Imperial stuff does work in nearly every field. That why guys who know how to do fractions can hand build rafters in houses with simple tools. I have yet to see a metric version of the imperial roofers square that "just works". In most of the building industry, very few things are the size they claim to be but they work together in a system. A modern 1/2 pipe has no dimension that is 1/2 inch (just like a modern 12 mm pipe or is that 13 mm?). Carpet is sold by yards but its width assumes normal installation wastage so a 12 foot roll will fit in a 12 ft room that might be 12'1" wide (which is typical). The metric world seems to be working hard on recreating the foot since nearly all building materials are based on multiplies of 300 mm units (what I call a "metric foot") I've never seen someone confuse feet and inches but I've seen lots of people drop a an order of magnitude in the metric system. I've even seen furniture at Ikea that claims it was 8 cm wide. I've also noticed that people who can estimate in feet tend to get their numbers about +/- 2 feet when guessing at room sizes but metric people tend to be +/- 2 meters. I'm not convinced that humans and metric are such a good match anymore.
A few years ago there were over 20 different "miles" used in the US. A mile of rail line was different than a mail of railroad bridge but if you put a mile (down to the inch) of railroad on a mile of rail bridge (also down to the inch) it would fit and thermal expansion wouldn't cause problems ever.
A litre just happens to be the average of all the pints used at the time down to about 3 decimal places.
That 250g steak was about 100g+/-. 2x4's aren't 2" or 4" but they are real handy for making 6 inch (nominal- which it isn't) thick walls. Was that 1/2 inch router or 12 mm or 13mm? It seems that the metric bits are 13mm but the chuck is 12mm so it won't fit. The cops in Australia figure if you say someone is 5'11 they figure +/- 2 inches. If you say the suspect is 1.8m they will figure 1.8 +/-.2. I wonder why that is....
The Egyptians didn't have decimals but they did have fractions. There is also some evidence that they didn't convert units. That was sort of like the British up until the 1960's when you bought some things like groceries in pence, paid rent in shillings and bought property in pounds and rarely if ever needed to convert between the two and when you did, the bank would take care of all of it for you. It was like two (3 if you count pounds) different currencies at the same time.
Decimal arithmetic wasn't widely taught in all US schools until the introduction of "New Math" in the late '60s and early '70s. Feet and inches are trivial if you can deal with fractions.
What if Jobs convinced Cingular to flip their business model. Instead of the "free phone but we get you on the minutes" business plan, its a "free data but we get you on the phone"? The high end phones leave the factories at a cost of less then $100 so the phone subsidy business model won't be around much longer.
Re:I've been using vi for so long...
on
The Birth of vi
·
· Score: 2, Informative
Where does the:wq come from? It was superseded decades with the introduction of:x which gets rid of the ambiguity of:wq! meaning:w!q or:w q!
Does MRTG count as heavy floating point? I would think not. The T1000 was bought to replace the decade old box. The odd thing is just doing simple counting in unoptimized C using gcc sometimes the machine is quick and other times its very slow. Its almost like it had a turbo button like early PC's that gets pressed.
Why isn't iWeb included in iWork? Most small corps want a web page creation tool and they might buy iWork but they aren't going to be buying iLife too.
Which machines are you referring to? I've seen performance graphs of the 8 core x4600 being slower than a 4 core for some problems (it was way beyond abuse of the hyper channel bus but it showed the typical Amdahl's law curve) I've got a 6 core T1000 that runs a great deal of tasks slower than a decade old 4 CPU sparc server 1000. That code is highly optimized and in some cases hand tuned. The T1 cpu is great for massively inefficient code (like Java maybe?) but its not so impressive when coded to the metal. Their other bigger stuff (10k, 15k) is somewhat isolated from the Amdahl's law's point of view.
The x86 running CP/M 86 wouldn't run most CP/M 80 code since it wasn't very compatible for several reasons. Sure the op codes looked about the same and lots of instructions would result in some of the same byte code but it still required going through the entire program line by line checking every assembly instruction. Add in that the x86 assembler was more of a assembly language compiler than a macro assembler made it a real pain to port software. CP/M and nearly every application by that time required a Z80 system and wouldn't run on an 8080 based system so the 8088/8086 upgrade path was much different than the Z80 to Z8000 translation. The Z8000 would run the Z80 based CP/M code with no change. The 8088 choice had no technical advantage other than the cheaper. IBM had 3 different designs that they picked between. One was a Z8000, one was 680?0 and the other was 8088. They picked the only one that didn't make their minis look silly as well as the most expensive to make.
