AIUI the trouble is that fedora/redhat have a lot of influence with certain upstreams. This can make it difficult for other distros to resist the changes they are pushing.
f it's an intermittent fault and it's not happening when they visit, they assume it's your equipment at fault and charge you for the visit.
Worse than that they charge your service provider for the visit who then charges you. So AIUI it's your service provider who has to fight to try and get you to pay a fee that should never have been charged (and who ultimately may end up having to decide between eating the cost and losing a customer) not openreach.
What used to happen with debian is that the resident geek (aka me) would set a system up then the system's owner/user would install updates that introduced a new "abi version" of the kernel and on teh next reboot X would fail to load rendering the system unsable for them. Yeah it wasn't exactly difficult to fix if you knew what you were doing (just re-run module-assistant auto-install nvidia)but it was a PITA if you were setting up systems for other people.
Nowadays there is dkms which provided you get things setup right in the first place (e.g. make sure you install the kernel headers metapackage so the new kernel headers get pulled in when the new kernel does) mostly mitigates this issue but that is a fairly recent development.
There are also some miguided souls who use NVIDIA's installer to install the driver rather than the debian packages. That WILL end badly when a kernel update comes down.
I'd expect a load of computers behind NAT to create a big traffic storm if they were all rebooted at once but it would then subside as the computers backed off to. Also the traffic would likely be relatively irregular.
A regular request every second sounds like a mark of a client developed by someone who either didn't understand how NTP was supposed to work or didn't care about the load they were putting on donators
Afaict they are generally known as "convertible tablets" or "tablet PCs" and you can still buy them from a few vendors, the trouble with them has been
1: they are VERY expensive, somewhere arround double the price of an equivilently specced regular laptop. 2: they were too bulky and heavy for practical tablet use 3: touchscreen interfaces (both the software and the hardware) were really not very well developed, so most of the time you had to use a stylus.
Of course, if you don't like your ISP, you may not want to do that. However, why are you dealing with them at all in that case?
IME large ISPs give by far the best bang for your buck on connectivity BUT the competent people are burried behind a wall of support drones. If I was going to get a mailing list hosted commercially (personally i'd rather do it myself on a rented vm or dedicated server) I would want it to be done by a smaller company where competant people are more likely to be accessible.
I think google must have realised they threw the baby out with the bathwater when they brought in nofollow and either started ignoring it or at least treating it as merely an advisory note along the lines of "this link may be user contributed and hence lower importance than others on the site".
While the "old web" of traditional websites set up by individuals and linking to each other still exists more and more content is moving to wikis, forums and similar or to pages that are primerally accessed by following links (or sometimes copy/pasting urls if the forum doesn't allow links) from those wikis and forums. Heck many projects are now using wikis as their main websites. A search engine that wants to find the best results doesn't want to completely ignore this information just because there is some risk it is spam-contaiminated.
mmm, tfa doesn't make it clear whether the latency figures they are using are one-way latency or round trip time. Either way though 200ms is a pretty low value to use as the highest latency testcase. UK to australia is about 300ms round trip time datacenter to datacenter and a sattelite connection or crappy cellular connection can easily have over a second's latency.
The parent is semi-correct, HTTP pipelining allows the client to send multiple requests sent at a time but the responses to those requests have to be sent sequentially and in the same order that the requests were sent.
This means that if one request is slow then all requests behind it in the queue are also held up.
The arm instruction set is the original instruction set (extended many times over the years) of the arm series and is 32-bit fixed width.
Thumb1 is a 16-bit fixed width instruction set, it was handy when you had very limited memory and/or very limited memory bandwidth but the performance penalty from the reduced instruction set was too high to make it a good choice for general use.
Thumb2 is a variable width instruction set with a mixture of 16-bit and 32-bit instructions. Raw core performance is lower than with the arm instruction set but cache pressure is also lower so AIUI overall performance is comparable to the arm instruction set. Afaict this is the instruction set that arm is pushing people to use at the moment.
Some older electronics with PSUs that use mains frequency transformers and whose design was close to the edge may have problems as may some stuff that uses mains as a time reference but mostly electronics should be fine.
Clocks (whether electronic or mechanical) that derive their timebase from the mains would be a nuisance but ultimately if it was the main issue I think they would have forced a transition through by now.
Afaict the real problem is the big stuff, big motors and generators are usually at least somewhat locked to grid frequency and a 10% change in operating speed is probablly not acceptable. Transformers can also be problematic as a lower frequency can cause core saturation leading to overheating. Replacing that stuff would be seriously expensive.
