All modern astrophotography done in an observatory is done with digital CCD technology.
True but misleading. CCD's used in astrophotography are not like those in digital camers.
They have active cooling systems. Active cooling greatly reduces noise, especially on long exposures. Needless to say, digital SLR's do not have active cooling systems. Astronomers also use a technique called "binning". Adjacent sensor elements are combined to make one pixel. Resolution is reduced but noise/sensetivity is improved. This technique is not possible with the Bayer filter arrangement used in virtually all digital cameras.
Can someone explain to me why it's suddenly bad because it's us and not nature. No matter that we ourselves are one of natures mutations and what we do is merely the natural result of that mutation.
We know that nature can handle "natural" disasters because it was handeling them long before we existed. Some natural disasters may even be necessary to keep the system operating. It isn't so clear with man made changes. There is a severe shortage of precident
Then there is the issue of biodiversity. Every extinction we cause means a species, a habitat, an ecosystem will no longer exist for us to study. All those solved problems that we will never know. Oh, nature is resiliant. It is almost certain that no matter how badly we fuck things up, nature and it's biodiversity will recover eventually. But nature is also slow. Many of us would rather not wait the millions of years that nature would take to recover from our abuses.
The business reason behind this is that the phones are subsidised. You pay an up front fee for the phone, but that is nowhere near the actual cost of the phone. The provider then recovers the rest of the cost over the lifetime of your contract.
That's what contracts are for. There is no need to lock the phone except to discourage customers from leaving *after* their contract is up and the phone is paid for.
I don't see a whole lot point to running fiber to the home until backhaul costs come down.
The speed of most consumer broadband services is limited by the cost of the backhaul, not the performance of the local loop. If my area, 6Mbps DSL is available for those who can afford it. It's the same wire and the same hardware at both ends. Most people stick with 1.5Mbps becuase that is all they can afford.
"Consumer" fiber optic systems are required to maintain eye safety. Usually, this just means low enough power so that if you do point a fiber at your retina, there is no dammage. That's that gigabit ethernet does. Around the time that 10Gb Ethernet was being developed, there were some proposals to have the system detect a fiber cut and shut down the laser. I don't know if this has been implimented.
The Amiga was a fanstastic machine, way ahead of it's time.
However, the world ignored the Amiga. It's influence on today's machine is negligable.
Amiga style light weight multitasking never caught on. Machines just became fast enough to not need it. No machines today that support multiple resolutions simultaneously. Native NTSC timing died with the Amiga as well.
Let's face it. trying to email angel22@hotmail.com has a good chance of hitting a valid box simply due to the number of accounts.
Wrong. angel22 has a very low chance of being hit if angel1-21 don't exist. That's what it means to have a sparce name space. angle4135 might exist and so might a couple dozen others but they are spread randomly between 1 and 5000.
Also the router can not detect the attacks you mention because the router does not know what the recipients are or whether the transmissions failed due to nosuchuser.
No. I got your point exactly. You just aren't understanding. Adding digits on the right side of the @ sign has the exact same effect on the name space as adding digits on the left side of the @ sign.
Worse, your method will not detect technician@mail0001, technician@mail0002, technician@mail0003, etc. because each server only sees one attempt.
Background noise only matters if the checker is a human. You wouldn't do it that way, even with only 5000 users. Even if you did want to break up the logs and check them by hand, segregating by mail server is not a particularly effective method. For instance, alphabetically by receipient would work better.
I think one of the things that is overlooked in reducing the success of dictionary attacks is what mega ISP's can do. They need to divide up their mail servers so each would have no more than say about 5000 mailboxes. It would make the addresses a little longer maybe. Instead of having an address such as technician126@msn.com, I would have an address like technician@mail3275.msn.com
It doesn't matter which side of the @ site you put the number on. technician@mail3275.msn.com is no harder to hit than technician3275@msn.com. The important part is that the real users need to scattered randomly among a sparce name space rather than assigned sequentially among a dense one.
technician@mail3275.msn.com is actualy more vulnerable than technician3275@msn.com. In the former, the size of the name space can be found with DNS lookups and need only be done once for all usernames. For the later, the spammer has to actually send mail.
All you need to detect a dictionary attack is a sparcely populated name space and the ability to detect attempts to send to the missing addreseses. If a single host, in a short time, racks up a high percentage of failures with a large number of email addresses, then you just found a dictionary attack. .
Small domains are less vulnerable to dictionary attacks becuase their name space is already more sparce than megaISP's can practically be.
