One problem: using cell phones in NYC is tricky, they frequently drop signal.
Walk along a block and signal strength varies enourmously. A means to autoreconnect and re-establish an authenticated session would be needed both in the server and the software on the palmtop/laptop/mobile PC.
When I last worked at IBM (a long time back) we had many quads of S/390's and 3090's. Our departmental quad consisted of 4 mainframes, as the term suggests, this was because mainframes crash. Strange really that even big computers crash, but they do.
So in the TCO calculations we have 11 PC servers with one being a backup (wholly inadequate in my opinion) compared with a single mainframe, without a backup.
Toss in a new backup mainframe and these TCO costs are very different.
In 1991 I wrote a white paper to our management expressing how mainframes could be used as SuperServers. Of course, it was expected the clients have the hardware - no one in their right mind would buy one for anything but a massive company. If your company has a mainframe in the z class hanging out forlornly in the corner then it might pay to try this approach. To go out and buy a mainframe from IBM to host your email on would be economic suicide. And afterall, what else could a normal company based in PC servers or Unix systems put on a mainframe that might make economic sense?
If they built roads like we make software...
on
Software Aesthetics
·
· Score: 1
A server is meant to stay up for 99.999% of some time period, normally a year.
If a Motorway or Interstate were so built it would require 3.5 hrs of maintenance in 40 years, it would never be stopped for snow, car pileups, dear crossings, or grannies loosing their marbles and driving the wrong way down them.
Interstates are the complex mix of rebar and concrete laid in more or less straight lines. They have no natural beauty.
So why the heck need software that has to operate under the same rules need beauty?
And why does this dolt think that goto's are pretty.
Ellegance in software creation is at the writers discretion, if it is specified in the requirements then use a code beautifier, if its not then just make sure that the next sucker to maintain it has enough guidance to be able to carry out the job effectively.
I suppose the cosmonauts that rode MIR for more than a year were some how in a different form of Zero-G. I suppose that Russian Space being different to US Space, their research is useless... And assuming this means that some Dunderhead can still get Federal Tax Dollars to do research that someone else has already done.... better.
But by this definition, Microsoft is well within reason doing the dreaded bundling of IE. N'est pas?
And just because I feel adversity coming on...
M$ = Win2K Server = $800 (say, 800 quid - Oh! How did that happen?) And 5 of yer best friends can use the server, too!
And that means...
webserver = IIS (though I wish they would fix it, my Apache server is getting tired of rejecting requests for default.ida),
DNS server = MS-DNS (It's not that bad),
Kerberos KDC = Errr Sort of,
DHCP = The home of...,
SSHd = ya don't like NT network logon?,
Samba = Err, I don't think it needs this.
1 MSDN subscription = $500 (= 10 licenses)
2 SuSE releases, 2 RH releases, and a zillion Linux distro's to test (@ 1.95 each) = c. $150.
Of course, other than Kylix, I have not bought any Linux software, while I have splurged on various Windoze games, apps and development tools (many shareware goodies too - Yes, I paid for my WinZip!)
Looking at the pictures it's amazing to think that in 20 years the weight of a laptop has shrunk from 15lbs to... 14lbs?
They got me a Compaq Portable, never a real screamer, but the 2x5.25" drives by the green phosphur screen were very useful. It was such a delight to transport it places (Not!). We eventually used it as a serial bus analyzer, where it did fine duty till it died in late 88.
As old laptops go, the best of an early breed was the Tandy/Radio Shack T1000. It had a vast 320x240 pixel screen, NiCad's that lasted just long enough for you to walk away from the recharger, and a heaving great 80c88. Wow, powwa!
Oddly we used this laptop in late 1987 to do downloads across a network to update databases in a real time control system. How many patents do you think this would fubar?
Autogyros are not new, they were invented in the 1920's. They haven't, and will not, change the way things are done. New technology will improve them and make them somewhat more useful, or should I say less useless.
Runway slots are the limiting factor at airports, the time it takes to roll down the runway with the brakes on, then turn off, and still have a safe distance from the aircraft in front of, and behind you. Airspeed does not play much of a role in this, A modern jet has a stall speed of about 160kts, they can even be landed in about 1000ft (with passengers smeared on the forward bulkhead).
Autogyros are not magic carpets, they cannot arbitrarily stop. They hover, by the grace of God, and loose huge amounts of energy doing it. Nor are Autogyros VTOL, they just have short take off runs and short landing runs.
