was a leak - to make sure the broom burned before they did. He said that they couldn't see the flame.
The use of open/free/whatever software is not popular just because of the politics. Remember - it's just easier to get your hands on it, get familiar with it, and have warm cuddly feelings 'bout it. Users understand that.
Why not? I've been given a ticket for "littering" in new york city for having a half-full bottle of juice at my feet while talking to a friend.
I believe my real crime was (from the officers comments) having long hair and "not sitting like a normal human being" (I had my legs crossed and I was slouching on the bench).
So in my experience, citations, violations, and arrests can happen for just walking down the street without a cell phone (a co-worker of mine got jumped by 4 cops in the "street crimes unit", the division that a couple of months later killed Amadu Diallo - see here for some satire). He had done nothing.
Next question: if you're running a 100GB database for a $400 million company, would you put it on Linux, with a filesystem that will need 20 minutes to fsck in the event of an emergency reboot? Is "experimental" support for a shared fiber-channel disk array good enough to allow you to sleep soundly? Calling Sun or another "expensive" high-end vendor starts making more sense here.
Funny thing is that most companies that use Sun (and I'm only picking on sun here) as their mission critical platform use solaris 2.6, which suffers all of the weaknesses that you harp on. Here's the solaris stuff 12 months ago:
DMP for FCAL arrays under 2.6 was broken (last solution I saw was that FC/scsi emulation drivers were disabling one ring to work reliably on a dual-ringed JBOD). I don't know if this has been fixed.
Veritas' filesystem or volume manager, unless you tune the kernel in a manner that isn't documented anywhere but in sunsolve (and was damn hard to find, at that), you're likely to cause kernel thread stack overflows and crash your kernel as soon as you get a really high I/O volume.
You see, you can't sleep easy with sun either. You still have to be on your toes, and be ready to be on call 24/7. Until you shake out all of their problems for them (remember, with sun's 5x00 arrays, veritas comes with it) you have to assume the worst, because it will happen to you at 4:00am in the morning.
You man OSX, which still doesn't seem very good at pre-empting the display? Watching ie crash on macosx is pretty funny. The app crashes, and the dialogue's frame comes up, but you the user have to switch to the finder to tell the OS that it's OK, you figured out it crashed when the main window dissapeared.
Slashdot Mirror
Absolutely the colors on some seem to be out of balance with what we expect to see. Some of the plates, because of a slight time lapse, have rgb artifacts becuase of moving objects.
I choose to believe that some of the brilliant colors may stand out because the colors that were being recorded were absolutely amazing. Some of the blues in the architecture section are stunning, and the beauty of the scenes are touching and beyond words.
This has got to be the coolest thing I've seen posted on slashdot in months.
-Peter
We might wait for the israelis to find a salt lake in a valley and move in so we have someone to sell food to.
However procmail (and maybe other local delivery agents?) can be used with the maildir format, so you get all of the advantages without worrying about running qmail if you're not ready yet.
-Peter
I grew up (and still live on) Roosevelt Island in New York City. The abandoned ruins on the south side of the island were the playground for most of the male pre-adolescents on the island at one point or another.
It was fantastic. Huge former hospitals, former prisons. Stone structures partly laid bare, rotting, falling apart. A true adventure and novelty each time you visited. 6 or seven buildings in all, and never a day where exertion and discovery didn't overcome boredom.
A few years after I stopped playing in those ruins a kid died falling through a rotted floor, and security there finally became serious, and many of the buildings were torn down, for safety reasons.
I cherish my memories of exploring those buildings, finding iron lungs and other odd contraptions.
-Peter
Well, there's plenty of silicon there...
I thought that part of cryptonomicon was speculation on Stephens' part. It's cool to find out that this was not only real but still documented.
What a clever idea.
-Peter
"Hey ma"
Uuuuuuurrrrrp
"Excuse me"
Based on some limited investigation of chiliware, I agree with bruce here. I don't think they're capable of writing 100 useable applications in the next year. I think they may be capable of packaging other people's software, though.
