For a long time, Windows allowed pretty long samba passwords. Except it didn't make a hash from the whole password supplied, but sequenced it into 8-char pieces which it then hashed and concatenated the hashes.
In most cases, a 9-char password is some 96 times (number of printable characters) harder than an 8-char password, and 10-char password is 96 times harder than 9-char password and so on. In their case, a 16-char password was twice as hard as 8-char password, and a 10-char password was a simple sum of difficulty of an 8-char password and a 2-char one.
Of course if we're talking only about competent implementations, then it's a different matter...
I thought the problem was that there was an infinite number of matching passphrases producing invalid results. Like, only a very simple hash or CRC - 1 or 2 bytes checks the validity of the passphrase to protect from common typos, but if you try even semi-hard, you will get a hash collision, the data decrypts, but it decrypts to garbage - a standard GIGO filter with a very weak anti-garbage protection on input.
This way, on top of one correct result you should get an infinite number of incorrect results and unless you have a clue how the correct result should look like and use some heuristics to distinguish it from garbage, you'll be no wiser than before... (and if it was additionally encrypted with anything that makes it look like white noise, there is simply no way to tell it apart from pure garbage.)
Sounds then, like they were faking work they were paid for. It's completely identical to stuffing a cold piece of meat in a hamburger you sell, or skipping washing the car before applying wax.
That's not a validation issue. That's a disciplinary issue. The employee is creating faulty products. If they are not aware of that, they must be made aware. If they are aware of that, they must be fired.
I wouldn't be surprised the least bit if it was the recommended power source with battery backup.
100h on battery power is a lot for any device. It would run either off mains or off a car-based generator, into which the whole battery of missiles would be plugged, providing the same synchronization.
You don't need more precision to get the missile into the radar cone (which was the problem here). You don't need much more precision if your missile doesn't hit the other missile, but creates a 50m wide cloud of shrapnel on its route
1/10 of a second has the same magic properties as 1/50 and 1/100 of a second, if obtained from 50Hz power grid: every device plugged into the same grid gets the same number of ticks. The frequency may float a little up or down, but remains consistent throughout the whole grid, meaning no costly, unreliable and difficult to implement synchronization subsystems.
The number provided area of the sky where to aim the radar, which then provided exact tracking for the missile. 229 meters would surely be a far miss with the missile but quite enough for the radar cone. 687 not quite so.
is when it's much better to use fixed-point arithmetic, If you're working with 0,1s ticks, make your clock an integer counting these ticks and use them universally throughout your software. Whenever you face the operation of division in your program, think twice whether it wouldn't be better to replace the basic unit by the one pre-divided and use integer multiplication elsewhere instead. No mess associated with floating point operations.
Re:Windows 7 Will beThe Death Knell For Microsoft
on
A Tale of Two Windows 7s
·
· Score: 3, Insightful
You missed the point. It's not the learning curve gp was complaining about. It's: - speed - stability - requirements - actual substantial improvements over XP.
In gp's experience both Vista and 7 failed on all 4 fronts. Slow, crashy, expensive and not better in any way.
...depends on how they are used - if they are a kind of extremely common pass-through elements that play a support role to others, then yes, it makes no sense. If they are used as end-of-the-line devices, like memory cells, arranged into memory banks, if one cell has 99.99% success rate, a bank of 1K has 90% success rate, and you'd have to produce only about 10% more of them than the device requires. Then passing the address bus through a remap might prove profitable. Unless of course the new feature provides less than 10% value improvement over the old, reliable one...
That's only if given action is classified as punitive rather than preventive. If you are drunk, you can't drive cars. Doesn't matter how inconvenient that is to you, without court, without judge, even without specific administrative order you lose right to drive until you are sober. (and notice, it's a part of freedom to travel, one of fundamental freedoms.) But that's not to punish you for drinking, it's to protect others from your dangerous driving. So if you are a notorious pirate, the government may take an action to protect authors from your irresponsible behavior - it's not that -you- are being punished, it's that -they- are being protected from you. (and as much as this may be total bullshit for you and me, lawyers and politicians, and even Joe Average may buy into it.)
It isn't entirely true. It just requires chips with partially programmable logic to switch features off. Lots of modern GPUs have around 50% yield per one vertex/pixel shader unit. Then they get sold as "LE", standard and "GT" versions, depending on how many shader units work - the silicon is the same, but the firmware disables failed units and the cheap version has 4 of them, the medium has 8 and the deluxe has 12 working units.
Simply use redundancy and disable failed parts of the chip.
Oh, but you're still allowed to use TV and radio. And books create fire risk and are environmentally unfriendly! Also, only lone people read books and we don't want that in a healthy society!
Access to a whole lot of information media is restricted currently. Restricted radio bands. Classified documents. Paywall to access archives. Currently analog TV is being dismantled. These are all information media.
