You'd also need at least a few readers and a tape library style robotic storage / retrieval system which would add to the weight considerably. But if you didn't mind the huge random access latency, you might be able to build something fairly light.
Good point, I wonder if there is a link between browser usage and the state of the economy. With all the people being fired from work in the last 6-12 months, there's less people being forced to browse the net with IE from their work desktop and the remaining staff may be too busy and / or scared of being fired to browse too much...
More likely you trade slightly above and below the current price to trick GS into trades that manipulate the price further in ways that you can profit from.
Here's the main problem. Most of the people working in economics don't have a clue how the economy actually functions. They didn't predict this crisis, and they are still saying we are going to recover soon when the data really doesn't support that point of view. I've been reading Steeve Keen's blog for a couple of years. He did see this coming, and his predictions of just how bad this will turn out have been holding so far.
I would really like things to go back to how they were in the 30's to 50's in terms of legislation. Back then everyone remembered the Great Depression, and people had a much better understanding of the risks of widespread leverage. But since that generation died out we've been convincing ourselves we know better and have repealed all the laws that held the financial / banking system in check.
The 2 biggest issues I have with sql used in application development;
Slightly changing the requirements often involves completely restructuring the query. eg, A group by and having clause no longer cut it, now you need to use a work table and at least 3 separate statements.
Queries on the same tables are difficult to reuse for similar tasks, resulting in many copies of similar statements that must be maintained.
Don't get me wrong, I'm not judging myself. The point I was trying to make (remember, I'm a pretty poor communicator myself) is that I could have been in his shoes. If there's a genetic disposition, I've got it, maybe presenting as aspergers. But I didn't have the same environment as an infant / toddler, that's the biggest difference I see between myself and my son.
For me growing up, maths was like breathing, english like squeezing blood from a stone.
But I doubt austism has any single cause. I think a significant contributor to my sons diagnosis was how he dealt with frustration when he was unable to express what he really wanted.
My son, our second child, has an austism diagnosis. Personally I think his difficulty with language and resulting early behavioral issues stem from him being too smart, his mum and older sister being too talkative, and me being too lazy.
To clarify a little; as a toddler he didn't need any help with anything he could reach, he'd grunt and point for anything he wanted that he couldn't reach, and we'd play 20 questions to work out what he needed help with. But mostly he didn't seem to need or what our help so he avoided learning to communicate with us.
I see a lot of myself in my son. I can usually guess what he's thinking, because that's what I would be thinking. I'm very strong in maths / spacial problem solving, and very weak in creating writing and communication skills. But when I was growing up I had 2 teachers for parents who encouraged me all the time. My son doesn't have that advantage mainly because me and my wife were always busy with work or entertainment and being so self sufficient he was never very demanding of our time.
I wasn't surprised or angry by his diagnosis as I don't think it changed my understanding of who he is at all. I think his learning difficulties are more my fault for not knowing how to teach him, than the fault of any environmental agent or genetic disorder.
Sony are writing all the test cases one at a time, drip feeding the hackers with examples of how their VM implementation is flawed so they can fix it. It's like the ideal implementation of extreme programming;)
What I would really like to see in firefox (or an extension) for this type of use case is using tabs more like an automatic temporary bookmarking system.
AFAIK firefox already keeps pages in a number of different states between "rendered and ready to display" to "must re-download from host website". When you close a tab or navigate to a new page, the last page may stay at "rendered and ready to display" for quite a while before being cleaned up.
If I haven't even looked at a specific tab for an hour or so, there's no point keeping it ready to display. So go ahead and free up some memory, heck even flush the page state to the disk cache, just don't discard it until I close the tab.
Actually flushing the page to disk cache isn't a silly idea, in the event of a crash the browser could then reload the page without hitting the network...
Actually missing the last packet or two of a TCP stream is common if you shutdown the stream incorrectly. I ran into this problem when I tried to write an ftp client on windows. My guess is either their ftp client / server has a bug or the unreliable network he talks about is causing enough dropped packets to abort the connection early and their download client is not retrying.
So my recommendation would be to use some kind of download manager that can open multiple streams, that will retry on any connection failure, and ideally a protocol that does end to end checksums of the whole file to ensure it downloaded correctly.
I suppose you'd also prefer that their OS didn't ship with a file system, or at least preferred that the file system was very poorly coupled to the operating system and the user experience?
How about a shell interface that uses internal interfaces that aren't documented at all so you must effectively use the shell written by that company.
Back in the win95 days I tried writing an explorer task bar replacement. I got pretty far on my own. I built a start menu, a task bar, a system tray (with a little help from the doco in litestep). But I was completely stumped on implementing app-bar support (looks like litestep finally solved that one around mid 2007).
I'm certainly not going to complain about an OS that includes a html GUI control by default. But there are a lot of areas of their OS that can't be easily extended, replaced or customised because they haven't documented them well enough.
Also putting all your DRM eggs in one convenient basket makes a nice big juicy target for piracy. While this is a continuously moving target, the tools I've seen lately are getting fairly good.
Trust me, most developers writing new experimental filesystems have this feature on their radar. You might be able to build a hybrid device in hardware, but the filesystem layer is a much better place to put a feature like this.
Slashdot Mirror
You'd also need at least a few readers and a tape library style robotic storage / retrieval system which would add to the weight considerably. But if you didn't mind the huge random access latency, you might be able to build something fairly light.
Assuming of course that the battery doesn't have any net loss or gain of electrons...
Good point, I wonder if there is a link between browser usage and the state of the economy. With all the people being fired from work in the last 6-12 months, there's less people being forced to browse the net with IE from their work desktop and the remaining staff may be too busy and / or scared of being fired to browse too much...
There was a recent piece of "journalism" in australia that tried to make a tenuous link that "if you buy a DVD at a flea market you are funding terrorism", sigh.
