I want a search engine that only indexes items excluded in the robots.txt file:-)
What's interesting is that I've heard of robots that do that exclusively. It may of been here on slashdot, but I've heard of people putting stuff in their exclude list in robots.txt and some robots _ONLY_ searched those files.
As soon as I typed the first 'S', up pops good old google autocomplete:
"STD clinic london"
I turn off all of that autocomplete crap, but does anyone really find the lack of privacy, potential for embarrassment, and the frequency of auto completing the wrong thing better than having Google say "Did you mean ___?" or just typing a few words?
I've simply never found any of this autocomplete stuff a feature. I search for so much stuff (99.999% of it in a Google search bar accessible via Apple-L, TAB, type hit return) that I just don't see how it could be any quicker any other way.
Now, the fun of seeing "STD clinic london" merely enough for everyone to not have this feature. For all we know this was for her cousin or something or for a book she is writing, but its still embarrassing and unnecessary to have such info flying back at you all the time.
To play devil's advocate, child porn possession is HIGHLY illegal. By your definition, possession of it is not advocating anything and should not be illegal? What's the difference (just speaking to possession here, not creation).
There is a big difference here. Child porn is almost universally tabooed across the planet. Children have no knowledge or experience regarding sex. Sex with children is not a human need.
OTOH, violent sex is almost universally done across the planet to some degree (at least after fight/makeup sex). Adults are aware and have knowledge regarding sex. Sex among adults is a human need.
I hope this is not your example of how Linux is ready for the mainstream.
No, this is:
for i in `seq -f 'node%04.0f' 1 2048` do
rsh $i 'echo 100 >/proc/sys/vm/swappiness' done
Granted its a cheesy example, but I would rather do that on 2,000+ Linux machines vs playing around with a registry key on 2,000+ Windows machines any day.
exactly... and some OSs (read: OSX) caches less-frequently used data (cached window contents, and other images, etc) to the drive to free up real RAM; it doesn't matter how much RAM is installed on the machine, it'll still use the swap. Even my machine at work with 8GB of RAM frequently uses the swap even before 1/4 of the RAM has been touched
exactly back at you:)
Its also good for an OS to page out "dead code". Things like initialization code or stale data or whatever so that real RAM can be used for disk cacheing and allowing new apps to do their thing.
It frustrates me when a user runs something that takes all of the memory on a box, but it happens. Hell, I've even done it compiling some code where the preprocessor ran my box out of memory. Some seriously nasty C macros. Gotta love those new CS guys:)
Sure, but its what you are locking into and to what degree.
UNIX and Linux (excluding OS X) is much more stable in terms of APIs, backwards compatibility, open standards, and the like than Apple OSes or Microsoft OSes. No, this is not absolute. Yes, I've frantically debugged some code I wrote after applying patches to an AIX box. Yes, I have had much more issues with Windows and OS X (with and before OS X) with gotchas after updates.
To put this in perspective, Sun has a current patch set for Solaris 2.5.1 that came out August 16th of this year. 2.5.1 came out over 10 years ago. Where I work, it takes months to validate a Windows service pack and document its gotchas. When I update my Macs, its a crapshoot if everything is going to be OK.
UNIX and Linux are THE OSes for server side "real work" (TM). Migration within and between them is relatively easy. You are making some kind of a lock-in, but that is a generic platform lock-in, not a specific vendor lock-in. Apache or any other web server will run just fine on any UNIX or Linux box. Sure, it runs on Windows too, but there are tons of gotchas and differences between UNIX and Linux vs Windows. NFS is basically the same on AIX, Solaris, *BSD, Linux. NFS is available on Windows too, but its going to cost you and its robustness is going to be on your reputation, not mine.
you never know when some runaway process is going to eat all yer RAM and need to use swap... no matter how much RAM you've got.
Personally, I prefer a runaway process to run out of resources and stop vs take over my whole system. It takes a long time to page out 1+ Gigs of RAM. It takes a long time to unpage all of that at shutdown or even when an app is closed.
Swap completely depends on the computer's real RAM available and the purpose of the computer and the OS on said computer.
To adequately answer the question, "How much Virtual Memory is Enough?" The correct answer is "It depends".
Having too much swap on a HPC type of machine is a nightmare and will kill performance. Having too little swap on a general purpose server (moreso real RAM) is going to hurt performance. Paging out too much on a laptop with a slow disk can be very painful and slow down the shutdown process.
