Unfortunately that means you have to run a Windows server for your website, and Windows is dropping like a lead balloon in terms of web server statistics.
The current enterprise web stack is actually something along the lines of Java + Spring (Beans, DI, Security) + Apache Wicket (OO Ajax UI) + Hibernate (DB). The DAO layers being abstracted enough that they could be direct database access or remote method calls via your favoured protocol (RMI, SOAP, JSON, PB, etc).
Firstly, you might change ISP, so you lose your old email. Secondly, they pull tricks like this. Thirdly, they won't provide as good a solution as a dedicated email provider.
What I am wondering is if you can set up a new personal GMail account, and get it to sync your old emails from the ISP's own gmail service.
Indeed it is. People are getting their knickers in a twist over microcode for the GPU (which is probably just a microcode update) that probably target a custom architecture, compiled by a custom compiler, and which in themselves provide the advertised functionality (decoding h.264, etc). It's firmware, it's conjoined to the hardware (which is also closed-source) to make the hardware do something useful. By having firmware, you can add features down the line (e.g., VP8, H.265) as well, that you couldn't do with totally fixed hardware.
Note that many CPUs include lots of microcode themselves, where the advertised ISA instructions are actually small microcode programs. Of course they're less complex than video decode microcode, but they're the same thing overall. The CPU microcode is probably also generated via a custom assembler/compiler or hand-written.
So it could come down to the parcels being marked as suspicious, going to the suspicious parcels department (leading to a 3 day delay), where one guy now has a nice line in selling second hand shoes in the bar after work (leading to higher incidence of "lost" packages).
For a one or two story arched roof that is printed (sintered) in place as one solid construction? In 1/6th the Earth's gravity? I'm sure the lack of rebar will not be a problem.
Now when they're printing domes to cover entire craters, then they might need a means to strengthen the structure. And much as I would like that to happen within twenty years, I suspect it will be more like two hundred years before this happens.
I would imagine that the human trip to the moon would be *after* all the robotic colony printers have done their work in building the external structures.
If the structures that are built are robust (especially in regard to flaking away when rubbed/hit/etc) enough, then even furniture could be printed using this mechanism.
And these robots can keep on building stuff, as long as they are powered. As soon as the first module is built, it can build another module, and structures between modules. All we need to do is get the inflatable protective air-tight internal structures on the moon to be installed inside them. And lots of solar panels to mount on the outside. And some means of extracting water and air from the rocks (although initially these would be sent from Earth as well).
For each year of refinement to the equipment done here before launch, I'm sure many many years of work on the moon would be saved, as the equipment becomes more reliable, robust, useful, flexible, and faster.
I had a Palm IIIc (which I won, thanks ZDNet) which was really quite neat. And it was easy to use, and snappy. And the desktop software was pretty good.
Downsides - Serial connection was slow, low resolution display, lack of central repository for software - although some websites did step up to the mark.
AMD is positioning the quad-core Jaguars against low-end i3s actually. It's not "that slow".
I'm guessing that overall it will have 4x to 8x the processing power of the 3.2GHz, but in-order, PPE in the PS3's Cell. Not quite the generational leap it could have been, IMO.
Up until two or three days ago, the leaks were saying 4GB RAM, and a month or two ago, 2GB.
Clearly the XBox 3's (I'm not guessing what name they will use this time round) 8GB RAM has made Sony gamble that they need 8GB too. 8GB of fast costly GDDR5.
The video recording capability is done by the VCE unit that is in AMD's GPUs. It's free (it won't be used for anything else whilst playing a game), apart from memory bandwidth and packet processing/file I/O. I also presume that this unit is used in the Wii U for it's gamepad functionality, as that uses an AMD GPU as well. And I'm fairly certain that the XBox 3 will also include it, because it too will include an AMD GPU.
Your ship isn't going to come in building and fixing PCs. Just saying. You'll be lucky to get a dingy. That's what the reality is. You are conditioned to believe that hard work leads to loads of money, guaranteed, but if it did, tens of millions of people in the US would be rich. If you don't make your main money fixing PCs, then stop fixing PCs and get some rest, or hire someone to do the menial tasks.