The Kidall's theory goes out the window when you consider IBM already had a license agreement for CPM and was already making better x86 based machines than the PC (like the DisplayWrite which had a faster CPU, more memory potential and CPM) but instead of using a well engineered machine to take on what they considered "toy computer". Their intent was to put Apple and Tandy out of the computer business and then regain their market on the "real computers".
I took more than one call from IBM's sales people trying to get me to consider upgrading to a better computer at 10 times the cost. A friend had a job of putting a flyer in boxes at the factory that offered to take the band new PC as a trade in for some thing like a 360 or whatever was their current mini at the time. The temporary workers at the factory (in Boca Raton) were told that the assembly line on the new PC would shut down with very little notice and were offered a very nice severance package. Even IBM's management had odd attitudes about the product until about two years after the XT.
IBM picked the x86 for the original IBM PC because they wanted something better than an Apple ]{ and TRS-80 but not too much better because they didn't want to sell $5000 computers, they preferred selling $30,000 computers. So we got stuck with mediocrity.
They were dimmable back in '91 or so. If you have a typical incandescent dimmer, you will need CFL with the right kind of inverter and those are even more inefficient that the typical cheap type so your power usage won't go down much. A 60W bulb at 1/2 power (and less than 1/2 light) will burn about 30 w (except that modern 60W bulbs use less than 60W) while a dimmed CFL will use even more power as its dimmed and 15W dimmable CFL may easily take up to 30W to produce 1/2 its normal light. There are dimmer switches that are made for CFLs that you might consider putting in.
The 5 year old bulbs are more likely to last longer than the ones made last month. The last batch of bulbs I got from Ikea are not outlasting the last set of traditional globes I already have installed. I've got 4 down-lights in my kitchen on the same switch and 2 are CFL and two are incandescent and the incandescent lights haven't been replaced in two years but I've put in 5 CFL.
The last 15 W CFL I took apart put 18 W into the tube. Its inverter was massively inefficient so it was on par with a modern 60 w incandescent for power use. If they new bulbs are like the one I disassembled, then I'm not saving any energy at all with the more expensive CFLs and I'm creating even more toxic waste in the process. The energy costs to make the CFL means it started out 500 hours behind the green curve before it ever was turned on.
It will be cheaper to build your own data centres in the long run than to contract them out. For a large gov't project, you have to have two for disaster recovery. Selecting the sites will be a political mess but the real issues are "1) where is power and air conditioning cheap?" and "2) where can we get staff" For this I would propose you put your main data center near a major hydro dam and have your secondary site near where most of your existing staff is. That way when your main site goes down, your people are where they need to fix the most problems. It also helps build up the staff into proper procedures of working remotely which can lead to all sorts of advantages that the federal gov't isn't too happy about even considering. There are also a number of nice military data centers that the US gov't already owns.
If there is any data in the que when the helo line is processed, the message can be discarded. RFC pipelining happens after the server claims it can do it. Turning off pipelineing will only slow down some mailing lists
Maybe the best solution is to stop filtering at all for a bit. Let everyone know just how bad the problem is. This was a technique used in the Usenet community every once in a while to let more people know just how much work is being done behind the scenes.
I propose that we turn off all RBLs and filters for 24 hrs the day before congress sits for the 1st time in the new year.
Laws won't help because its an international issue. The only way to stop it is to follow the money and while two systems exist to do that, they aren't even used by the banks that pay for them. I'm talking about MasterCard and Visa anti fraud departments that have the ability to pull merchant accounts so fast that it will make a dodgy ISP's head spin. Both groups need to get together (which they do anyway) and decided that ISPs that host phishing sites will deal with them on their terms instantly and not in the slow ways that tend to happen now. The same goes for domain registrars. Do you think godaddy and NSI will start getting a human to read the applications if they have have their merchant account turned off? I'm guessing both companies would be bankrupt in a month if their ability to take credit cards was pulled. I'm wondering why the banks won't push this. The banks own MasterCard and Visa and the banks are the ones that pay when the scammers win.