Well, given that 80% of their $5.6 billion in operating profit (from TFA) is derived from ink and toner, they are not going to allow it to slip away. Otherwise they could potentially lose about $4bn in profit.
I'm not defending it, cause I don't agree with it. But they're not going to let that bone go any time soon.
But the thing is that people are getting pissed off at the cheap, but low quality hardware and the chronic extortion-like scheme for keeping ink and toner in the printers.
So, for yet another poor analogy, car companies are going to be owned by oil companies, and cars would suffer in quality and be loss leaders that were inefficient in fuel consumption to recoup their costs.
Personally, I believe the printer market is just at a low point driven by dumb consumers who demand cheap crap. They are starting to learn, and the market will change to have quality printers again, but that will take time. Look at the quality of cars today. Cars today are much better than cars from 15 years ago. Same with computers. This is not true for your average consumer level printer. They suck.
Cool, but it's really nothing more than a bunch of servers in a single rack with big hard drives.
Actually, it is or can be one big disk or "just" a JBOD. Yes, each box is just a regular server running Linux, but the setup can go like this: internal RAID5 or so within each box, then concatenate the drives with RAID 0 over each of the raid5.
RAID cards are another example of a commercially successful offloading hardware device. It is possible to do it in software, but most systems do it in hardware.
True, but only on the high end, price has no object range.
Cheaper RAID cards are not worth using because in the event of a card failure the odds of you losing your data is still great. I've heard that some of these things cannot read drives that were setup with a different firmware revision of the same brand of card.
Again, for an in box card, I would opt to use the general purpose CPU over a dedicated RAID card. Now whatever is in my hardware RAID array is a different story. I don't know or care. Its a black box to me, and its from a reputable vendor. But I would trust generic SCSI or (S)ATA cards and software RAID from someone like OS X or Linux or Solaris over something like one of those cheap raid cards. If I care that much about my data, I'm going to spend over $200 for a solution. Otherwise, I clearly don't care about my data.
I'm not sure what that smell is, but its familiar.
Yes, TCP/IP offloaders, crypto offloaders, physics offloaders, FFT offloaders, have all existed. The only accepted offloader that has succeeded is the GPU, and that is because it was subsidized by the high end graphics people and people with game addictions. The cost/benefit of the other offloaders has not proven itself. Especially when you consider the rate of increase of the CPU speeds and the bottleneck of getting the data to the offloaded chip and back again. The Linux kernel mailinglist has been pretty much anti TCP/IP offloading because the time spent optimizing the drivers and the performance increase was often surpassed in a few months with a faster general purpose CPU and a generic driver that worked with all cards. This is also confounded when you have OS level TCP/IP stuff like ipchains or iptables that need their code in the OS and not on the card. FFT and physics offloaders have not taken off because of the cost/benefit loss when one takes into account the speed loss of getting the data onto the card and then the cost of the cache RAM to put on the card (if its ever enough), and then the time to get the data off of the card.
Now, what will make these things work?
A bus that is near or at the speed of memory bandwidth or a problem where the data does not need to go back to the host system. A GPU falls into the second half there. The display needs to know what its told, the computer does not need to know the details. TCP/IP offloaders have failed to catch on because of the price/performance benefit and their lack of ubiquity and commonality. Crypto offloaders are cool. Especially for the geek factor of having the keys stored on the card and zeroed out when tampered with. But again, the cost of writing specific drivers vs known CPU drivers for doing crypto and the rate of increase of cheap commodity CPUs over time often exceeds the cost/benefit of the crypto card.
Now, give me a fast bus and the ability to use my generic system RAM with one of these offloading cards, and things could change, but until then I expect this battle between the CPU and specific PUs to continue with no real winner.
If you want value and performance custom built is the way to go, not prebuilt crap.
If you want a car considering value and performance custom built is the way to go, not prebuilt crap. False
If you want a home considering value and performance custom built is the way to go, not prebuilt crap. False
If you want a piece of furniture considering value and performance custom built is the way to go, not prebuilt crap. False
If you want a piece of clothing considering value and performance custom built is the way to go, not prebuilt crap. False
If you want a wristwatch considering value and performance custom built is the way to go, not prebuilt crap. False
If you want a television considering value and performance custom built is the way to go, not prebuilt crap. False
If you want a computer considering value and performance custom built is the way to go, not prebuilt crap. True???