Otherwise all you are doing is burning yourself out. The only plus side is that you are doing it for yourself.
Seriously, I do hope that your ship comes in because you work hard and do a good job (excepting the lack of sleep). It's just that the odds are so stacked against you...
Hmm, 1000 total rewrites. Let's assume that your fileserver (writing at a constant 2.4MB/s) is just writing, and capacity is somehow not an issue (e.g., overwriting logs). You have a 512GB SSD with a 1000 erase cycles. It takes two days to fill the SSD, hence it will take 2000 days before failure. That's around six years. I'm sure some other cheap component on the drive will have failed before then. Of course, with such low write speeds, but large amounts of data you would use the largest capacity SATA HDD so the whole exercise is moot.
Yes, the article should have talked about such drives - writing an article without even checking the current state of SSD technology was quite odd.
What you've written is great if you are worried about putting SSDs in your server farm. A decent RAID of 256GB SSDs should be safe for even the most writiest of server loads, for the expected lifetime of a server.
It would be great to see an article that including a spectrum of SSD write endurances for consumer use NAND (MLC and TLC), but also with office/consumer write expectations, e.g., 8 hours a day, average 10MB/s writes (i.e., perfect torrenting on a 100mbps internet link). Hell, do 100MB/s - that would cover most workstation usages.
You would still recharge more than the bare minimum. The "range anxiety" isn't worth it. Okay, maybe not 50% more than required, but 20% extra would be common sense, even if you are in a rush.
It shouldn't be a problem for most owners who don't do massively long journeys and for whom the overnight recharge is fine - no more refilling stations ever.
That would be the nightmare scenario for an EV currently. Fortunately it's not a common scenario! A decade of battery capacity improvements and second generation supercharger stations would solve it. Or we'll all be driving hydrogen pwoered cars.
I agree, except the thread updating the animation that things are working is probably totally disconnected from the thread that is doing stuff (and may be hung). So in reality, it's useless. Even the ones that are actually accurate are rendered useless because you don't know that it is a decent implementation! But yes, I'm sure they make people wait longer before failing to terminate the process because they feel it is doing something.
I think it's fairly reasonable to assume that if the person is 150 miles from their destination they're going to wait till the range estimator reads 155 then take off.
And you'd do that in a standard car?
No, no you wouldn't. You'd fill up at least 50% more than required.
And in both cars I bet the range estimator under-estimates by a fair margin in case someone does do the above.
alternately slow down and speed up to take advantage of regenerative braking
I do not believe advice like this would be given. Not one bit. I can believe "regenerative braking will extend the range some more", but not "you should speed up and brake to benefit from regenerative braking". Not one second - because the speeding up part takes far more power than is regained from the braking. If the advice was given (and so far there is no evidence it was, and I'm sure the call would have been recorded) then yes, the support guys need far more training.
Surely a car like the Tesla (in record everything mode) will also record GPS coordinates over time, and thus you can get accurate speed estimates from these, without relying on wheel shaft rotation rate to work out the speed - which may be inaccurate depending on the tire diameter (although I'm sure that the tire diameter setting is settable on the car fairly painlessly, and that the correct setting would have been used for this test drive).
However, he also admits the he "cannot account for the discrepancy" in that first 60 mph log. He thinks it might have to do with the tire.
FFS, I would seriously hope that accurate speedometers are a solved problem by now, and I'm sure these logs record GPS measured speeds in addition to wheelshaft rotation measured speeds.
I'm guessing the GPS logs are being saved for later. You think Musk showed his entire hand in the first salvo?
However I do believe the reporter was trying to find the supercharger station in this case - unless he truly was driving around a car park that is not anywhere near the charging station, and just hoping that 0 miles range meant 0 miles remaining, because it would make a juicy paragraph in his review.
Slashdot Mirror
Why would you ever let any social network trawl your email accounts (and other social networks)?
I'm annoyed that I can't "facet" my Facebook account into family, friends and work, and hide things from each of these. Google's thing can do that.
Last thing I want is exes popping up again.
Unfortunately that means you have to run a Windows server for your website, and Windows is dropping like a lead balloon in terms of web server statistics.