Note the slew of sites that chilliware wants to use to drive its name - none of which have anything on them. Chilliware looks like it's another internet play with linux as it's hole card.
If you want a good commercial application suite, I'd say it's not chilliware you'll be getting it from.
-Peter
I've put linux at a lot of workplaces, and found that IS is very scared of it at first, but tend to forget about it until they *don't* have to worry about it - i.e. when the latest virus, os update, etc. comes through.
However, there are times when linux has presented valid problems - having samba set up and declare itself the master browser in such a way as to disrupt windows systems from connecting to shares on the domain, etc. (this may no longer be a problem, but it has happened). However this just points to the need to document how new elements must behave, and make it clear that new things must be tried in an environment that is not critical to production. This can teach real lessons and improve networks a lot.*
IMO an IT department that resists change above all else is one that will resist itself out of existance. It's the users that try new things for themselves, and from these evaluations learn new things, that make systems and networks useable.
-Peter
* Anyone else remember when doom first hit big novell networks? Networks were being crashed by the flood of ipx broadcast traffic. Even though this killed business in many places, in the end it fixed configs across the world to keep private networks private. Disruptive events like that, and like linux have to be evaluated and lessons learned. They can't be squashed.
I thought we're the working hard and moderately comfortable, though highly motivated.
-Peter
Most journals are written asynchronously for performance reasons. The performance of doing otherwise is considered unacceptable.
So what make you feel that a journal necessarily has any more current and consistant information then a tree that's 2 phases out at the point in time of a disaster?
-Peter
Linux can use swapfiles, but they're less effecient the a swap partition because the various VM and FS policies that deal with normal user files are very different from the optimal use of a swap file. I.e. you really don't want or need a cache layer between your memory and swap file.
:)
Also, dynamicly creating and removing swap files (or extending and shrinking them) is going to cause your filesystem to become massivly fragmented very quickly, causing many multiples worse performance then the already horrendously nasty and unreasonable performance of having to page out 600MB, then do something, then page it back in.
What kind of glutton for punishment are you that you want to do this to yourself?
Actually, can you give an example of an application that really does require this kind of swap? When you run a large database you try to pin its memory (shm, cache, and table buffers) into active memory so it can't be paged out. Graphics rendering systems would be slowed down many orders of magnitude if this much data had to be swapped per frame.
The only reason I can think of for the need for 600 MB of swap in most systems is because of an application with a lot of leaking memory. Please let slashdot know if you've got another reason
-Peter
I believe that the speedup is the result of gathering writes into groups. The more data you can throw on the disk in sequence, the better off you are in terms of speed. This is one of the ways that write-back caches speed up your life. All writes can be arranged so that data is written in a linear fashion to the disk.
Using a data and metadata log can make some operations faster, but you do suffer with sun's ufs+logging the problem of having to write to the log, then copy the log to the filesystem. This can be an issue in an environment where the filesystem is used for a lot of transactions.
IMO I believe that tux2 can do better in situations like this because of the elimination of an unneeded copy.
Also note that sun's ufs filesystem has long been horribly, horribly slow relative to it's cousin, the bsd FFS. I'm going to guess that when adding the logging feature, sun's developers decided it could be seen as an opportunity to add other speed enhancements besides just logging.
-Peter
In the interview, they used buzz-terms like "programming by contract" which is a practice that can make a lot of problems go away, but doesn't address the question of how the system was designed. If it's still a monolithic binary with authoratitve and caching built into one, w/o a well defined seperation then it seems like the default is still going to have trouble with preveninting attacks that poison the cache. If the credability rules still exist, then it'll again be easy to start an attack through BIND.
I do hope BIND9 is better then 4 and 8, but I don't think I'm going to use it now that I've got everything I need in djbdns.
-Peter
I'm not sure you've got all the facts WRT immigrants coming here to "be americans".
It was a common practise for immigrants to come to america, make money, and return to their home towns to settle down. I.e. the patriarch of the Trump family (i.e. the builder of New Yorks City's ugliest buildings) made their fortune, returned to europe (germany or austria I believe) with money, and then was chased out because he had returned just after he was over mandatory military service age.