I simply do not believe that computers are the only commodity consumer device that requires the knowledge and skill and desire to hand build said commodity item better than any company with assembly lines in China, Japan, or the United States can do.
What is so special about computers? And is there a market for a commodity company to provide such value and performance products since there is no current company providing such things in the world today?
For those who have not escaped mom's basement in years, we would like to know...
Re:Don't People Bother to "Search" Before Posting?
on
What's On Your Thumbdrive?
·
· Score: 2, Insightful
Could do a long post... but easier just to point to this/. post that was already up with MANY MANY good links.
100 million is a pretty big payout for an obvious way to navigate through music that I myself invented when I was a kid. This method is: "The patent covers an interface that lets users navigate through a tree of expanding options, such as selecting an artist, then a particular album by that artist, then a specific song from that album."
I mean, isn't that how the stuff is organized in the record store too?
"No wireless. Less space than a nomad. Lame." 100 million dollars in patent taxes lame.
I think its great that everyone thinks Venture Capitalists are complete idiots. If thats true then where did they get the money to invest in the first place?
Is that your variation of the "Test this for a while and if you like it, buy it!"
Yes. And also, I cannot try most software for free. They make me give away my life information just to see if their crappy software works or not, and usually it doesn't.
Case in point. I currently burn my music CDs with a script that I wrote and cdrecord. I read on the flac website that the latest and greatest Roxio Toast Titanium allows you to burn directly from flac, thus avoiding the uncompress step. Cool I thought. So I downloaded a version via bittorent. Well, it was a waste of my time and not worth the $0 that I paid for it. The songs had a pop or gap or something between the tracks, so I actually lost on my CD media and my time trying out this crappy software, and I don't use it any more in favor of my scripts.
Now, when software is like any other product where I can just return it to the store and say "It sucks, give me a refund", then I will change my behavior. But as it stands now, I've paid thousands of dollars in software, and if I could have used it first, I would not have done so. To me, the software industry, like the music industry, has taught me that its a dumb decision to pay my hard earned money for their products. And so, I go for alternatives or create my own.
Yeah, I'm relatively sure that 99% of torrent download ARE legitimate things like linux distros..
99% of the torrents I download are legitimate things. Those that aren't are usually software that I want to try and then don't use because they are inferior to freely available alternatives.
This is a recipe for a nightmare when it comes to backup, security, and disaster recovery. It's usually also a terribly inefficient (and therefore, expensive) use of disk space and in my experience it often results in document versioning problems which take hours to unravel--you know, Fred and Alice are both working on that proposal, each of them has it on their desktop, making mutually exclusive changes simultaneously.
The disk space comes with the desktop machines and is cheap. If Fred and Alice are both modifying a document on a shared drive doesn't versioning come into play or aren't the documents locked out? Also, its up to the users to put their data on the shared and backed up network drive because their desktop is not backed up.
Now if you wanted to make a bomb out of liquids - why not just bring a bottle of water and a piece of rubidium or cesium? Remember what happens when alkali metals hit water? BOOM! Two grams of cesium and a quart of water is enough to make an explosion roughly equivalent to about three or so hand grenades going off. Water and rubidium can blow apart a bathtub. Cesium is far, far more reactive.
Thanks for now limiting the number of things I can now bring on a plane. I thought the latest thing with not allowing liquids on board was bad, but now you're telling people that bringing 2 grams or more of metal onto a plane is now banned.
Slashdot Mirror
I want a search engine that only indexes items excluded in the robots.txt file :-)
What's interesting is that I've heard of robots that do that exclusively. It may of been here on slashdot, but I've heard of people putting stuff in their exclude list in robots.txt and some robots _ONLY_ searched those files.
As soon as I typed the first 'S', up pops good old google autocomplete:
"STD clinic london"
I turn off all of that autocomplete crap, but does anyone really find the lack of privacy, potential for embarrassment, and the frequency of auto completing the wrong thing better than having Google say "Did you mean ___?" or just typing a few words?
I've simply never found any of this autocomplete stuff a feature. I search for so much stuff (99.999% of it in a Google search bar accessible via Apple-L, TAB, type hit return) that I just don't see how it could be any quicker any other way.