The current enterprise web stack is actually something along the lines of Java + Spring (Beans, DI, Security) + Apache Wicket (OO Ajax UI) + Hibernate (DB). The DAO layers being abstracted enough that they could be direct database access or remote method calls via your favoured protocol (RMI, SOAP, JSON, PB, etc).
Never, ever, use ISP provided email.
Firstly, you might change ISP, so you lose your old email.
Secondly, they pull tricks like this.
Thirdly, they won't provide as good a solution as a dedicated email provider.
What I am wondering is if you can set up a new personal GMail account, and get it to sync your old emails from the ISP's own gmail service.
Apache Wicket for the web UI (bye bye JSPs), and Spring (or Guice) for DI, beans, etc.
Indeed it is. People are getting their knickers in a twist over microcode for the GPU (which is probably just a microcode update) that probably target a custom architecture, compiled by a custom compiler, and which in themselves provide the advertised functionality (decoding h.264, etc). It's firmware, it's conjoined to the hardware (which is also closed-source) to make the hardware do something useful. By having firmware, you can add features down the line (e.g., VP8, H.265) as well, that you couldn't do with totally fixed hardware.
Note that many CPUs include lots of microcode themselves, where the advertised ISA instructions are actually small microcode programs. Of course they're less complex than video decode microcode, but they're the same thing overall. The CPU microcode is probably also generated via a custom assembler/compiler or hand-written.
So it could come down to the parcels being marked as suspicious, going to the suspicious parcels department (leading to a 3 day delay), where one guy now has a nice line in selling second hand shoes in the bar after work (leading to higher incidence of "lost" packages).
For a one or two story arched roof that is printed (sintered) in place as one solid construction? In 1/6th the Earth's gravity? I'm sure the lack of rebar will not be a problem.
Now when they're printing domes to cover entire craters, then they might need a means to strengthen the structure. And much as I would like that to happen within twenty years, I suspect it will be more like two hundred years before this happens.
I would imagine that the human trip to the moon would be *after* all the robotic colony printers have done their work in building the external structures.
If the structures that are built are robust (especially in regard to flaking away when rubbed/hit/etc) enough, then even furniture could be printed using this mechanism.
And these robots can keep on building stuff, as long as they are powered. As soon as the first module is built, it can build another module, and structures between modules. All we need to do is get the inflatable protective air-tight internal structures on the moon to be installed inside them. And lots of solar panels to mount on the outside. And some means of extracting water and air from the rocks (although initially these would be sent from Earth as well).
For each year of refinement to the equipment done here before launch, I'm sure many many years of work on the moon would be saved, as the equipment becomes more reliable, robust, useful, flexible, and faster.
I had a Palm IIIc (which I won, thanks ZDNet) which was really quite neat. And it was easy to use, and snappy. And the desktop software was pretty good.
Downsides - Serial connection was slow, low resolution display, lack of central repository for software - although some websites did step up to the mark.
Upsides - http://www.palminfocenter.com/news/560/palm-simcity-in-color/
It's eight cores, AMD processors don't bother with SMT / hyperthreading.
It's two Jaguar-based compute modules, if you feel like googling.
AMD is positioning the quad-core Jaguars against low-end i3s actually. It's not "that slow".
I'm guessing that overall it will have 4x to 8x the processing power of the 3.2GHz, but in-order, PPE in the PS3's Cell. Not quite the generational leap it could have been, IMO.
Up until two or three days ago, the leaks were saying 4GB RAM, and a month or two ago, 2GB.
Clearly the XBox 3's (I'm not guessing what name they will use this time round) 8GB RAM has made Sony gamble that they need 8GB too. 8GB of fast costly GDDR5.
The video recording capability is done by the VCE unit that is in AMD's GPUs. It's free (it won't be used for anything else whilst playing a game), apart from memory bandwidth and packet processing/file I/O. I also presume that this unit is used in the Wii U for it's gamepad functionality, as that uses an AMD GPU as well. And I'm fairly certain that the XBox 3 will also include it, because it too will include an AMD GPU.
It's 8 Jaguar cores. I don't know what this makes it. Lots of cores, but each core isn't exactly a powerhouse.