Remove the pilot from the loop and every possible failure mode you can think of, and the 1,000,000 others you haven't a clue about, all have to be programmed into the computer. Now let's see... A pilot takes 5 years and $1million to train... a room full of programmers, living on nothing but Jolt and Twinkies for 20 years, would cost... humm Big Number! The airforce tried this one, keeps trying it, and doesn't actually have to worry about the plane returning to the ground... gravity will assist if the computer goes AWOL. I think your 400 passengers might be a bit litigious if they found gravity was in total control of the landing criteria.
It has been said (by whom I do not recall) of WWII fighter pilots that 10% had all the kills, 10% always had something wrong with their plane and returned to base, and the other 80% just enjoyed the blue skies and tried to avoid being a zap on the side of some else's cockpit.
Whether this is true or not, it does represent what we see in business today. Mediocrity is with us. The meek have indeed inherited the world, and look what they are doing to it.
We have to use this bulk of people in the same way a dam is made of cold gray concrete. We have to use these marginally capable people to stem the flood. There are too many jobs for most IT departments to actually complete. We don't have enough good programmers to complete them all. We have to pick our targets well, complete them with panache, and hope that the vast unwashed horde of programmers we have working on everything else holds back the rabid users.
I feel for your position, but the true Prima Donna (in Programming or Management) would being saying... This is a no-win situation, Adios banditos, I'm outta here. Which is exactly what you hoped all your mediocre players would say.
Of course if the thought of that pension is the thing that is keeping you put... hunker down, put up with the shit, and count the days. It's what everyone else in Government seems to be doing.
No way is Java a C++ alternative, nor a C alternative. It is an application programming tool. It does the fluffy bits well, if slowly.
If I need to set a Pentium's CR0 register, it's hard in C++, impossible in Java.
C++ & C are systems programming tools meant to create not just applications but the systems that the applications sit on.
Java, VB. PB, Perl, etc, are application languages that are easily used to implement higher level functionality - just don't try writing OS's, or system gadgets, in them and you find out what inefficient means.
It is this basic understanding that is being frittered away, and to which the original article was aimed: the erosion of understanding of what the essential elements of our world are. We can make management organisations because they don't have to think well ("throw enough peoples ideas and comments together, it's bound to be right") but we have problems putting engineering teams together because basic and essential skills, hopefully learnt at an early age, are missing.
Make a robot in Meccano/Erector Set, and you have some idea about why there are limits on joint movement. Make it in Lego and you see no reason why we can't all have one.
Make a BIOS in assembler and you know why you need interrupts, write it in Java and you wonder why anyone would want a BIOS, it's just so slow!
As many companies are now outsourcing their systems to ASP's and other forms of providers, the ability to arbitrarily hack the data becomes moot. It's hard enough for most of the managers that decide on the outsoucing to comprehend what they have committed their company to, let along hack in and alter scandalous data.
In this neck of the woods, a company I worked for (whose stock symbol rhymes with dirty) was stuck in the middle of two warring Pharma companies. One believed the other had exceeded their contracted limits on pimping some drug to hospitals. So, we had to search the database for references to hospital visits, and the comments made. This, as you might imagine, was a fairly heady piece of SQL.
I doubt such data alone would be used to prove a legal point, but to provide background info it is without a doubt very useful. In this instance, the resulting data set was megabytes. I doubt a jury could be kept alive, let along awake, long enough to trudge through it all.
I think it may have been Knuth that was called in to a court room a decade ago to give testimony on code that had been stolen. His observation was that the stolen code had the same spacetabspace structure that the originating companies code had. Tell tale marks like this (the proverbial smoking gun) can make high court drama. While code and data in our eyes (as programmers) look very different, to the lay person they probably look quite similar. In this instance code was data.
As the hacking court cases have often fallen to the display or at least analysis of third party logs, I would think that the place of raw data in the court room is well established. How much a lawyer can safely display is an altogether, and entirely different question.
And sadly it wasn't even Olivetti that did the original work, I do believe it was the old Acorn boys (of BBC computer fame). There was a Tomorrows World show about it in the late 80's.
When this mess started I felt that it was a no win situation. If MS won then Java was going to be fragmented, if Sun won then Java was going to be left in the hands of a bunch of clowns. So here we are; Microsoft lost the battle, Sun will lose the war.