Now, the fun of seeing "STD clinic london" merely enough for everyone to not have this feature. For all we know this was for her cousin or something or for a book she is writing, but its still embarrassing and unnecessary to have such info flying back at you all the time.
To play devil's advocate, child porn possession is HIGHLY illegal. By your definition, possession of it is not advocating anything and should not be illegal? What's the difference (just speaking to possession here, not creation).
There is a big difference here. Child porn is almost universally tabooed across the planet. Children have no knowledge or experience regarding sex. Sex with children is not a human need.
OTOH, violent sex is almost universally done across the planet to some degree (at least after fight/makeup sex). Adults are aware and have knowledge regarding sex. Sex among adults is a human need.
You can't ban things based on what people do after the fact.
Remind me why marijuana is illegal again?
not quite hitler-like but certainly the worst president in all of US history. yes, far worse than even nixon.
Have a heart!
The politically correct response here is that Bush is clearly the best president since Clinton.
Personally, I believe he and Rumsfeld should be outed.
#echo [0-100] > /proc/sys/vm/swappiness
/proc/sys/vm/swappiness'
I hope this is not your example of how Linux is ready for the mainstream.
No, this is:
for i in `seq -f 'node%04.0f' 1 2048`
do
rsh $i 'echo 100 >
done
Granted its a cheesy example, but I would rather do that on 2,000+ Linux machines vs playing around with a registry key on 2,000+ Windows machines any day.
exactly... and some OSs (read: OSX) caches less-frequently used data (cached window contents, and other images, etc) to the drive to free up real RAM; it doesn't matter how much RAM is installed on the machine, it'll still use the swap. Even my machine at work with 8GB of RAM frequently uses the swap even before 1/4 of the RAM has been touched
:)
:)
exactly back at you
Its also good for an OS to page out "dead code". Things like initialization code or stale data or whatever so that real RAM can be used for disk cacheing and allowing new apps to do their thing.
It frustrates me when a user runs something that takes all of the memory on a box, but it happens. Hell, I've even done it compiling some code where the preprocessor ran my box out of memory. Some seriously nasty C macros. Gotta love those new CS guys
ANY choice made in IT means some kind of lock-in.
Sure, but its what you are locking into and to what degree.
UNIX and Linux (excluding OS X) is much more stable in terms of APIs, backwards compatibility, open standards, and the like than Apple OSes or Microsoft OSes. No, this is not absolute. Yes, I've frantically debugged some code I wrote after applying patches to an AIX box. Yes, I have had much more issues with Windows and OS X (with and before OS X) with gotchas after updates.
To put this in perspective, Sun has a current patch set for Solaris 2.5.1 that came out August 16th of this year. 2.5.1 came out over 10 years ago. Where I work, it takes months to validate a Windows service pack and document its gotchas. When I update my Macs, its a crapshoot if everything is going to be OK.
UNIX and Linux are THE OSes for server side "real work" (TM). Migration within and between them is relatively easy. You are making some kind of a lock-in, but that is a generic platform lock-in, not a specific vendor lock-in. Apache or any other web server will run just fine on any UNIX or Linux box. Sure, it runs on Windows too, but there are tons of gotchas and differences between UNIX and Linux vs Windows. NFS is basically the same on AIX, Solaris, *BSD, Linux. NFS is available on Windows too, but its going to cost you and its robustness is going to be on your reputation, not mine.
you never know when some runaway process is going to eat all yer RAM and need to use swap... no matter how much RAM you've got.
Personally, I prefer a runaway process to run out of resources and stop vs take over my whole system. It takes a long time to page out 1+ Gigs of RAM. It takes a long time to unpage all of that at shutdown or even when an app is closed.
Swap completely depends on the computer's real RAM available and the purpose of the computer and the OS on said computer.
To adequately answer the question, "How much Virtual Memory is Enough?" The correct answer is "It depends".
Having too much swap on a HPC type of machine is a nightmare and will kill performance. Having too little swap on a general purpose server (moreso real RAM) is going to hurt performance. Paging out too much on a laptop with a slow disk can be very painful and slow down the shutdown process.
There is no right answer.
Well, given that 80% of their $5.6 billion in operating profit (from TFA) is derived from ink and toner, they are not going to allow it to slip away. Otherwise they could potentially lose about $4bn in profit.