That's a common misconception. A 1.4 crop means only half the light falls on the sensor. It only seems free becuase of bit of slight of hand in the way sensor sensitivity is measured.
Digital sensor elements are activated by photons. More photons, less noise. However, ISO numbers are based on photon *density*. Shrink the sensor while keeping the optics the same and fewer photons reach the sensor. The noise level goes up but the ISO number stays the same.
Thus, the noise level that can be achived at ISO200 with full frame, requirers dropping to ISO100 with a 1.4 crop.
Some routes are unwalkable. Freeways, for instance. Others are technically possible but a really bad idea.
What is needed is the ability to select roads *not* to use. This would make bicycle routing much more effective. It would also be useful for driving directions. You may know that a particular road is undesireable do to construction, predictable congestion, etc. It would be really handy to able to say "I know 101 is jammed, what's the next best way"
Ignoring one way streets is good but correcting for speed is just unnecessary. Just read the distance.
1) It isn't scroll lock that is equivilient to Xoff. That's Pause.
2) The extra keys were not added in response to programs that took control of the keyboard. The extra keys were there from the very first IBM PC keyboard when it was expected that all IO would go through the BIOS.
In late January, my server rebooted for reasons unknown (probably a power outage). The web server didn't come up properly. And that means no email.
I sat down at a Nairobi internet cafe and downloaded putty off the net. After about an hour of painful editing and debug, I had the problem fixed.
Originally, I thought I would carry around a floppy with putty on it. But I discovered that it didn't really help. If it was at all possible to use putty, it was easy and fast enough to download it.
The hypothetical backpacker's laptop is a tool to facilitate email communications when all your posessions must fit into a backpack and all Internet access takes place in Internet cafes.
The requirements are:
1) cheap (in case it gets lost/stolen) 2) light 3) durable (whether on airplanes or chicken
trucks, luggage tends to get kicked around) 4) Has a keyboard and reasonable size screen. 5) Has a floppy drive.
The last one may not be obvious to those who haven't traveled in this manner. The only way you can reliably move data between a laptop and the PC at an Internet cafe is with a floppy. For everything else, the machines will either be too primitive or policies too restrictive.
A side note: Has anyone managed to get a USB equiped PDA to talk directly to a USB floppy?
Second, that's a hell of a lot of TV; I don't want to let the thing fill up, because when will I possibly find the 100 hours to watch everything it records?
I think that's the point. Or at least it's my point. With a TIVO, my own tendency to procrastinate works toward the goal of watching less TV. If I can watch the show anytime I want then I don't need to watch it right now. So, instead I'll do something else that is not so flexible. Repeat a few times, and maybe I'll never get arround to watching the show. That's probably a good thing. When I do get arround to watching anything, I cherry pick the good stuff and discard the rest.
The trouble comes if the storage is too small. Then I face the delima: If I don't watch it now, it will be gone later. Maybe I should watch it now. To prevent this from happenning, make the disk real big.:-)
Simple, really. And, unlike using colored windows, the prompt actually stays correct whenever you ssh from one machine to another.
I also put the hostname in the window titlebar. 'Useful for finding the apropriate window when it is iconified. I haven't figured out how to make the title bar actually track the hostname though.
Faster than what Netscape? Netscape 6 maybe, I'll give you that. But IE5 renders much more slowly than NS4.*. As for reliability, it's hard to say. IE runs too slowly for normal use on my SS10HS125. Thus, I never run it for any length of time. There is also the total lack of plug-ins.
At this point, it is not ecomical to run ADSL at full rate. The hardware can run at 7Mbps, but actual service plans cap that to 1.5Mps (if you are lucky). Further, it is unusual to see close to 1.5Mhps performance in real life.
So what would fiber buy you? $$$$ price. 10x the theoretical bandwidth. And no change in actual performance.
Re:Could you imagine...
on
Star In A Jar
·
· Score: 1
If people have eternal life then maybe they won't be in such a hurry to have kids. When inter-generation time streaches toward infinity, growth gets real slow.
I have an email address that I stopped useing entirely 4 years ago. It still receives more than twice as much spam as my zone contact address. The address I use here and on Usenet gets about the same as my nic address.
Slashdot Mirror
All modern astrophotography done in an observatory is done with digital CCD technology.
True but misleading. CCD's used in astrophotography are not like those in digital camers.