There are mining towns in Pennsylvania where the entire town came from one village in italy and they came here to work, but brought over the culture etc. of their old world. Certianly not to become americans.
The current state of america's amazingly broad background of people seems to me to be at least as much the result of people who were determined they were never going to stay but were seduced by the better life here as it was people who up and moved.
Don't believe the happy crap in the history books. Most of it is lies. Lots of people moved to america because the alternative was to be put on a firing line or starving. I believe that if a person moves just to make money, in most cases they will believe they will return to their home. It's desperation that causes people to uproot themselves completely.
-Peter
Also, AFAIK akamai doesn't do full peering in their colo facilities. They may get a full routing table, but they're not propegating their own routes, are they?
-Peter
Internap just buys enough transit and sees enough routes to actually make informed decisions on their own routes. They are in a position to send any packet to it's destination via the best path as far as can be determined via BGP.
Whether they take full advantage of this or not is really the question. Obviously they're trying.
-Peter
(Former internap customer. Quit that job, but really liked their service. They were the most proactive ISP I've ever seen, bar none).
http://www.acm.org/pubs/citations/proceedings/mod/ 93597/p387-cha/
This URL points to an abstract of a system that seems to assume some of the prior art that you are asking for. However it's been done on an old Xerox LISP machine, not on a UNIX box using the X window system.
However the abstract does say that the database is running on a sun server with a relational database, and the article is from 1990.
I don't have an ACM account with which to access the complete article, and I'm not going to pay for the download. Perhaps your library has this on the shelf?
-Peter
It still smells like toxic chemicals. Maybe the car companies have found a way to insert magic smellazoids in the exhaust?
-Peter
Yah. I had a high school teacher who used to work in power plants. He recounted engineers in places where hydrogen was used walking down aisles with wooden brooms waving in front of them when there was a leak - to make sure the broom burned before they did. He said that they couldn't see the flame.
-Peter
The use of open/free/whatever software is not popular just because of the politics. Remember - it's just easier to get your hands on it, get familiar with it, and have warm cuddly feelings 'bout it. Users understand that.
-Peter
So, you would be one of those good christians, that believes in forgiveness and turning the other cheek.
Unless, of course, the situation is one where you judge that such basic levels of understanding and good manners aren't called for?
-Peter
I believe my real crime was (from the officers comments) having long hair and "not sitting like a normal human being" (I had my legs crossed and I was slouching on the bench).
So in my experience, citations, violations, and arrests can happen for just walking down the street without a cell phone (a co-worker of mine got jumped by 4 cops in the "street crimes unit", the division that a couple of months later killed Amadu Diallo - see here for some satire). He had done nothing.
-Peter
Funny thing is that most companies that use Sun (and I'm only picking on sun here) as their mission critical platform use solaris 2.6, which suffers all of the weaknesses that you harp on. Here's the solaris stuff 12 months ago:
DMP for FCAL arrays under 2.6 was broken (last solution I saw was that FC/scsi emulation drivers were disabling one ring to work reliably on a dual-ringed JBOD). I don't know if this has been fixed.
Veritas' filesystem or volume manager, unless you tune the kernel in a manner that isn't documented anywhere but in sunsolve (and was damn hard to find, at that), you're likely to cause kernel thread stack overflows and crash your kernel as soon as you get a really high I/O volume.
You see, you can't sleep easy with sun either. You still have to be on your toes, and be ready to be on call 24/7. Until you shake out all of their problems for them (remember, with sun's 5x00 arrays, veritas comes with it) you have to assume the worst, because it will happen to you at 4:00am in the morning.
-Peter
You man OSX, which still doesn't seem very good at pre-empting the display? Watching ie crash on macosx is pretty funny. The app crashes, and the dialogue's frame comes up, but you the user have to switch to the finder to tell the OS that it's OK, you figured out it crashed when the main window dissapeared.
-Peter