Your ship isn't going to come in building and fixing PCs. Just saying. You'll be lucky to get a dingy. That's what the reality is. You are conditioned to believe that hard work leads to loads of money, guaranteed, but if it did, tens of millions of people in the US would be rich. If you don't make your main money fixing PCs, then stop fixing PCs and get some rest, or hire someone to do the menial tasks.
Otherwise all you are doing is burning yourself out. The only plus side is that you are doing it for yourself.
Seriously, I do hope that your ship comes in because you work hard and do a good job (excepting the lack of sleep). It's just that the odds are so stacked against you...
I was responding to a post that wrote "I have a fileserver that writes 200gb per day".
Bonus points for using "writier" when comparing different use cases!
Hmm, 1000 total rewrites. Let's assume that your fileserver (writing at a constant 2.4MB/s) is just writing, and capacity is somehow not an issue (e.g., overwriting logs). You have a 512GB SSD with a 1000 erase cycles. It takes two days to fill the SSD, hence it will take 2000 days before failure. That's around six years. I'm sure some other cheap component on the drive will have failed before then. Of course, with such low write speeds, but large amounts of data you would use the largest capacity SATA HDD so the whole exercise is moot.
Yes, the article should have talked about such drives - writing an article without even checking the current state of SSD technology was quite odd.
What you've written is great if you are worried about putting SSDs in your server farm. A decent RAID of 256GB SSDs should be safe for even the most writiest of server loads, for the expected lifetime of a server.
It would be great to see an article that including a spectrum of SSD write endurances for consumer use NAND (MLC and TLC), but also with office/consumer write expectations, e.g., 8 hours a day, average 10MB/s writes (i.e., perfect torrenting on a 100mbps internet link). Hell, do 100MB/s - that would cover most workstation usages.
You would still recharge more than the bare minimum. The "range anxiety" isn't worth it. Okay, maybe not 50% more than required, but 20% extra would be common sense, even if you are in a rush.
It shouldn't be a problem for most owners who don't do massively long journeys and for whom the overnight recharge is fine - no more refilling stations ever.
That would be the nightmare scenario for an EV currently. Fortunately it's not a common scenario! A decade of battery capacity improvements and second generation supercharger stations would solve it. Or we'll all be driving hydrogen pwoered cars.
I agree, except the thread updating the animation that things are working is probably totally disconnected from the thread that is doing stuff (and may be hung). So in reality, it's useless. Even the ones that are actually accurate are rendered useless because you don't know that it is a decent implementation! But yes, I'm sure they make people wait longer before failing to terminate the process because they feel it is doing something.
I think it's fairly reasonable to assume that if the person is 150 miles from their destination they're going to wait till the range estimator reads 155 then take off.
And you'd do that in a standard car?
No, no you wouldn't. You'd fill up at least 50% more than required.
And in both cars I bet the range estimator under-estimates by a fair margin in case someone does do the above.
alternately slow down and speed up to take advantage of regenerative braking
I do not believe advice like this would be given. Not one bit. I can believe "regenerative braking will extend the range some more", but not "you should speed up and brake to benefit from regenerative braking". Not one second - because the speeding up part takes far more power than is regained from the braking. If the advice was given (and so far there is no evidence it was, and I'm sure the call would have been recorded) then yes, the support guys need far more training.
Surely a car like the Tesla (in record everything mode) will also record GPS coordinates over time, and thus you can get accurate speed estimates from these, without relying on wheel shaft rotation rate to work out the speed - which may be inaccurate depending on the tire diameter (although I'm sure that the tire diameter setting is settable on the car fairly painlessly, and that the correct setting would have been used for this test drive).
Let's just wait for Musk's reply.
However, he also admits the he "cannot account for the discrepancy" in that first 60 mph log. He thinks it might have to do with the tire.
FFS, I would seriously hope that accurate speedometers are a solved problem by now, and I'm sure these logs record GPS measured speeds in addition to wheelshaft rotation measured speeds.
I'm guessing the GPS logs are being saved for later. You think Musk showed his entire hand in the first salvo?
However I do believe the reporter was trying to find the supercharger station in this case - unless he truly was driving around a car park that is not anywhere near the charging station, and just hoping that 0 miles range meant 0 miles remaining, because it would make a juicy paragraph in his review.