I'm not defending it, cause I don't agree with it. But they're not going to let that bone go any time soon.
But the thing is that people are getting pissed off at the cheap, but low quality hardware and the chronic extortion-like scheme for keeping ink and toner in the printers.
So, for yet another poor analogy, car companies are going to be owned by oil companies, and cars would suffer in quality and be loss leaders that were inefficient in fuel consumption to recoup their costs.
Personally, I believe the printer market is just at a low point driven by dumb consumers who demand cheap crap. They are starting to learn, and the market will change to have quality printers again, but that will take time. Look at the quality of cars today. Cars today are much better than cars from 15 years ago. Same with computers. This is not true for your average consumer level printer. They suck.
Cool, but it's really nothing more than a bunch of servers in a single rack with big hard drives.
/ 0418253&from=rss Bunches of good replies about the box there.
:)
Actually, it is or can be one big disk or "just" a JBOD. Yes, each box is just a regular server running Linux, but the setup can go like this: internal RAID5 or so within each box, then concatenate the drives with RAID 0 over each of the raid5.
Its up to you. For at least one of the previous slashdot stories about this take a look here: http://linux.slashdot.org/article.pl?sid=05/06/22
Personally, I want one or 10 of them
RAID cards are another example of a commercially successful offloading hardware device. It is possible to do it in software, but most systems do it in hardware.
True, but only on the high end, price has no object range.
Cheaper RAID cards are not worth using because in the event of a card failure the odds of you losing your data is still great. I've heard that some of these things cannot read drives that were setup with a different firmware revision of the same brand of card.
Again, for an in box card, I would opt to use the general purpose CPU over a dedicated RAID card. Now whatever is in my hardware RAID array is a different story. I don't know or care. Its a black box to me, and its from a reputable vendor. But I would trust generic SCSI or (S)ATA cards and software RAID from someone like OS X or Linux or Solaris over something like one of those cheap raid cards. If I care that much about my data, I'm going to spend over $200 for a solution. Otherwise, I clearly don't care about my data.
Hmm... Do I smell WinModem?
I'm not sure what that smell is, but its familiar.
Yes, TCP/IP offloaders, crypto offloaders, physics offloaders, FFT offloaders, have all existed. The only accepted offloader that has succeeded is the GPU, and that is because it was subsidized by the high end graphics people and people with game addictions. The cost/benefit of the other offloaders has not proven itself. Especially when you consider the rate of increase of the CPU speeds and the bottleneck of getting the data to the offloaded chip and back again. The Linux kernel mailinglist has been pretty much anti TCP/IP offloading because the time spent optimizing the drivers and the performance increase was often surpassed in a few months with a faster general purpose CPU and a generic driver that worked with all cards. This is also confounded when you have OS level TCP/IP stuff like ipchains or iptables that need their code in the OS and not on the card. FFT and physics offloaders have not taken off because of the cost/benefit loss when one takes into account the speed loss of getting the data onto the card and then the cost of the cache RAM to put on the card (if its ever enough), and then the time to get the data off of the card.
Now, what will make these things work?
A bus that is near or at the speed of memory bandwidth or a problem where the data does not need to go back to the host system. A GPU falls into the second half there. The display needs to know what its told, the computer does not need to know the details. TCP/IP offloaders have failed to catch on because of the price/performance benefit and their lack of ubiquity and commonality. Crypto offloaders are cool. Especially for the geek factor of having the keys stored on the card and zeroed out when tampered with. But again, the cost of writing specific drivers vs known CPU drivers for doing crypto and the rate of increase of cheap commodity CPUs over time often exceeds the cost/benefit of the crypto card.
Now, give me a fast bus and the ability to use my generic system RAM with one of these offloading cards, and things could change, but until then I expect this battle between the CPU and specific PUs to continue with no real winner.
If you want value and performance custom built is the way to go, not prebuilt crap.
If you want a car considering value and performance custom built is the way to go, not prebuilt crap. False
If you want a home considering value and performance custom built is the way to go, not prebuilt crap. False
If you want a piece of furniture considering value and performance custom built is the way to go, not prebuilt crap. False
If you want a piece of clothing considering value and performance custom built is the way to go, not prebuilt crap. False
If you want a wristwatch considering value and performance custom built is the way to go, not prebuilt crap. False
If you want a television considering value and performance custom built is the way to go, not prebuilt crap. False
If you want a computer considering value and performance custom built is the way to go, not prebuilt crap. True???
