why leave your basement, hack the security camera around him to take the picture, and the cellphone company to find out his phone number. send the picture to the phone number and call him to tell him the codeword easy, nothing to sweat about:)
what happens if the guy in front of you jams on the brakes or the car in the other lane swerves towards you
The computer which is always fully attentive of road condition through it's radar and cameras, detects the trajectory change or the rapid shortening of the distance with the other car or even the lighting of the brake lights on the car in front and reacts to it a thousand time faster than a human driver would ?
I don't think the problem really is the software. The problem more likely is cost : putting all the sensors on a car would most likely cost a lot driving car prices by a fair amount. this isn't a problem with planes because, a plane carries more people, has a longer life expectancy, has a better maintenance plan, etc etc, than an individual car has.
(Not to mention that the computer which does not bend or break rules will have left a safe security distance between the two cars)
Thanks for the great post, makes me wonder how they test their system, do you know if they use some kind of fuzzy testing ? like simulating completely random input combinations and applying them to the system to see how it reacts ?
just with your last sentence :
AND KNOWING IS HALF THE BATTLE! (Come to think of it, did they ever say what the other half was?)
They did actually, the other half is fighting the battle. You can know everything there is to know about doing something, if you don't actually engage in doing it, your knowledge is useless.
Is image recognition still a problem ? I thought it had been successfully solved by crowd sourcing it to spammers and bots writers.... there was an article about this a while ago but I can't find it right now.
I personnaly own a Logitech MX900 it does come with a usb pluggable pod/receiver but it is fully bluetooth compliant. I never pugged the pod's usb cable anywhere, just the power cord to recharge the mouse. And it has always worked flawlessly.
wouldn't that also mess with a human ? I mean, a tester on the other side would probably think you are a machine and start sending random stuff as well no ?
Have you tried looking down at the bottom of the page where, on every single google web search front page, whatever the language it is setup for, there is a link called "Google.com in English" ?
that is not a problem in itself : you are already used to wait while the system buffers the stream. If multicast allows a more efficient management of the bandwidth all you have to do is schedule sessions every 30 seconds or say 50 users and start the multicast.
I respectfully disagree.
Making the "commit" a _2_ step process (commit + push) is a risk because the second step is not _needed_ from the developper's point of view, unless once in a long while when he wants to send his modifications to the QA team. They already get the full power of VCS with only the first step (commit), thus it encourages a behaviour pattern where the second _additionnal_ step (if compared to a CVCS) is likely to be overlooked.
This is what the whole discussion is about : Yes the DVCS does support the CVCS pattern but it requires an additionnal step which may easily be overlooked because it doesn't bring a lot of value to the person it burdens.
Again you may be working in this wonderfull team spread all over the world at home and all in which case DVCS _is_ for you, but there are many many many teams who simply work in the same building if not on the same floor or even in the same room, who have a fast corporate network available at all times to the central repository. For those people DVCS introduces more overhead than worth.
This confuses me. Why can't you just have a corporate server with a repository that is the "official" repository (i.e., a centralized repository), but individuals can checkout a version, then work on it at the beach or wherever--including "committing" at the beach on their local laptop. However, until they push that change back to the central server, it is not considered completed work? This doesn't seem to be any different than having a true centralized svn repository, checking out code, working at the beach, then committing it later.
Even if you have a centralized svn repository, someone can checkout, make changes at the beach, then spill a margarita on the laptop and lose that work anyway.
On the surface it isn't any different, but think about it this way : there is _no_ need for the developper to push to the centralized location. the developper would still be able to enjoy all the advantages of the version control. The only thing left to make him push to the centralized repository is policy. And you should know that a policy won't be followed unless it is backed up by usefulness for the undertaker. That's just human nature. Therefore people will work locally and work will be lost. Sure the failure rate of computer is significantly lower now than it was, but this has the potential to make losses occur that would have been prevented using a centralized VCS.
As for the people saying it is hard to branch and merge in subversion, please try using it again. Especially now that it has basic merge tracking support. This goes a long way to cover the 80% of the cases that matter. I do agree that CVS is obselete.
Now that i have said this: I also enjoy using DVCS and there are even cases when they _are_ great for corporate work. not so long ago I was working on a project where part of the team was in india, part in tunisia and part in france at different locations. We seriously considered implementing a DVCS solution (svk based actually) but were prevented from doing it because of "corporate security" (yeah I know, don't tell me about it:( some)
this is not as ironic or impossible as it sounds at first sight, Sachs is not a dreamer, what he wants to achieve is not suppressing all of poverty, but to suppress life threatening poverty. To do this he proposes to help the poor countries get back on the development ladder by using slight modifications to the market forces. once they get on the development ladder he argues, extreme poverty should disappear pretty fast (his proposed time frame is 20 years )
Maybe the cpu problem could be solved using a distributed tracker system ala azureus ? offload some of the tracking to peers inside the secure network... just dreaming:)
noone is chump enough to make something totally different that nothing runs on. I guess that's why IBM did not develop the cell processor which is therefor not used in PS3s or why no supercomputer is built using it.