Unfortunately, as long as you -can- access some information, your right isn't violated if some or many channels of access to it are forbidden. It doesn't matter that you'd have to travel 400 miles and pay $5000 to obtain the same information you can get here and now for free - you still have access to that information.
- that some law shouldn't be passed because it's a really bad idea doesn't mean it can't be passed. Of course the "licensed to use Internet" is a horrible idea and wouldn't pass because it would mean a political suicide to any party that did it. But it doesn't mean governments aren't legal to pass it.
- "whatever isn't forbidden, is allowed" combined with limited size of law and nearly unlimited scale of human imagination means vastly more things are allowed than forbidden and we can safely take for granted a lot of them despite lack of official law protection for them - simply because forbidding them would be a very bad idea. It still doesn't mean the government isn't legal to forbid them, just that it would be a very stupid thing to do so. Still, outbreaks of stupidity on the side of the government are pretty common.
tl;dr, you're confusing what should be with what can be.
The problem is access to the Internet is not any of elementary human rights or constitution-granted freedoms. The government may regulate, restrict and forbid access to it in any arbitrary way just like they may regulate sales of tobacco or speed limits on roads. They don't need a court sentence, they don't even need suspicion. They are allowed to pass a bill that says you need a special government-issued permit to access the Internet and any government clerk may revoke it on discretionary basis, and they aren't breaking any fundamental laws, because there weren't any laws granting you access to the Internet in the first place.
Not today - 5 days ago. I retried today, it failed just the same. Independent developers are killing themselves by preventing people willing to give them money from doing so.
I intended to pay $5 because that's how much I could afford and how much the game was worth for me. Unfortunately the site was down, timing out and giving me server errors. So I got World of Goo from Rapidshare, but it didn't provide options of payment to the authors. Pity.
Unfortunately if your (rich) competition is willing to spend on marketing way more than they ever expect to earn back, you'd better flee such market. One thing is competing by merit and minimal profit margins, another is operating at significant loss to deprive the competition of any chance to profit and compete.
I wonder how are they going to guarantee it to reindeer shepherds in the far north of Finland, living in the taiga good 100km away from nearest electric power...
The reason is that test makers will not accept calculators with very powerful abilities. They want the student to solve the problem and not the calculator. And here the students did the exact opposite: solved the calculator.
wait - the other party in the communication is not a person but a computer. In most cases the pinged person would never know they have been contacted, unless they installed dedicated software that would inform them of the fact. It's more like calling their modem with a voice phone, they would never know you did unless they attach a phone to the line which they don't normally use for phone communications.
Scale difficulty, but scale rewards too. Always allow to scale back and never let the game overwhelm the player.
This is precisely what Oblivion did wrong. It decided about your difficulty basing on your character level, never caring if your character mastered in Speechcraft, Mercantile, lockpicking and Acrobatics. It still threw strongest ogres and meanest trolls at you. OTOH if you -avoided- levelling up, you could beat the enemies better. Also, by making the whole world levelled flat with your progress, it removed incentive for exploration and made it hard to scale difficulty yourself by picking your battles.
So instead of arbitrarily deciding "this player is good, let's give him a hard path", present the player with three paths of various difficulty, AND various rewards. Say, you can only get the "best" ending if you finish on "hard", because only then you will have to snipe the main boss precisely instead of blowing up the whole place, and the collateral damage will be reduced - and best if the difficulty is chosen by the player by gameplay decisions. Picking the right opponents, choosing the right weaponry etc....also, never deprive the player of the pleasure of squashing the strongest enemy like a bug, if they earned it by hard work. My fav moment of STALKER-SOTC? The final assault on the Reactor, armed with the Bulldog grenade launcher and a stash of grenades saved over the whole gameplay. They would make common battles way easier, but I saved them and then the over-the-top weapon made the final difficult battle a breeze, elite enemies thrown left and right.
It was a very surprising experience, moving from small services where you get 10 hits per minute maybe, to a corporation that receives several thousands hits per second.
There was a layer of cache between each of 4 application layers (database, back-end, front-end and adserver), and whenever a generic cache wouldn't cut it, a custom one was applied. On my last project there, the dedicated caching system could reduce some 5000 hits per second to 1 database query per 5 seconds - way overengineered even for our needs but it was a pleasure watching the backend compressing several thousands requests into one, and the frontend split into pieces of "very strong cache, keep in browser cache for weeks", "strong caching, refresh once/15 min site-wide", "weak caching, refresh site-wide every 30s" and "no caching, per visitor data" with the first being some 15K of Javascript, the second about 5K of generic content data, the third about 100 bytes of immediate reports and the last some 10 bytes of user prefs and choices.
It isn't always true.