Once phishing is fixed, then they can use the same system to nail spam.
The delay won't help if they already sent the entire data. I wrote a patch to log the size of the rest of the data in the TCP buffer for sendmail a long time ago and that was a great way to locate the high speed spam senders. Another trick is change the hello message to be more than one line but that breaks some major mailers.
I've seen the VOR needle move when the GSM clicks started in the speakers and my instructors phone rang. The error was about 5 degrees. With VORs being about 100 miles apart, that could lead to a large enough error to wander into a no-fly zone.
I don't see a problem in good weather or at cruise. I do have a problem with phones being used during ILS approaches in the clouds with tight tolerances. A phone inside a big tin can isn't going get very good reception until it gets close to an airport. Of course thats thats when the instruments are most essential. As the plane descends into the clouds the pilot needs to be able to see the runway at somewhere between 200 and 800 feet. The last think you want is a phone waking up and transmitting at full power trying to say hello to a tower when you need to make sure your radar altimeter and glide slope gauges are exactly where they should be. The real problem is that once you allow phones to be turned on, there will be no way to get them turned off for the real nasty approaches.
They mention 4500 artists and many of them haven't done anything in years. A few years ago a Melbourne radio station had a contest where bands could send them a CD for some prize. Their listening area was about 3 million people and they got 3000 albums. Their rules required that at least one song not be a cover and the CD had to have been made in the previous year. That would imply that for every 1000 people, you would have at least 1 band that made an album with at least one original song in the last year. Now multiply that by the population of the countries that give us popular music and you end up with English speaking countries making over 400,000 new albums a year. Sure lots of them will be bad but how many are good if more than few under people every get a chance to hear them? The 4500 artists at the top are worried about those number and the RIAA is deathly afraid of that many new albums entering their marketplace and are doing everything in their power to stop it.
Slashdot Mirror
Imperial stuff does work in nearly every field. That why guys who know how to do fractions can hand build rafters in houses with simple tools. I have yet to see a metric version of the imperial roofers square that "just works".
In most of the building industry, very few things are the size they claim to be but they work together in a system. A modern 1/2 pipe has no dimension that is 1/2 inch (just like a modern 12 mm pipe or is that 13 mm?). Carpet is sold by yards but its width assumes normal installation wastage so a 12 foot roll will fit in a 12 ft room that might be 12'1" wide (which is typical). The metric world seems to be working hard on recreating the foot since nearly all building materials are based on multiplies of 300 mm units (what I call a "metric foot")
I've never seen someone confuse feet and inches but I've seen lots of people drop a an order of magnitude in the metric system. I've even seen furniture at Ikea that claims it was 8 cm wide. I've also noticed that people who can estimate in feet tend to get their numbers about +/- 2 feet when guessing at room sizes but metric people tend to be +/- 2 meters. I'm not convinced that humans and metric are such a good match anymore.
French wine comes in 750ml bottles and 12 to a case. Didn't they invent the metric system?
A few years ago there were over 20 different "miles" used in the US. A mile of rail line was different than a mail of railroad bridge but if you put a mile (down to the inch) of railroad on a mile of rail bridge (also down to the inch) it would fit and thermal expansion wouldn't cause problems ever.
A litre just happens to be the average of all the pints used at the time down to about 3 decimal places.
1cm^3 of pure water does weigh 1g at sea level but 1.000 cm ^3 of water at sea level doesn't weigh 1.000g does it?
That 250g steak was about 100g+/-. .2.
2x4's aren't 2" or 4" but they are real handy for making 6 inch (nominal- which it isn't) thick walls.
Was that 1/2 inch router or 12 mm or 13mm? It seems that the metric bits are 13mm but the chuck is 12mm so it won't fit.
The cops in Australia figure if you say someone is 5'11 they figure +/- 2 inches. If you say the suspect is 1.8m they will figure 1.8 +/-
I wonder why that is....