They have active cooling systems. Active cooling greatly reduces noise, especially on long exposures. Needless to say, digital SLR's do not have active cooling systems. Astronomers also use a technique called "binning". Adjacent sensor elements are combined to make one pixel. Resolution is reduced but noise/sensetivity is improved. This technique is not possible with the Bayer filter arrangement used in virtually all digital cameras.
We know that nature can handle "natural" disasters because it was handeling them long before we existed. Some natural disasters may even be necessary to keep the system operating. It isn't so clear with man made changes. There is a severe shortage of precident
Then there is the issue of biodiversity. Every extinction we cause means a species, a habitat, an ecosystem will no longer exist for us to study. All those solved problems that we will never know. Oh, nature is resiliant. It is almost certain that no matter how badly we fuck things up, nature and it's biodiversity will recover eventually. But nature is also slow. Many of us would rather not wait the millions of years that nature would take to recover from our abuses.
The business reason behind this is that the phones are subsidised. You pay an up front fee for the phone, but that is nowhere near the actual cost of the phone. The provider then recovers the rest of the cost over the lifetime of your contract.
That's what contracts are for. There is no need to lock the phone except to discourage customers from leaving *after* their contract is up and the phone is paid for.
I don't see a whole lot point to running fiber to the home until backhaul costs come down.
The speed of most consumer broadband services is limited by the cost of the backhaul, not the performance of the local loop. If my area, 6Mbps DSL is available for those who can afford it. It's the same wire and the same hardware at both ends. Most people stick with 1.5Mbps becuase that is all they can afford.
"Consumer" fiber optic systems are required to maintain eye safety. Usually, this just means low enough power so that if you do point a fiber at your retina, there is no dammage. That's that gigabit ethernet does. Around the time that 10Gb Ethernet was being developed, there were some proposals to have the system detect a fiber cut and shut down the laser. I don't know if this has been implimented.
The Amiga was a fanstastic machine, way ahead of it's time.
However, the world ignored the Amiga. It's influence on today's machine is negligable.
Amiga style light weight multitasking never caught on. Machines just became fast enough to not need it. No machines today that support multiple resolutions simultaneously. Native NTSC timing died with the Amiga as well.
Let's face it. trying to email angel22@hotmail.com has a good chance of hitting a valid box simply due to the number of accounts.
Wrong. angel22 has a very low chance of being hit if angel1-21 don't exist. That's what it means to have a sparce name space. angle4135 might exist and so might a couple dozen others but they are spread randomly between 1 and 5000.
Also the router can not detect the attacks you mention because the router does not know what the recipients are or whether the transmissions failed due to nosuchuser.
No. I got your point exactly. You just aren't understanding. Adding digits on the right side of the @ sign has the exact same effect on the name space as adding digits on the left side of the @ sign.
Worse, your method will not detect technician@mail0001, technician@mail0002, technician@mail0003, etc. because each server only sees one attempt.
Background noise only matters if the checker is a human. You wouldn't do it that way, even with only 5000 users. Even if you did want to break up the logs and check them by hand, segregating by mail server is not a particularly effective method. For instance, alphabetically by receipient would work better.
I think one of the things that is overlooked in reducing the success of dictionary attacks is what mega ISP's can do. They need to divide up their mail servers so each would have no more than say about 5000 mailboxes. It would make the addresses a little longer maybe. Instead of having an address such as technician126@msn.com, I would have an address like technician@mail3275.msn.com
It doesn't matter which side of the @ site you put the number on. technician@mail3275.msn.com is no harder to hit than technician3275@msn.com. The important part is that the real users need to scattered randomly among a sparce name space rather than assigned sequentially among a dense one.
technician@mail3275.msn.com is actualy more vulnerable than technician3275@msn.com. In the former, the size of the name space can be found with DNS lookups and need only be done once for all usernames. For the later, the spammer has to actually send mail.
All you need to detect a dictionary attack is a sparcely populated name space and the ability to detect attempts to send to the missing addreseses. If a single host, in a short time, racks up a high percentage of failures with a large number of email addresses, then you just found a dictionary attack. .
Small domains are less vulnerable to dictionary attacks becuase their name space is already more sparce than megaISP's can practically be.
I mean, come on! are we all that afraid that WE'RE overpaid?
Not me. I'm unemployed.
That's a common misconception. A 1.4 crop means only half the light falls on the sensor. It only seems free becuase of bit of slight of hand in the way sensor sensitivity is measured.
Digital sensor elements are activated by photons. More photons, less noise. However, ISO numbers are based on photon *density*. Shrink the sensor while keeping the optics the same and fewer photons reach the sensor. The noise level goes up but the ISO number stays the same.