All this also explains why IBM did not develop a new product line of cell based blade servers. And neither are grids being built around cell based servers.
Of course even if IBM did develop it and sony did use it in the PS3, it would be unable to run anything which is why there isn't any game for the PS3 or why there are not linux distribution for the PS3.
Sorry, but a different architecture doesn't mean nothing runs on it, nor does it mean noone will develop for it if the promised power is cheap and proficient enough.
Who cares if it's real time ? it doesn't matter at all !! it will _seem_ real time to us since we have no other reference points. But our time may be running at 1 billionth the speed of the universe which is hosting the simulator and we would have no way to know. We may even be running on time sharing, on when the supercalculator is available, off when other more important projects need it. Our perception of time continuity would not change since our entire universe would simply be in stasis (us included) when the program is not running...
Am I the only one who is reminded of the new unified theory called "An Exceptionally Simple Theory of Everything" which was published recently ?
I was reading an article about it where they explain that the author of the theory believes that all interactions in our univers can be explained by particles and forces placed on the corners of an exceptionnal mathematical figure (E8 geometry). And how all their interactions can be explained by simple projections of this figure in a lower dimension.
When I read that, the first thing that came to my mind was that it is too perfect : an exceptionnal mathematical figure ? matching perfectly our universe ? what if we were just a simulation based on this figure to see if a workable universe would emerge out of it?
Now what I would find highly interesting is for a project to start simulating "universes" using E6,E7, E8, etc and see if we can get a universe working like ours bu slower:)
This was just about 20 seconds of googling it. I count two professional photographer within my friends, and when discussing with one of them who does portraits, she told me that when she takes the picture she is already thinking of the things she can fix on photoshop later... that about sums it up.
About point 3 : I have a hard time seeing how you could breed with photoshop to create these extremely beautiful people, or rather I prefer not to imagine it.
Slashdot Mirror
What happens to the game/software industry ?
Not much... It goes entirely software as a service (think xbox live/steam/...) + some console games
Or for software used in business they continue as before to get their licensing fee.
why leave your basement, hack the security camera around him to take the picture, and the cellphone company to find out his phone number. send the picture to the phone number and call him to tell him the codeword easy, nothing to sweat about :)
Thanks for the great post, makes me wonder how they test their system, do you know if they use some kind of fuzzy testing ? like simulating completely random input combinations and applying them to the system to see how it reacts ?
just with your last sentence :
They did actually, the other half is fighting the battle. You can know everything there is to know about doing something, if you don't actually engage in doing it, your knowledge is useless.
Is image recognition still a problem ? I thought it had been successfully solved by crowd sourcing it to spammers and bots writers .... there was an article about this a while ago but I can't find it right now.
I personnaly own a Logitech MX900 it does come with a usb pluggable pod/receiver but it is fully bluetooth compliant. I never pugged the pod's usb cable anywhere, just the power cord to recharge the mouse. And it has always worked flawlessly.
http://www.amazon.com/Logitech-MX900-Bluetooth-Cordless-Optical/dp/B0000CEPDF
Would it be possible to make the Antikythera http://en.wikipedia.org/wiki/Antikythera_mechanism in LBP ?
It seems the irony of using an amazingly powerful digital computer to emulate a simple mechanical computer is completely lost on you ...
wouldn't that also mess with a human ? I mean, a tester on the other side would probably think you are a machine and start sending random stuff as well no ?
Have you tried looking down at the bottom of the page where, on every single google web search front page, whatever the language it is setup for, there is a link called "Google.com in English" ?
Sorry but I prefer ::1 myself
that is not a problem in itself : you are already used to wait while the system buffers the stream. If multicast allows a more efficient management of the bandwidth all you have to do is schedule sessions every 30 seconds or say 50 users and start the multicast.
This should already help right ?
I respectfully disagree. Making the "commit" a _2_ step process (commit + push) is a risk because the second step is not _needed_ from the developper's point of view, unless once in a long while when he wants to send his modifications to the QA team. They already get the full power of VCS with only the first step (commit), thus it encourages a behaviour pattern where the second _additionnal_ step (if compared to a CVCS) is likely to be overlooked. This is what the whole discussion is about : Yes the DVCS does support the CVCS pattern but it requires an additionnal step which may easily be overlooked because it doesn't bring a lot of value to the person it burdens. Again you may be working in this wonderfull team spread all over the world at home and all in which case DVCS _is_ for you, but there are many many many teams who simply work in the same building if not on the same floor or even in the same room, who have a fast corporate network available at all times to the central repository. For those people DVCS introduces more overhead than worth.
This confuses me. Why can't you just have a corporate server with a repository that is the "official" repository (i.e., a centralized repository), but individuals can checkout a version, then work on it at the beach or wherever--including "committing" at the beach on their local laptop. However, until they push that change back to the central server, it is not considered completed work? This doesn't seem to be any different than having a true centralized svn repository, checking out code, working at the beach, then committing it later.