For a long time, Windows allowed pretty long samba passwords. Except it didn't make a hash from the whole password supplied, but sequenced it into 8-char pieces which it then hashed and concatenated the hashes.
In most cases, a 9-char password is some 96 times (number of printable characters) harder than an 8-char password, and 10-char password is 96 times harder than 9-char password and so on. In their case, a 16-char password was twice as hard as 8-char password, and a 10-char password was a simple sum of difficulty of an 8-char password and a 2-char one.
Of course if we're talking only about competent implementations, then it's a different matter...
I thought the problem was that there was an infinite number of matching passphrases producing invalid results. Like, only a very simple hash or CRC - 1 or 2 bytes checks the validity of the passphrase to protect from common typos, but if you try even semi-hard, you will get a hash collision, the data decrypts, but it decrypts to garbage - a standard GIGO filter with a very weak anti-garbage protection on input.
This way, on top of one correct result you should get an infinite number of incorrect results and unless you have a clue how the correct result should look like and use some heuristics to distinguish it from garbage, you'll be no wiser than before... (and if it was additionally encrypted with anything that makes it look like white noise, there is simply no way to tell it apart from pure garbage.)
Sounds then, like they were faking work they were paid for.
It's completely identical to stuffing a cold piece of meat in a hamburger you sell, or skipping washing the car before applying wax.
That's not a validation issue. That's a disciplinary issue. The employee is creating faulty products. If they are not aware of that, they must be made aware. If they are aware of that, they must be fired.
I wouldn't be surprised the least bit if it was the recommended power source with battery backup.
100h on battery power is a lot for any device. It would run either off mains or off a car-based generator, into which the whole battery of missiles would be plugged, providing the same synchronization.
In Cracow we have 5 bridges (in reasonable distance) and if even one is blocked the whole city gets jammed beyond any hope.
You don't need more precision to get the missile into the radar cone (which was the problem here).
You don't need much more precision if your missile doesn't hit the other missile, but creates a 50m wide cloud of shrapnel on its route
1/10 of a second has the same magic properties as 1/50 and 1/100 of a second, if obtained from 50Hz power grid: every device plugged into the same grid gets the same number of ticks. The frequency may float a little up or down, but remains consistent throughout the whole grid, meaning no costly, unreliable and difficult to implement synchronization subsystems.
The number provided area of the sky where to aim the radar, which then provided exact tracking for the missile. 229 meters would surely be a far miss with the missile but quite enough for the radar cone. 687 not quite so.
is when it's much better to use fixed-point arithmetic,
If you're working with 0,1s ticks, make your clock an integer counting these ticks and use them universally throughout your software.
Whenever you face the operation of division in your program, think twice whether it wouldn't be better to replace the basic unit by the one pre-divided and use integer multiplication elsewhere instead. No mess associated with floating point operations.
You missed the point. It's not the learning curve gp was complaining about.
It's:
- speed
- stability
- requirements
- actual substantial improvements over XP.
In gp's experience both Vista and 7 failed on all 4 fronts. Slow, crashy, expensive and not better in any way.
...depends on how they are used - if they are a kind of extremely common pass-through elements that play a support role to others, then yes, it makes no sense. If they are used as end-of-the-line devices, like memory cells, arranged into memory banks, if one cell has 99.99% success rate, a bank of 1K has 90% success rate, and you'd have to produce only about 10% more of them than the device requires. Then passing the address bus through a remap might prove profitable. Unless of course the new feature provides less than 10% value improvement over the old, reliable one...
That's only if given action is classified as punitive rather than preventive.
If you are drunk, you can't drive cars. Doesn't matter how inconvenient that is to you, without court, without judge, even without specific administrative order you lose right to drive until you are sober. (and notice, it's a part of freedom to travel, one of fundamental freedoms.) But that's not to punish you for drinking, it's to protect others from your dangerous driving.
So if you are a notorious pirate, the government may take an action to protect authors from your irresponsible behavior - it's not that -you- are being punished, it's that -they- are being protected from you. (and as much as this may be total bullshit for you and me, lawyers and politicians, and even Joe Average may buy into it.)
It isn't entirely true. It just requires chips with partially programmable logic to switch features off.
Lots of modern GPUs have around 50% yield per one vertex/pixel shader unit. Then they get sold as "LE", standard and "GT" versions, depending on how many shader units work - the silicon is the same, but the firmware disables failed units and the cheap version has 4 of them, the medium has 8 and the deluxe has 12 working units.
Simply use redundancy and disable failed parts of the chip.
Oh, but you're still allowed to use TV and radio. And books create fire risk and are environmentally unfriendly! Also, only lone people read books and we don't want that in a healthy society!
Access to a whole lot of information media is restricted currently. Restricted radio bands. Classified documents. Paywall to access archives. Currently analog TV is being dismantled. These are all information media.