Thus, the noise level that can be achived at ISO200 with full frame, requirers dropping to ISO100 with a 1.4 crop.
Some routes are unwalkable. Freeways, for instance. Others are technically possible but a really bad idea.
What is needed is the ability to select roads *not* to use. This would make bicycle routing much more effective. It would also be useful for driving directions. You may know that a particular road is undesireable do to construction, predictable congestion, etc. It would be really handy to able to say "I know 101 is jammed, what's the next best way"
Ignoring one way streets is good but correcting for speed is just unnecessary. Just read the distance.
That's fine except for a couple of things:
1) It isn't scroll lock that is equivilient to Xoff. That's Pause.
2) The extra keys were not added in response to programs that took control of the keyboard. The extra keys were there from the very first IBM PC keyboard when it was expected that all IO would go through the BIOS.
Linux beer hike? That's outrageous! Linux beer is supposed to always be free as in, er... Whatever! Isn't that what the GPL is about?
If it's truly unlimited then surely any unlicensed band will do, no matter how small.
Why do you are what the FCC does with the rest?
In late January, my server rebooted for reasons unknown (probably a power outage). The web server didn't come up properly. And that means no email.
I sat down at a Nairobi internet cafe and downloaded putty off the net. After about an hour of painful editing and debug, I had the problem fixed.
Originally, I thought I would carry around a floppy with putty on it. But I discovered that it didn't really help. If it was at all possible to use putty, it was easy and fast enough to download it.
The hypothetical backpacker's laptop is a tool
to facilitate email communications when all your
posessions must fit into a backpack and all
Internet access takes place in Internet cafes.
The requirements are:
1) cheap (in case it gets lost/stolen)
2) light
3) durable (whether on airplanes or chicken
trucks, luggage tends to get kicked around)
4) Has a keyboard and reasonable size screen.
5) Has a floppy drive.
The last one may not be obvious to those who haven't traveled in this manner. The only way
you can reliably move data between a laptop and
the PC at an Internet cafe is with a floppy. For
everything else, the machines will either be too
primitive or policies too restrictive.
A side note: Has anyone managed to get a USB
equiped PDA to talk directly to a USB floppy?
I think that's the point. Or at least it's my point. With a TIVO, my own tendency to procrastinate works toward the goal of watching less TV. If I can watch the show anytime I want then I don't need to watch it right now. So, instead I'll do something else that is not so flexible. Repeat a few times, and maybe I'll never get arround to watching the show. That's probably a good thing. When I do get arround to watching anything, I cherry pick the good stuff and discard the rest.
The trouble comes if the storage is too small. Then I face the delima: If I don't watch it now, it will be gone later. Maybe I should watch it now. To prevent this from happenning, make the disk real big. :-)
Basic quality is more certainly not equal to VHS. In my experience, High quality is roughly equal to, maybe a little better than VHS EP.
Basic quality isn't just jerky, it distorts the colors. Even low frame rate animation looks bad. Useless.
I do most of my recording at High Quality. Some animation is OK at medium quality.
A "30 hour" TIVO records about hours of 15 hours at High Quality on a 36GB disk. So a good rule of thumb is 2GB/hour of usable quality.
Simple, really. And, unlike using colored windows, the prompt actually stays correct whenever you ssh from one machine to another.
I also put the hostname in the window titlebar. 'Useful for finding the apropriate window when it is iconified. I haven't figured out how to make the title bar actually track the hostname though.
Faster than what Netscape? Netscape 6 maybe, I'll give you that. But IE5 renders much more slowly than NS4.*. As for reliability, it's hard to say. IE runs too slowly for normal use on my SS10HS125. Thus, I never run it for any length of time. There is also the total lack of plug-ins.
At this point, it is not ecomical to run ADSL at full rate. The hardware can run at 7Mbps, but actual service plans cap that to 1.5Mps (if you are lucky). Further, it is unusual to see close to 1.5Mhps performance in real life.
So what would fiber buy you? $$$$ price. 10x the theoretical bandwidth. And no change in actual performance.
Even better, get them all running SETI@HOME
If people have eternal life then maybe they won't be in such a hurry to have kids. When inter-generation time streaches toward infinity, growth gets real slow.
I have an email address that I stopped useing entirely 4 years ago. It still receives more than twice as much spam as my zone contact address. The address I use here and on Usenet gets about the same as my nic address.