You can never prove someone is trustworthy and what you can get done without trusting people is largely limited to what you can do yourself.
If you want to get anything done that is too big to accomplish on your own you have to trust people. Extending that trust always carries an element of risk. There are two approaches to trying to reduce that risk, one is to try and foster mutual respect between you and those you share your secrets with. The other is to try and make it harder for those who have access to those secrets to maliciously release them. A third is to try and monitor for any releases of secrets so you can punish those responsible.
Getting back on topic monitoring SSL traffic like this has pros and cons
Pros: if information is leaked you may be able to figure out who leaked it by going back through the logs. The threat of this may stop people leaking information. You may be able to stop leaks through keyword triggers etc.
Cons The person in control of the monitoring equipment is in a massively abusable position. If they wanted they could trivially reconfigure the equipment to store passwords for every service your customers accessed including online banking etc. If your employees find out about this it fosters distrust in their relationship with you, especially if you did not tell them clearly in advance.
Suppose you have a regular laptop which has a hdd bay and an odd bay and none of this fancy msata shit, you have decided you want to upgrade your storage what are your options.
1: conventional hard drive only, lots of space for your buck but slow under random access 2: SSD only, great random access speeds but large SSDs start to get very expensive (though not as expensive as they were a year or so ago) 3: SSD and conventional HDD, fast random access for your programs and lots of storage for your data but you lose the optical drive and you have to order an adaptor from a specialist supplier to do it. I suspect many people, even fairly geeky people don't realise this is an option (I didn't until someone pointed it out to me on/.) 4: get a hybrid drive which should offer plenty of storage, give fast random access for the frequently used stuff, be reasonablly affordable and let you keep your optical drive.
So hybrids can be attractive, but only to the relatively small niche who want performance but also need or think they need lots of capacity.
up to about a year ago we saw a trenth of lots of volatility combined with a general expontential growth trend the value of bitcoin leading to a high of about $30 per bitcoin.
Then we saw a trend ot lots of volatility combined with a general exponential decay reaching a lot of arround $2 per bitcoin. After that we saw more peaks and troughs but each time they got smaller and it seems that bitcoin is now finally starting to settle on a stable value.
Looks like the fear and greed driven speculators are finally getting out of the market:)
Of course, the catch is no encryption for amateur radio
Another catch is the forbidding of commercial traffic.
But the biggest catch is scaling, meshes work fine for low data rates but you quickly reach a point where it is difficult to add more capacity in the places where you need it.
Furthermore the latency is likely to be bad due to the very large number of hops.
Personally I highly doubt that a mesh network will ever offer better quality of service than even "best effort" internet traffic over a typical DSL/cable connection.
And i'm saying the only the only reason that most threats are not viable is that the US wouldn't stand for it. Their security comes not from any active action by the US military but by potential aggressors knowing that if they attacked Canada they would find themselves dealing with the full force of the US military.
The problem I see with your idea is it would encourage massive vertical integration. To minimise taxes a company would want to do everything from raw mining of the ore to assembly of the final product within one corporation so that the only wealth transfers that happened were the purchase of the mining rights and the sale of the final product. Smaller companies who could only do one part of the value chain would have a massive tax disadvantage over megacorps who could do all of it.
The Internet is the worst example of all, it's entirely privately owned and operated.
The internet is a network of networks, some public some private. Nowadays most networks in the internet are private but it started out as a system for connecting together research networks, commercialisation came later.
This tool calculates the brute-force time on a character basis. It says that dictionary attacks still work and should be mitigated by policy and practice.
What it doesn't emphasise is that there is a world between the "dumb dictionary attack" (try all words in a dictionary) and the "dumb brute force attack" (try all combinations of characters in an alphabet). That world is what a smart attacker will inhabit.
Something like "Mother!fucker" would't be cracked by either the "dumb dictionary attack" and the "dumb brute force attack" but that does not mean it is a strong password.
Slashdot Mirror
The blocks in that picture look more line mini mega blocks than duplo to me (stud pitch is the same but studs are taller)
It's not Linux that's regressing, it's Fedora.
AIUI the trouble is that fedora/redhat have a lot of influence with certain upstreams. This can make it difficult for other distros to resist the changes they are pushing.
f it's an intermittent fault and it's not happening when they visit, they assume it's your equipment at fault and charge you for the visit.