Unfortunately, as long as you -can- access some information, your right isn't violated if some or many channels of access to it are forbidden. It doesn't matter that you'd have to travel 400 miles and pay $5000 to obtain the same information you can get here and now for free - you still have access to that information.
You're missing some details:
- that some law shouldn't be passed because it's a really bad idea doesn't mean it can't be passed. Of course the "licensed to use Internet" is a horrible idea and wouldn't pass because it would mean a political suicide to any party that did it. But it doesn't mean governments aren't legal to pass it.
- "whatever isn't forbidden, is allowed" combined with limited size of law and nearly unlimited scale of human imagination means vastly more things are allowed than forbidden and we can safely take for granted a lot of them despite lack of official law protection for them - simply because forbidding them would be a very bad idea. It still doesn't mean the government isn't legal to forbid them, just that it would be a very stupid thing to do so. Still, outbreaks of stupidity on the side of the government are pretty common.
tl;dr, you're confusing what should be with what can be.
The problem is access to the Internet is not any of elementary human rights or constitution-granted freedoms.
The government may regulate, restrict and forbid access to it in any arbitrary way just like they may regulate sales of tobacco or speed limits on roads. They don't need a court sentence, they don't even need suspicion. They are allowed to pass a bill that says you need a special government-issued permit to access the Internet and any government clerk may revoke it on discretionary basis, and they aren't breaking any fundamental laws, because there weren't any laws granting you access to the Internet in the first place.
Not today - 5 days ago.
I retried today, it failed just the same.
Independent developers are killing themselves by preventing people willing to give them money from doing so.
I intended to pay $5 because that's how much I could afford and how much the game was worth for me. Unfortunately the site was down, timing out and giving me server errors. So I got World of Goo from Rapidshare, but it didn't provide options of payment to the authors. Pity.
Unfortunately if your (rich) competition is willing to spend on marketing way more than they ever expect to earn back, you'd better flee such market. One thing is competing by merit and minimal profit margins, another is operating at significant loss to deprive the competition of any chance to profit and compete.
I wonder how are they going to guarantee it to reindeer shepherds in the far north of Finland, living in the taiga good 100km away from nearest electric power...
The reason is that test makers will not accept calculators with very powerful abilities. They want the student to solve the problem and not the calculator.
And here the students did the exact opposite: solved the calculator.
wait - the other party in the communication is not a person but a computer. In most cases the pinged person would never know they have been contacted, unless they installed dedicated software that would inform them of the fact. It's more like calling their modem with a voice phone, they would never know you did unless they attach a phone to the line which they don't normally use for phone communications.
Scale difficulty, but scale rewards too. Always allow to scale back and never let the game overwhelm the player.
This is precisely what Oblivion did wrong. It decided about your difficulty basing on your character level, never caring if your character mastered in Speechcraft, Mercantile, lockpicking and Acrobatics. It still threw strongest ogres and meanest trolls at you. OTOH if you -avoided- levelling up, you could beat the enemies better. Also, by making the whole world levelled flat with your progress, it removed incentive for exploration and made it hard to scale difficulty yourself by picking your battles.
So instead of arbitrarily deciding "this player is good, let's give him a hard path", present the player with three paths of various difficulty, AND various rewards. Say, you can only get the "best" ending if you finish on "hard", because only then you will have to snipe the main boss precisely instead of blowing up the whole place, and the collateral damage will be reduced - and best if the difficulty is chosen by the player by gameplay decisions. Picking the right opponents, choosing the right weaponry etc. ...also, never deprive the player of the pleasure of squashing the strongest enemy like a bug, if they earned it by hard work. My fav moment of STALKER-SOTC? The final assault on the Reactor, armed with the Bulldog grenade launcher and a stash of grenades saved over the whole gameplay. They would make common battles way easier, but I saved them and then the over-the-top weapon made the final difficult battle a breeze, elite enemies thrown left and right.
It was a very surprising experience, moving from small services where you get 10 hits per minute maybe, to a corporation that receives several thousands hits per second.
There was a layer of cache between each of 4 application layers (database, back-end, front-end and adserver), and whenever a generic cache wouldn't cut it, a custom one was applied. On my last project there, the dedicated caching system could reduce some 5000 hits per second to 1 database query per 5 seconds - way overengineered even for our needs but it was a pleasure watching the backend compressing several thousands requests into one, and the frontend split into pieces of "very strong cache, keep in browser cache for weeks", "strong caching, refresh once/15 min site-wide", "weak caching, refresh site-wide every 30s" and "no caching, per visitor data" with the first being some 15K of Javascript, the second about 5K of generic content data, the third about 100 bytes of immediate reports and the last some 10 bytes of user prefs and choices.
Would pinging the person's IP number constitute "communication"?