We all lose.
Now we will get brain dead attempts by M$ to cripple JVMs and strange new Java-like languages that only run (slowly) on M$ OS's. Java will become even more of an entrenched Sun-only toy, never getting the input it needs to make it reliable enough for the big league.
In my experience a role like CTO in a small company is one of the top 5 or 6 people in that organization. In such a rarified atmosphere the CTO has a lot of pull. The success of the company is based partly on your ability to do your job. If you are leaving this job because others at your level have out manoeuvered you politically, you probably have to think long and hard about the implication for what comes next.
I think that in any normal role you can quickly say that the circumstances left you no options: Quit. It's no skin off your nose if they go down.
However, in your position as a member of the executive, if that really is your role, then you have to ask yourself the really hoary question: Am I man enough to make a business successful, or am I just another worker ant that wants to be paid well and leave at 5pm?
You have given use a question, but I don't think you have stated all the facts in a manner that allows anyone to give a responsible answer - Maybe you are an executive!
Re:I'm no physics head, but isn't C a constant
on
Stop, Light.
·
· Score: 1
The speed of light is set by the ISO standards people as something to do with the oscilation of electrons in some noble gas. (I'm sure someone can furnish the twiddly details) Every so often, like every few months, some scientist refines this number by a few decimal places, and so c has become the classic "variable constant".
It's one of the reasons I hate the metric system; King Edwards feet were good enough as a constant for a thousand years! How did a system started by Napoleon, that defined the earths circumference at 1 million kilometers become the basis for todays standards!:)
A flash of inspiration...
on
Stop, Light.
·
· Score: 1
So here I am sitting at the console to the Deep Thought and we get a programming error - light beams going everywhere. We stop the program, you know,/usr/bin/Life_the_Universe_and_ Everything.ksh, and rerun in the cool light debugger, get to the offending photon, then just tweek its parameters by adding a subtle disturbance in the eddy current of its existence. Give it the Go laser, and a new planet is instantiated. We're running again. The Mice will get a whooping maintenance bill for this.
Another link to check out in search of odd holes in the ground and old anomolies... Weird New Jersey. The website supports a magazine that can occasionally be found in local book stores.
Proof positive that what the rest of the US has know for years is true... New Jersey is strange.
I'm not sure that much of that sort of run in, run out work is done by the RC-135's. These aircraft are used for ELINT. On the night in question an RC-135 (Cobra Ball - I Think) was in the general vacinity trying to snoop on a Russian missile test. There is some concensous that the CCCP thought it was going to shot down the RC. These aircraft do not enter Russian airspace, they just have to be close enough to either track the signals, or photograph the rentry vehicle (used in the missile test).
Alumni associations are not new, a certain, extremely litigious northern New Jersey Sales Force Automation Company, whose stock ticker symbol rhymes with DIRTY (apropo, really), has had such an Alumni for many years. It's a pity the employers pay more attention to these informal groups than the investors - why would a bunch of strangers put so much energy into shouting about the abuses at a company if there were nothing to shout about?
I think the Chunnel goes through extremely porous rock. The non-porous clay above the tunnel keeps it nominally dry. They might want porous rocks or rocks that are softer (not necessarily one and the same), to be able to dig the tunnel faster. Given that such a project would involve Russian and US Governments, maybe a tunneling rate of 1cm/decade is not out of the question:)
Anyway, how fast would the average dig rate for the chunnel be... it was first envisaged by Napolean in 1803 or somesuch... 22 miles / 180 years is about.1 mile a year. Statistics are wonderfully obscure.
I just reviewed BT's web site looking for backbone numbers, didn't find much. Except... They have a new Transatlantic cable, takes 320K calls at once. Assume 14Kbits/s per call, that gives us about 806Gbytes per day, or your 12-14 Dat tapes, ofcourse it would take about 2 days to write these tapes...
Given that this is just one low speed source, and you would be better off putting a big collector on the main trunk lines; not all the individual sources (duh, who cares about web pages if no one looks at them) need recording. We are probably talking something like 5 or 6 backbone trunks doing about 1Gigbit/s, 24hrs a day.
Sure stream it, keep it, just don't think you will ever be able to do anything with it.
Its because of the bureaucratic nincompoops that make stipulations like these that I left that damp and soggy island, not to mention the fact that English politicians are too stupid to know when to say "cut the taxes."