The Egyptians didn't have decimals but they did have fractions. There is also some evidence that they didn't convert units. That was sort of like the British up until the 1960's when you bought some things like groceries in pence, paid rent in shillings and bought property in pounds and rarely if ever needed to convert between the two and when you did, the bank would take care of all of it for you. It was like two (3 if you count pounds) different currencies at the same time.
Decimal arithmetic wasn't widely taught in all US schools until the introduction of "New Math" in the late '60s and early '70s. Feet and inches are trivial if you can deal with fractions.
What if Jobs convinced Cingular to flip their business model. Instead of the "free phone but we get you on the minutes" business plan, its a "free data but we get you on the phone"?
The high end phones leave the factories at a cost of less then $100 so the phone subsidy business model won't be around much longer.
Where does the :wq come from? It was superseded decades with the introduction of :x which gets rid of the ambiguity of :wq! meaning :w!q or :w q!
Does MRTG count as heavy floating point? I would think not. The T1000 was bought to replace the decade old box.
The odd thing is just doing simple counting in unoptimized C using gcc sometimes the machine is quick and other times its very slow. Its almost like it had a turbo button like early PC's that gets pressed.
Why isn't iWeb included in iWork? Most small corps want a web page creation tool and they might buy iWork but they aren't going to be buying iLife too.
Which machines are you referring to?
I've seen performance graphs of the 8 core x4600 being slower than a 4 core for some problems (it was way beyond abuse of the hyper channel bus but it showed the typical Amdahl's law curve)
I've got a 6 core T1000 that runs a great deal of tasks slower than a decade old 4 CPU sparc server 1000. That code is highly optimized and in some cases hand tuned. The T1 cpu is great for massively inefficient code (like Java maybe?) but its not so impressive when coded to the metal.
Their other bigger stuff (10k, 15k) is somewhat isolated from the Amdahl's law's point of view.
The x86 running CP/M 86 wouldn't run most CP/M 80 code since it wasn't very compatible for several reasons. Sure the op codes looked about the same and lots of instructions would result in some of the same byte code but it still required going through the entire program line by line checking every assembly instruction. Add in that the x86 assembler was more of a assembly language compiler than a macro assembler made it a real pain to port software. CP/M and nearly every application by that time required a Z80 system and wouldn't run on an 8080 based system so the 8088/8086 upgrade path was much different than the Z80 to Z8000 translation. The Z8000 would run the Z80 based CP/M code with no change. The 8088 choice had no technical advantage other than the cheaper. IBM had 3 different designs that they picked between. One was a Z8000, one was 680?0 and the other was 8088. They picked the only one that didn't make their minis look silly as well as the most expensive to make.
The Kidall's theory goes out the window when you consider IBM already had a license agreement for CPM and was already making better x86 based machines than the PC (like the DisplayWrite which had a faster CPU, more memory potential and CPM) but instead of using a well engineered machine to take on what they considered "toy computer". Their intent was to put Apple and Tandy out of the computer business and then regain their market on the "real computers".
I took more than one call from IBM's sales people trying to get me to consider upgrading to a better computer at 10 times the cost. A friend had a job of putting a flyer in boxes at the factory that offered to take the band new PC as a trade in for some thing like a 360 or whatever was their current mini at the time. The temporary workers at the factory (in Boca Raton) were told that the assembly line on the new PC would shut down with very little notice and were offered a very nice severance package. Even IBM's management had odd attitudes about the product until about two years after the XT.
Amdahl's law.
IBM picked the x86 for the original IBM PC because they wanted something better than an Apple ]{ and TRS-80 but not too much better because they didn't want to sell $5000 computers, they preferred selling $30,000 computers. So we got stuck with mediocrity.
They were dimmable back in '91 or so. If you have a typical incandescent dimmer, you will need CFL with the right kind of inverter and those are even more inefficient that the typical cheap type so your power usage won't go down much. A 60W bulb at 1/2 power (and less than 1/2 light) will burn about 30 w (except that modern 60W bulbs use less than 60W) while a dimmed CFL will use even more power as its dimmed and 15W dimmable CFL may easily take up to 30W to produce 1/2 its normal light. There are dimmer switches that are made for CFLs that you might consider putting in.