Even if you have a centralized svn repository, someone can checkout, make changes at the beach, then spill a margarita on the laptop and lose that work anyway.
On the surface it isn't any different, but think about it this way : there is _no_ need for the developper to push to the centralized location. the developper would still be able to enjoy all the advantages of the version control. The only thing left to make him push to the centralized repository is policy. And you should know that a policy won't be followed unless it is backed up by usefulness for the undertaker. That's just human nature. Therefore people will work locally and work will be lost. Sure the failure rate of computer is significantly lower now than it was, but this has the potential to make losses occur that would have been prevented using a centralized VCS.
As for the people saying it is hard to branch and merge in subversion, please try using it again. Especially now that it has basic merge tracking support. This goes a long way to cover the 80% of the cases that matter. I do agree that CVS is obselete.
Now that i have said this: I also enjoy using DVCS and there are even cases when they _are_ great for corporate work. not so long ago I was working on a project where part of the team was in india, part in tunisia and part in france at different locations. We seriously considered implementing a DVCS solution (svk based actually) but were prevented from doing it because of "corporate security" (yeah I know, don't tell me about it :( some)
Jeffrey sachs a famous american economist who was for a time special advisor to UN secretary general Kofi Annan wrote a book published in 2005 titled "The end of poverty" where he details just such a revision. see http://www.amazon.com/End-Poverty-Jeffrey-Sachs/dp/0141018666/ref=sr_11_1?ie=UTF8&qid=1201185744&sr=11-1
this is not as ironic or impossible as it sounds at first sight, Sachs is not a dreamer, what he wants to achieve is not suppressing all of poverty, but to suppress life threatening poverty. To do this he proposes to help the poor countries get back on the development ladder by using slight modifications to the market forces. once they get on the development ladder he argues, extreme poverty should disappear pretty fast (his proposed time frame is 20 years )
Maybe the cpu problem could be solved using a distributed tracker system ala azureus ? offload some of the tracking to peers inside the secure network ... just dreaming :)
I guess that's why IBM did not develop the cell processor which is therefor not used in PS3s or why no supercomputer is built using it.
All this also explains why IBM did not develop a new product line of cell based blade servers. And neither are grids being built around cell based servers.
Of course even if IBM did develop it and sony did use it in the PS3, it would be unable to run anything which is why there isn't any game for the PS3 or why there are not linux distribution for the PS3.
Sorry, but a different architecture doesn't mean nothing runs on it, nor does it mean noone will develop for it if the promised power is cheap and proficient enough.
Who cares if it's real time ? it doesn't matter at all !! it will _seem_ real time to us since we have no other reference points. ...
But our time may be running at 1 billionth the speed of the universe which is hosting the simulator and we would have no way to know. We may even be running on time sharing, on when the supercalculator is available, off when other more important projects need it. Our perception of time continuity would not change since our entire universe would simply be in stasis (us included) when the program is not running
Am I the only one who is reminded of the new unified theory called "An Exceptionally Simple Theory of Everything" which was published recently ?
:)
I was reading an article about it where they explain that the author of the theory believes that all interactions in our univers can be explained by particles and forces placed on the corners of an exceptionnal mathematical figure (E8 geometry). And how all their interactions can be explained by simple projections of this figure in a lower dimension.
When I read that, the first thing that came to my mind was that it is too perfect : an exceptionnal mathematical figure ? matching perfectly our universe ? what if we were just a simulation based on this figure to see if a workable universe would emerge out of it?
Now what I would find highly interesting is for a project to start simulating "universes" using E6,E7, E8, etc and see if we can get a universe working like ours bu slower
here is the original
http://www.pvponline.com/2007/01/24/turbo-boost/
Sorry this was not the original link : http://demo.fb.se/e/girlpower/retouch/retouch/index.html
... that about sums it up.
additionnal hint for this site, click on the picture to start playing.
Secondary bonus for bearing with me this far a post with many other links about the subject: http://www.diet-blog.com/archives/2006/03/26/celebrity_retouching_10_reasons_to_revise_your_reality.php
This was just about 20 seconds of googling it. I count two professional photographer within my friends, and when discussing with one of them who does portraits, she told me that when she takes the picture she is already thinking of the things she can fix on photoshop later
About point 3 : I have a hard time seeing how you could breed with photoshop to create these extremely beautiful people, or rather I prefer not to imagine it.
Here, see for yourself : http://www.i-am-bored.com/bored_link.cfm?link_id=14537
Some of them at least believe in peace and are trying to make it happen. Try this :
http://www.israelipalestinianproject.com/
Optimism is good for morale, cheer up!
Hmm the text-only version is recommended since the images are loaded from the original site which is undergoing a nuclear meltdown right now.
http://www.google.com/search?q=cache:http://www.environmentalgraffiti.com/?p=592&strip=1
To answer your last question : process explorer is available here : http://www.microsoft.com/technet/sysinternals/utilities/processexplorer.mspx which seems to be part of the microsoft website (if you trust URLs and DNS I mean).