Slashdot Mirror
I assume the SysProg's were IBMers.
Walk along a block and signal strength varies enourmously. A means to autoreconnect and re-establish an authenticated session would be needed both in the server and the software on the palmtop/laptop/mobile PC.
When I last worked at IBM (a long time back) we had many quads of S/390's and 3090's. Our departmental quad consisted of 4 mainframes, as the term suggests, this was because mainframes crash. Strange really that even big computers crash, but they do.
So in the TCO calculations we have 11 PC servers with one being a backup (wholly inadequate in my opinion) compared with a single mainframe, without a backup.
Toss in a new backup mainframe and these TCO costs are very different.
In 1991 I wrote a white paper to our management expressing how mainframes could be used as SuperServers. Of course, it was expected the clients have the hardware - no one in their right mind would buy one for anything but a massive company. If your company has a mainframe in the z class hanging out forlornly in the corner then it might pay to try this approach. To go out and buy a mainframe from IBM to host your email on would be economic suicide. And afterall, what else could a normal company based in PC servers or Unix systems put on a mainframe that might make economic sense?
A server is meant to stay up for 99.999% of some time period, normally a year.
If a Motorway or Interstate were so built it would require 3.5 hrs of maintenance in 40 years, it would never be stopped for snow, car pileups, dear crossings, or grannies loosing their marbles and driving the wrong way down them.
Interstates are the complex mix of rebar and concrete laid in more or less straight lines. They have no natural beauty.
So why the heck need software that has to operate under the same rules need beauty?
And why does this dolt think that goto's are pretty.
Ellegance in software creation is at the writers discretion, if it is specified in the requirements then use a code beautifier, if its not then just make sure that the next sucker to maintain it has enough guidance to be able to carry out the job effectively.
Expert: Ex, Has been; Spurt, Drip under pressure.
I suppose the cosmonauts that rode MIR for more than a year were some how in a different form of Zero-G. I suppose that Russian Space being different to US Space, their research is useless... And assuming this means that some Dunderhead can still get Federal Tax Dollars to do research that someone else has already done.... better.
Agreed. I have 71 hits over the last 3 days, many more than the first outbreak. 19 on wednesday, 33 yesterday and 19 so far today (10:40 Eastern).
Maybe the thing is getting through its random IP generation faster??
But by this definition, Microsoft is well within reason doing the dreaded bundling of IE. N'est pas?
And just because I feel adversity coming on...
M$ = Win2K Server = $800 (say, 800 quid - Oh! How did that happen?) And 5 of yer best friends can use the server, too!
And that means...
8-)
1 MSDN subscription = $500 (= 10 licenses)
2 SuSE releases, 2 RH releases, and a zillion Linux distro's to test (@ 1.95 each) = c. $150.
Of course, other than Kylix, I have not bought any Linux software, while I have splurged on various Windoze games, apps and development tools (many shareware goodies too - Yes, I paid for my WinZip!)
Looking at the pictures it's amazing to think that in 20 years the weight of a laptop has shrunk from 15lbs to... 14lbs?
They got me a Compaq Portable, never a real screamer, but the 2x5.25" drives by the green phosphur screen were very useful. It was such a delight to transport it places (Not!). We eventually used it as a serial bus analyzer, where it did fine duty till it died in late 88.
As old laptops go, the best of an early breed was the Tandy/Radio Shack T1000. It had a vast 320x240 pixel screen, NiCad's that lasted just long enough for you to walk away from the recharger, and a heaving great 80c88. Wow, powwa!
Oddly we used this laptop in late 1987 to do downloads across a network to update databases in a real time control system. How many patents do you think this would fubar?
Ummm. I am astonished.
Autogyros are not new, they were invented in the 1920's. They haven't, and will not, change the way things are done. New technology will improve them and make them somewhat more useful, or should I say less useless.
Runway slots are the limiting factor at airports, the time it takes to roll down the runway with the brakes on, then turn off, and still have a safe distance from the aircraft in front of, and behind you. Airspeed does not play much of a role in this, A modern jet has a stall speed of about 160kts, they can even be landed in about 1000ft (with passengers smeared on the forward bulkhead).
Autogyros are not magic carpets, they cannot arbitrarily stop. They hover, by the grace of God, and loose huge amounts of energy doing it. Nor are Autogyros VTOL, they just have short take off runs and short landing runs.