Worse than that they charge your service provider for the visit who then charges you. So AIUI it's your service provider who has to fight to try and get you to pay a fee that should never have been charged (and who ultimately may end up having to decide between eating the cost and losing a customer) not openreach.
1 binary terabyte ~= 1.100 metric gigabytes
oops typo that should have said "1 binary terabyte ~= 1.100 metric terabytes"
The other thing that happened is that with each prefix we moved up the difference got bigger.
1 binary kilobyte = 1.024 metric kilobytes
1 binary megabyte ~= 1.049 metric megabytes
1 binary gigabyte ~= 1.074 metric gigabytes
1 binary terabyte ~= 1.100 metric gigabytes
What used to happen with debian is that the resident geek (aka me) would set a system up then the system's owner/user would install updates that introduced a new "abi version" of the kernel and on teh next reboot X would fail to load rendering the system unsable for them. Yeah it wasn't exactly difficult to fix if you knew what you were doing (just re-run module-assistant auto-install nvidia)but it was a PITA if you were setting up systems for other people.
Nowadays there is dkms which provided you get things setup right in the first place (e.g. make sure you install the kernel headers metapackage so the new kernel headers get pulled in when the new kernel does) mostly mitigates this issue but that is a fairly recent development.
There are also some miguided souls who use NVIDIA's installer to install the driver rather than the debian packages. That WILL end badly when a kernel update comes down.
I'd expect a load of computers behind NAT to create a big traffic storm if they were all rebooted at once but it would then subside as the computers backed off to. Also the traffic would likely be relatively irregular.
A regular request every second sounds like a mark of a client developed by someone who either didn't understand how NTP was supposed to work or didn't care about the load they were putting on donators
The criminals don't have to use the stolen details in the country they stole them from.
Afaict they are generally known as "convertible tablets" or "tablet PCs" and you can still buy them from a few vendors, the trouble with them has been
1: they are VERY expensive, somewhere arround double the price of an equivilently specced regular laptop.
2: they were too bulky and heavy for practical tablet use
3: touchscreen interfaces (both the software and the hardware) were really not very well developed, so most of the time you had to use a stylus.
Note that this ikea "house" is actually a 1 bedroom bungalow ......
Of course, if you don't like your ISP, you may not want to do that. However, why are you dealing with them at all in that case?
IME large ISPs give by far the best bang for your buck on connectivity BUT the competent people are burried behind a wall of support drones. If I was going to get a mailing list hosted commercially (personally i'd rather do it myself on a rented vm or dedicated server) I would want it to be done by a smaller company where competant people are more likely to be accessible.
I think google must have realised they threw the baby out with the bathwater when they brought in nofollow and either started ignoring it or at least treating it as merely an advisory note along the lines of "this link may be user contributed and hence lower importance than others on the site".
While the "old web" of traditional websites set up by individuals and linking to each other still exists more and more content is moving to wikis, forums and similar or to pages that are primerally accessed by following links (or sometimes copy/pasting urls if the forum doesn't allow links) from those wikis and forums. Heck many projects are now using wikis as their main websites. A search engine that wants to find the best results doesn't want to completely ignore this information just because there is some risk it is spam-contaiminated.
mmm, tfa doesn't make it clear whether the latency figures they are using are one-way latency or round trip time. Either way though 200ms is a pretty low value to use as the highest latency testcase. UK to australia is about 300ms round trip time datacenter to datacenter and a sattelite connection or crappy cellular connection can easily have over a second's latency.
The parent is semi-correct, HTTP pipelining allows the client to send multiple requests sent at a time but the responses to those requests have to be sent sequentially and in the same order that the requests were sent.
This means that if one request is slow then all requests behind it in the queue are also held up.
The arm instruction set is the original instruction set (extended many times over the years) of the arm series and is 32-bit fixed width.
Thumb1 is a 16-bit fixed width instruction set, it was handy when you had very limited memory and/or very limited memory bandwidth but the performance penalty from the reduced instruction set was too high to make it a good choice for general use.
Thumb2 is a variable width instruction set with a mixture of 16-bit and 32-bit instructions. Raw core performance is lower than with the arm instruction set but cache pressure is also lower so AIUI overall performance is comparable to the arm instruction set. Afaict this is the instruction set that arm is pushing people to use at the moment.
Some older electronics with PSUs that use mains frequency transformers and whose design was close to the edge may have problems as may some stuff that uses mains as a time reference but mostly electronics should be fine.
Clocks (whether electronic or mechanical) that derive their timebase from the mains would be a nuisance but ultimately if it was the main issue I think they would have forced a transition through by now.