More likely you trade slightly above and below the current price to trick GS into trades that manipulate the price further in ways that you can profit from.
Black Hat Hacker lesson of the day;
If you're going to steal source code you need to find a deniable way of copying the code to a machine you own before you send it outside the company.
Here's the main problem. Most of the people working in economics don't have a clue how the economy actually functions. They didn't predict this crisis, and they are still saying we are going to recover soon when the data really doesn't support that point of view. I've been reading Steeve Keen's blog for a couple of years. He did see this coming, and his predictions of just how bad this will turn out have been holding so far.
I would really like things to go back to how they were in the 30's to 50's in terms of legislation. Back then everyone remembered the Great Depression, and people had a much better understanding of the risks of widespread leverage. But since that generation died out we've been convincing ourselves we know better and have repealed all the laws that held the financial / banking system in check.
And then there's google gears which attempts the same disconnected operation with straight html & javascript.
The 2 biggest issues I have with sql used in application development;
Slightly changing the requirements often involves completely restructuring the query. eg, A group by and having clause no longer cut it, now you need to use a work table and at least 3 separate statements.
Queries on the same tables are difficult to reuse for similar tasks, resulting in many copies of similar statements that must be maintained.
Searching DNA is a very subtle and complex CS problem. This was written by a good friend of mine; DASH: Localising Dynamic Programming for Order of Magnitude Faster, Accurate Sequence Alignment
Don't get me wrong, I'm not judging myself. The point I was trying to make (remember, I'm a pretty poor communicator myself) is that I could have been in his shoes. If there's a genetic disposition, I've got it, maybe presenting as aspergers. But I didn't have the same environment as an infant / toddler, that's the biggest difference I see between myself and my son.
For me growing up, maths was like breathing, english like squeezing blood from a stone.
But I doubt austism has any single cause. I think a significant contributor to my sons diagnosis was how he dealt with frustration when he was unable to express what he really wanted.
My son, our second child, has an austism diagnosis. Personally I think his difficulty with language and resulting early behavioral issues stem from him being too smart, his mum and older sister being too talkative, and me being too lazy.
To clarify a little; as a toddler he didn't need any help with anything he could reach, he'd grunt and point for anything he wanted that he couldn't reach, and we'd play 20 questions to work out what he needed help with. But mostly he didn't seem to need or what our help so he avoided learning to communicate with us.
I see a lot of myself in my son. I can usually guess what he's thinking, because that's what I would be thinking. I'm very strong in maths / spacial problem solving, and very weak in creating writing and communication skills. But when I was growing up I had 2 teachers for parents who encouraged me all the time. My son doesn't have that advantage mainly because me and my wife were always busy with work or entertainment and being so self sufficient he was never very demanding of our time.
I wasn't surprised or angry by his diagnosis as I don't think it changed my understanding of who he is at all. I think his learning difficulties are more my fault for not knowing how to teach him, than the fault of any environmental agent or genetic disorder.
Sony are writing all the test cases one at a time, drip feeding the hackers with examples of how their VM implementation is flawed so they can fix it. It's like the ideal implementation of extreme programming ;)
What I would really like to see in firefox (or an extension) for this type of use case is using tabs more like an automatic temporary bookmarking system.
AFAIK firefox already keeps pages in a number of different states between "rendered and ready to display" to "must re-download from host website". When you close a tab or navigate to a new page, the last page may stay at "rendered and ready to display" for quite a while before being cleaned up.
If I haven't even looked at a specific tab for an hour or so, there's no point keeping it ready to display. So go ahead and free up some memory, heck even flush the page state to the disk cache, just don't discard it until I close the tab.
Actually flushing the page to disk cache isn't a silly idea, in the event of a crash the browser could then reload the page without hitting the network...
Actually missing the last packet or two of a TCP stream is common if you shutdown the stream incorrectly. I ran into this problem when I tried to write an ftp client on windows. My guess is either their ftp client / server has a bug or the unreliable network he talks about is causing enough dropped packets to abort the connection early and their download client is not retrying.
So my recommendation would be to use some kind of download manager that can open multiple streams, that will retry on any connection failure, and ideally a protocol that does end to end checksums of the whole file to ensure it downloaded correctly.
if google were to start making deals with all OEM's that their default search engine was google search
Start making deals? They already have.
I suppose you'd also prefer that their OS didn't ship with a file system, or at least preferred that the file system was very poorly coupled to the operating system and the user experience?
How about a shell interface that uses internal interfaces that aren't documented at all so you must effectively use the shell written by that company.
Back in the win95 days I tried writing an explorer task bar replacement. I got pretty far on my own. I built a start menu, a task bar, a system tray (with a little help from the doco in litestep). But I was completely stumped on implementing app-bar support (looks like litestep finally solved that one around mid 2007).
I'm certainly not going to complain about an OS that includes a html GUI control by default. But there are a lot of areas of their OS that can't be easily extended, replaced or customised because they haven't documented them well enough.
Just send a small transparant container of sand on the vomit comit or a space shuttle...
Also putting all your DRM eggs in one convenient basket makes a nice big juicy target for piracy. While this is a continuously moving target, the tools I've seen lately are getting fairly good.
"contact me via email at *"
I'm sure you could come up with other search terms...
Google obscure email addresses in groups, why can't they obscure them in search results?
How much of the recorded memory usage of Chrome is actually shared between processes?
Wouldn't a braking car apply more force to the plate? How much energy could it absorb from slowing a car down by 10km/h?
Trust me, most developers writing new experimental filesystems have this feature on their radar. You might be able to build a hybrid device in hardware, but the filesystem layer is a much better place to put a feature like this.
Unfortunately the speed of the interface to read and write to magnetic disks hasn't been increasing at the same rate as the volume of the disk.