Remove the pilot from the loop and every possible failure mode you can think of, and the 1,000,000 others you haven't a clue about, all have to be programmed into the computer. Now let's see... A pilot takes 5 years and $1million to train... a room full of programmers, living on nothing but Jolt and Twinkies for 20 years, would cost... humm Big Number! The airforce tried this one, keeps trying it, and doesn't actually have to worry about the plane returning to the ground... gravity will assist if the computer goes AWOL. I think your 400 passengers might be a bit litigious if they found gravity was in total control of the landing criteria.
It has been said (by whom I do not recall) of WWII fighter pilots that 10% had all the kills, 10% always had something wrong with their plane and returned to base, and the other 80% just enjoyed the blue skies and tried to avoid being a zap on the side of some else's cockpit.
Whether this is true or not, it does represent what we see in business today. Mediocrity is with us. The meek have indeed inherited the world, and look what they are doing to it.
We have to use this bulk of people in the same way a dam is made of cold gray concrete. We have to use these marginally capable people to stem the flood. There are too many jobs for most IT departments to actually complete. We don't have enough good programmers to complete them all. We have to pick our targets well, complete them with panache, and hope that the vast unwashed horde of programmers we have working on everything else holds back the rabid users.
I feel for your position, but the true Prima Donna (in Programming or Management) would being saying... This is a no-win situation, Adios banditos, I'm outta here. Which is exactly what you hoped all your mediocre players would say.
Of course if the thought of that pension is the thing that is keeping you put... hunker down, put up with the shit, and count the days. It's what everyone else in Government seems to be doing.
Yes, plains, big flat things. They are real easy in Lego. Planes however were for real boys, and Meccano.
Sorry, couldn't resist. %-)
No way is Java a C++ alternative, nor a C alternative. It is an application programming tool. It does the fluffy bits well, if slowly.
If I need to set a Pentium's CR0 register, it's hard in C++, impossible in Java.
C++ & C are systems programming tools meant to create not just applications but the systems that the applications sit on.
Java, VB. PB, Perl, etc, are application languages that are easily used to implement higher level functionality - just don't try writing OS's, or system gadgets, in them and you find out what inefficient means.
It is this basic understanding that is being frittered away, and to which the original article was aimed: the erosion of understanding of what the essential elements of our world are. We can make management organisations because they don't have to think well ("throw enough peoples ideas and comments together, it's bound to be right") but we have problems putting engineering teams together because basic and essential skills, hopefully learnt at an early age, are missing.
Make a robot in Meccano/Erector Set, and you have some idea about why there are limits on joint movement. Make it in Lego and you see no reason why we can't all have one.
Make a BIOS in assembler and you know why you need interrupts, write it in Java and you wonder why anyone would want a BIOS, it's just so slow!
To paraphrase a line from "My Cousin Vinny."
As many companies are now outsourcing their systems to ASP's and other forms of providers, the ability to arbitrarily hack the data becomes moot. It's hard enough for most of the managers that decide on the outsoucing to comprehend what they have committed their company to, let along hack in and alter scandalous data.
In this neck of the woods, a company I worked for (whose stock symbol rhymes with dirty) was stuck in the middle of two warring Pharma companies. One believed the other had exceeded their contracted limits on pimping some drug to hospitals. So, we had to search the database for references to hospital visits, and the comments made. This, as you might imagine, was a fairly heady piece of SQL.
I doubt such data alone would be used to prove a legal point, but to provide background info it is without a doubt very useful. In this instance, the resulting data set was megabytes. I doubt a jury could be kept alive, let along awake, long enough to trudge through it all.
I think it may have been Knuth that was called in to a court room a decade ago to give testimony on code that had been stolen. His observation was that the stolen code had the same space tab space structure that the originating companies code had. Tell tale marks like this (the proverbial smoking gun) can make high court drama. While code and data in our eyes (as programmers) look very different, to the lay person they probably look quite similar. In this instance code was data.
As the hacking court cases have often fallen to the display or at least analysis of third party logs, I would think that the place of raw data in the court room is well established. How much a lawyer can safely display is an altogether, and entirely different question.
And sadly it wasn't even Olivetti that did the original work, I do believe it was the old Acorn boys (of BBC computer fame). There was a Tomorrows World show about it in the late 80's.