Afaict the real problem is the big stuff, big motors and generators are usually at least somewhat locked to grid frequency and a 10% change in operating speed is probablly not acceptable. Transformers can also be problematic as a lower frequency can cause core saturation leading to overheating. Replacing that stuff would be seriously expensive.
You can never prove someone is trustworthy and what you can get done without trusting people is largely limited to what you can do yourself.
If you want to get anything done that is too big to accomplish on your own you have to trust people. Extending that trust always carries an element of risk. There are two approaches to trying to reduce that risk, one is to try and foster mutual respect between you and those you share your secrets with. The other is to try and make it harder for those who have access to those secrets to maliciously release them. A third is to try and monitor for any releases of secrets so you can punish those responsible.
Getting back on topic monitoring SSL traffic like this has pros and cons
Pros:
if information is leaked you may be able to figure out who leaked it by going back through the logs. The threat of this may stop people leaking information.
You may be able to stop leaks through keyword triggers etc.
Cons
The person in control of the monitoring equipment is in a massively abusable position. If they wanted they could trivially reconfigure the equipment to store passwords for every service your customers accessed including online banking etc.
If your employees find out about this it fosters distrust in their relationship with you, especially if you did not tell them clearly in advance.
Suppose you have a regular laptop which has a hdd bay and an odd bay and none of this fancy msata shit, you have decided you want to upgrade your storage what are your options.
1: conventional hard drive only, lots of space for your buck but slow under random access /.)
2: SSD only, great random access speeds but large SSDs start to get very expensive (though not as expensive as they were a year or so ago)
3: SSD and conventional HDD, fast random access for your programs and lots of storage for your data but you lose the optical drive and you have to order an adaptor from a specialist supplier to do it. I suspect many people, even fairly geeky people don't realise this is an option (I didn't until someone pointed it out to me on
4: get a hybrid drive which should offer plenty of storage, give fast random access for the frequently used stuff, be reasonablly affordable and let you keep your optical drive.
So hybrids can be attractive, but only to the relatively small niche who want performance but also need or think they need lots of capacity.
It's certainly interesting to look at what has happened with the value of bitcoin.
http://bitcoincharts.com/charts/mtgoxUSD#tgSzm1g10zm2g25zvzl
up to about a year ago we saw a trenth of lots of volatility combined with a general expontential growth trend the value of bitcoin leading to a high of about $30 per bitcoin.
Then we saw a trend ot lots of volatility combined with a general exponential decay reaching a lot of arround $2 per bitcoin. After that we saw more peaks and troughs but each time they got smaller and it seems that bitcoin is now finally starting to settle on a stable value.
Looks like the fear and greed driven speculators are finally getting out of the market :)
Of course, the catch is no encryption for amateur radio
Another catch is the forbidding of commercial traffic.
But the biggest catch is scaling, meshes work fine for low data rates but you quickly reach a point where it is difficult to add more capacity in the places where you need it.
Furthermore the latency is likely to be bad due to the very large number of hops.
Personally I highly doubt that a mesh network will ever offer better quality of service than even "best effort" internet traffic over a typical DSL/cable connection.
Note that I specified viable threat.
And i'm saying the only the only reason that most threats are not viable is that the US wouldn't stand for it. Their security comes not from any active action by the US military but by potential aggressors knowing that if they attacked Canada they would find themselves dealing with the full force of the US military.
The russians could easilly bomb canada without flying over any other country.
Of course they won't actually do that because they know the US would not stand for it.
The problem I see with your idea is it would encourage massive vertical integration. To minimise taxes a company would want to do everything from raw mining of the ore to assembly of the final product within one corporation so that the only wealth transfers that happened were the purchase of the mining rights and the sale of the final product. Smaller companies who could only do one part of the value chain would have a massive tax disadvantage over megacorps who could do all of it.
The Internet is the worst example of all, it's entirely privately owned and operated.
The internet is a network of networks, some public some private. Nowadays most networks in the internet are private but it started out as a system for connecting together research networks, commercialisation came later.
This tool calculates the brute-force time on a character basis. It says that dictionary attacks still work and should be mitigated by policy and practice.
What it doesn't emphasise is that there is a world between the "dumb dictionary attack" (try all words in a dictionary) and the "dumb brute force attack" (try all combinations of characters in an alphabet). That world is what a smart attacker will inhabit.
Something like "Mother!fucker" would't be cracked by either the "dumb dictionary attack" and the "dumb brute force attack" but that does not mean it is a strong password.