The 5 year old bulbs are more likely to last longer than the ones made last month. The last batch of bulbs I got from Ikea are not outlasting the last set of traditional globes I already have installed. I've got 4 down-lights in my kitchen on the same switch and 2 are CFL and two are incandescent and the incandescent lights haven't been replaced in two years but I've put in 5 CFL.
The last 15 W CFL I took apart put 18 W into the tube. Its inverter was massively inefficient so it was on par with a modern 60 w incandescent for power use. If they new bulbs are like the one I disassembled, then I'm not saving any energy at all with the more expensive CFLs and I'm creating even more toxic waste in the process. The energy costs to make the CFL means it started out 500 hours behind the green curve before it ever was turned on.
It will be cheaper to build your own data centres in the long run than to contract them out.
For a large gov't project, you have to have two for disaster recovery. Selecting the sites will be a political mess but the real issues are "1) where is power and air conditioning cheap?" and "2) where can we get staff"
For this I would propose you put your main data center near a major hydro dam and have your secondary site near where most of your existing staff is. That way when your main site goes down, your people are where they need to fix the most problems. It also helps build up the staff into proper procedures of working remotely which can lead to all sorts of advantages that the federal gov't isn't too happy about even considering.
There are also a number of nice military data centers that the US gov't already owns.
If there is any data in the que when the helo line is processed, the message can be discarded. RFC pipelining happens after the server claims it can do it. Turning off pipelineing will only slow down some mailing lists
Maybe the best solution is to stop filtering at all for a bit. Let everyone know just how bad the problem is. This was a technique used in the Usenet community every once in a while to let more people know just how much work is being done behind the scenes.
I propose that we turn off all RBLs and filters for 24 hrs the day before congress sits for the 1st time in the new year.
Laws won't help because its an international issue. The only way to stop it is to follow the money and while two systems exist to do that, they aren't even used by the banks that pay for them. I'm talking about MasterCard and Visa anti fraud departments that have the ability to pull merchant accounts so fast that it will make a dodgy ISP's head spin. Both groups need to get together (which they do anyway) and decided that ISPs that host phishing sites will deal with them on their terms instantly and not in the slow ways that tend to happen now. The same goes for domain registrars. Do you think godaddy and NSI will start getting a human to read the applications if they have have their merchant account turned off? I'm guessing both companies would be bankrupt in a month if their ability to take credit cards was pulled. I'm wondering why the banks won't push this. The banks own MasterCard and Visa and the banks are the ones that pay when the scammers win.
Once phishing is fixed, then they can use the same system to nail spam.
Why not just buy an x.400 email server? It does all that already.
The delay won't help if they already sent the entire data. I wrote a patch to log the size of the rest of the data in the TCP buffer for sendmail a long time ago and that was a great way to locate the high speed spam senders.
Another trick is change the hello message to be more than one line but that breaks some major mailers.
I've seen the VOR needle move when the GSM clicks started in the speakers and my instructors phone rang. The error was about 5 degrees. With VORs being about 100 miles apart, that could lead to a large enough error to wander into a no-fly zone.
I don't see a problem in good weather or at cruise. I do have a problem with phones being used during ILS approaches in the clouds with tight tolerances. A phone inside a big tin can isn't going get very good reception until it gets close to an airport. Of course thats thats when the instruments are most essential. As the plane descends into the clouds the pilot needs to be able to see the runway at somewhere between 200 and 800 feet. The last think you want is a phone waking up and transmitting at full power trying to say hello to a tower when you need to make sure your radar altimeter and glide slope gauges are exactly where they should be. The real problem is that once you allow phones to be turned on, there will be no way to get them turned off for the real nasty approaches.
They mention 4500 artists and many of them haven't done anything in years.
A few years ago a Melbourne radio station had a contest where bands could send them a CD for some prize. Their listening area was about 3 million people and they got 3000 albums. Their rules required that at least one song not be a cover and the CD had to have been made in the previous year. That would imply that for every 1000 people, you would have at least 1 band that made an album with at least one original song in the last year. Now multiply that by the population of the countries that give us popular music and you end up with English speaking countries making over 400,000 new albums a year. Sure lots of them will be bad but how many are good if more than few under people every get a chance to hear them? The 4500 artists at the top are worried about those number and the RIAA is deathly afraid of that many new albums entering their marketplace and are doing everything in their power to stop it.