I simply do not believe that computers are the only commodity consumer device that requires the knowledge and skill and desire to hand build said commodity item better than any company with assembly lines in China, Japan, or the United States can do.
What is so special about computers? And is there a market for a commodity company to provide such value and performance products since there is no current company providing such things in the world today?
For those who have not escaped mom's basement in years, we would like to know...
Could do a long post... but easier just to point to this /. post that was already up with MANY MANY good links.
d =10155070
This could be the best post from that article: http://ask.slashdot.org/comments.pl?sid=120583&ci
This is probably the only time I would defend the slashdot editors about a dupe.
This whole Pluto thing has got to be the most ridiculous "news" event of my entire life.
At least in my life, it was the Terri Schiavo "story".
As Manson said, "I was crazy when crazy meant something". Well, I remember when controversy was controversial.
100 million is a pretty big payout for an obvious way to navigate through music that I myself invented when I was a kid. This method is: "The patent covers an interface that lets users navigate through a tree of expanding options, such as selecting an artist, then a particular album by that artist, then a specific song from that album."
I mean, isn't that how the stuff is organized in the record store too?
"No wireless. Less space than a nomad. Lame." 100 million dollars in patent taxes lame.
I think its great that everyone thinks Venture Capitalists are complete idiots. If thats true then where did they get the money to invest in the first place?
Their father.
Is that your variation of the "Test this for a while and if you like it, buy it!"
Yes. And also, I cannot try most software for free. They make me give away my life information just to see if their crappy software works or not, and usually it doesn't.
Case in point. I currently burn my music CDs with a script that I wrote and cdrecord. I read on the flac website that the latest and greatest Roxio Toast Titanium allows you to burn directly from flac, thus avoiding the uncompress step. Cool I thought. So I downloaded a version via bittorent. Well, it was a waste of my time and not worth the $0 that I paid for it. The songs had a pop or gap or something between the tracks, so I actually lost on my CD media and my time trying out this crappy software, and I don't use it any more in favor of my scripts.
Now, when software is like any other product where I can just return it to the store and say "It sucks, give me a refund", then I will change my behavior. But as it stands now, I've paid thousands of dollars in software, and if I could have used it first, I would not have done so. To me, the software industry, like the music industry, has taught me that its a dumb decision to pay my hard earned money for their products. And so, I go for alternatives or create my own.
The bands explicitly say the stuff is:
http://wiki.etree.org/index.php?page=TradeFriendl
Yeah, I'm relatively sure that 99% of torrent download ARE legitimate things like linux distros..
99% of the torrents I download are legitimate things. Those that aren't are usually software that I want to try and then don't use because they are inferior to freely available alternatives.
Oh, and 99% of the legitimate 99% is music. Its just legal http://bt.etree.org/ and http://www.archive.org/audio for starts.
If I recall correctly, Heroin was originally designed the same way, or at least to help people get off of a morphine addiction.
Oops! It turned out to be even more addictive, oh well, let's try again. hehe
Rinse and repeat with methadone.
I have 7 letters to add.
JVC LCoS
Granted, its not as thin as plasma, but no picture is better.
This is a recipe for a nightmare when it comes to backup, security, and disaster recovery. It's usually also a terribly inefficient (and therefore, expensive) use of disk space and in my experience it often results in document versioning problems which take hours to unravel--you know, Fred and Alice are both working on that proposal, each of them has it on their desktop, making mutually exclusive changes simultaneously.
The disk space comes with the desktop machines and is cheap. If Fred and Alice are both modifying a document on a shared drive doesn't versioning come into play or aren't the documents locked out? Also, its up to the users to put their data on the shared and backed up network drive because their desktop is not backed up.
Now if you wanted to make a bomb out of liquids - why not just bring a bottle of water and a piece of rubidium or cesium? Remember what happens when alkali metals hit water? BOOM! Two grams of cesium and a quart of water is enough to make an explosion roughly equivalent to about three or so hand grenades going off. Water and rubidium can blow apart a bathtub. Cesium is far, far more reactive.
Thanks for now limiting the number of things I can now bring on a plane. I thought the latest thing with not allowing liquids on board was bad, but now you're telling people that bringing 2 grams or more of metal onto a plane is now banned.
Thanks.