When this mess started I felt that it was a no win situation. If MS won then Java was going to be fragmented, if Sun won then Java was going to be left in the hands of a bunch of clowns. So here we are; Microsoft lost the battle, Sun will lose the war.
We all lose.
Now we will get brain dead attempts by M$ to cripple JVMs and strange new Java-like languages that only run (slowly) on M$ OS's. Java will become even more of an entrenched Sun-only toy, never getting the input it needs to make it reliable enough for the big league.
Yuck!
In my experience a role like CTO in a small company is one of the top 5 or 6 people in that organization. In such a rarified atmosphere the CTO has a lot of pull. The success of the company is based partly on your ability to do your job. If you are leaving this job because others at your level have out manoeuvered you politically, you probably have to think long and hard about the implication for what comes next.
I think that in any normal role you can quickly say that the circumstances left you no options: Quit. It's no skin off your nose if they go down.
However, in your position as a member of the executive, if that really is your role, then you have to ask yourself the really hoary question: Am I man enough to make a business successful, or am I just another worker ant that wants to be paid well and leave at 5pm?
You have given use a question, but I don't think you have stated all the facts in a manner that allows anyone to give a responsible answer - Maybe you are an executive!
The speed of light is set by the ISO standards people as something to do with the oscilation of electrons in some noble gas. (I'm sure someone can furnish the twiddly details) Every so often, like every few months, some scientist refines this number by a few decimal places, and so c has become the classic "variable constant".
It's one of the reasons I hate the metric system; King Edwards feet were good enough as a constant for a thousand years! How did a system started by Napoleon, that defined the earths circumference at 1 million kilometers become the basis for todays standards! :)
So here I am sitting at the console to the Deep Thought and we get a programming error - light beams going everywhere. We stop the program, you know, /usr/bin/Life_the_Universe_and_ Everything.ksh, and rerun in the cool light debugger, get to the offending photon, then just tweek its parameters by adding a subtle disturbance in the eddy current of its existence. Give it the Go laser, and a new planet is instantiated. We're running again. The Mice will get a whooping maintenance bill for this.
Another link to check out in search of odd holes in the ground and old anomolies... Weird New Jersey. The website supports a magazine that can occasionally be found in local book stores.
Proof positive that what the rest of the US has know for years is true... New Jersey is strange.
I'm not sure that much of that sort of run in, run out work is done by the RC-135's. These aircraft are used for ELINT. On the night in question an RC-135 (Cobra Ball - I Think) was in the general vacinity trying to snoop on a Russian missile test. There is some concensous that the CCCP thought it was going to shot down the RC. These aircraft do not enter Russian airspace, they just have to be close enough to either track the signals, or photograph the rentry vehicle (used in the missile test).
Alumni associations are not new, a certain, extremely litigious northern New Jersey Sales Force Automation Company, whose stock ticker symbol rhymes with DIRTY (apropo, really), has had such an Alumni for many years. It's a pity the employers pay more attention to these informal groups than the investors - why would a bunch of strangers put so much energy into shouting about the abuses at a company if there were nothing to shout about?
Our Alumni links, just two of several.
I think the Chunnel goes through extremely porous rock. The non-porous clay above the tunnel keeps it nominally dry. They might want porous rocks or rocks that are softer (not necessarily one and the same), to be able to dig the tunnel faster. Given that such a project would involve Russian and US Governments, maybe a tunneling rate of 1cm/decade is not out of the question :)
Anyway, how fast would the average dig rate for the chunnel be... it was first envisaged by Napolean in 1803 or somesuch... 22 miles / 180 years is about .1 mile a year. Statistics are wonderfully obscure.
Data record = 0x00
I just reviewed BT's web site looking for backbone numbers, didn't find much. Except... They have a new Transatlantic cable, takes 320K calls at once. Assume 14Kbits/s per call, that gives us about 806Gbytes per day, or your 12-14 Dat tapes, ofcourse it would take about 2 days to write these tapes...
Given that this is just one low speed source, and you would be better off putting a big collector on the main trunk lines; not all the individual sources (duh, who cares about web pages if no one looks at them) need recording. We are probably talking something like 5 or 6 backbone trunks doing about 1Gigbit/s, 24hrs a day.
Sure stream it, keep it, just don't think you will ever be able to do anything with it.
Its because of the bureaucratic nincompoops that make stipulations like these that I left that damp and soggy island, not to mention the fact that English politicians are too stupid to know when to say "cut the taxes."