So just use regular soap, sterilize things with alcohol or bleach
Bad idea. Some bacteria will happily become immune to bleach. Since they don't have to compete with other bacteria, you will have a lot of them in your bleach-based environment. Is that really what you want?
Just let the nasty things be outcompeted by things you can take. That means that you don't sterilize things when you don't need to, which for most people mean almost never. Use a moderate amount of regular soap when washing to remove any grease, and learn how to wash effectively. Most people just push the dirt around, probably to let it make new friends.
Right, and the airplane's transmitters are outside the cage. (They must be, or they wouldn't work, right?) Thus, the plane's transmitters should theoretically be shielded from the passengers' interference.
The instrument receivers are placed inside the plane, even if the antennas themselves are placed outside. The wiring for power and signal are also inside the plane. You can filter, but putting filters on everything adds weight, the same as better shielded cables would. Added airframe weight means a lower payload or shorter range. There is also a fundamental problem in filtering the bits of the spectrum you actually want to receive things in. You either have to go for a larger transmitter or a system-level gain (CDMA-like techniques) to get a workable signal. Neither of these are relevant in legacy systems.
I'll take that bet. I think the plane wouldn't survive long enough to crash.
Hell, you don't see planes crash every time the sun decides to belch out a few terajoules of flare in our direction. Not to mention the basic design of a fuselage bears some resemblance to an EM-blocking faraday cage.
Correct. But this transmitter is inside a faraday cage, or in the lap of someone sitting in a seat inside the plane, if you will.
I would be willing to bet a beer that if you placed the sun in the same position, the plane would crash.
People oversell their connectivity all the time, and if people tell you they don't, you are going to pay more. Erlang's formula, properly adapted and applied, will make sure there are as few expensive bits of equipment lying around idle as possible by estimating the traffic you are likely to get. The downside is that you get a statistical probability of your users peaking out during rush hours or if the formula is applied in error. The upside is a cheaper connection because your vendor isn't sitting on redundant equipment.
It all depends what you are after. I would go for the cheaper service if I was satisfied with them, especially since I have a knee-jerk reaction to salespeople who try to shuffle the last fifty years or so of traffic estimation techniques under the carpet to sell a more expensive solution.
Other people have told you to ask for the usage graphs of your present provider. That will tell you if his resources are maxed out for any reasonable amounts of time. If they are, take it up with him. If they aren't, give the salesdroid a kick on the shin from me.
I think the question shows a fair bit of ignorance on the part of the poster.
To make any reasonably fast computer hardware, you need access to design tools. There are no free production-level ASIC design tools available. This will set you back about $100k, at least.
Then you need access to a production line. If you can't tell the manufacturers that you want >1M parts, you will only get access to older technology. Expect power and speed problems, and you will need to sign an NDA.
Then you need to design and test the bits you want to make. Good luck. This will probably take enough time that the design has been surpassed a number of times by commercial vendors. The result is a netlist, tested against your chosen process.
The next step is layout of your gate-level netlist. This tool will set you back at least $100k. You will also need to learn to use it. This is an even more non-trivial task than synthesizing the netlist.
Then you need to actually make the parts. If you haven't got access to one of the prototype houses who make wafers with a number of chips on them, you will need to pick up the tab for the masks yourself. In addition, there is the silicon bits you need, and you will have to have them packaged.
Five or six years ago "IP" was the big buzzword in the ASIC business. You couldn't turn around at a conference without some vendor stuffing loads of IP buzzwords in your ear. Apart from a few notable exceptions (ARM being one of them up to a few years back), this business has flopped. It is just too hard and expensive to make a general bit that can be used across processes, and it will run too slow.
That standard already exists, and is called EAN 128. It is meant to be used for... well... just about everything.
However, current hardware can't cope with these, and label designers would have fits if they had to make room for a typical EAN 128 code. They are not small by any definition of the word if applied in the way they were designed to be applied, even if the codeset is compact.
However, if EAN 128 were to be widespread, the barcode could contain lots of other nifty data, like expiry dates, traceability info (who made the bit of meat you hold in your hand), the lot number of the delivery. The code is extensible, and would have no problem accomodating a new, n-numbered identification number.
However, 13 digits seem to be sufficient in Europe right now.
really. duh. hard coded means you can't change it. so why didn't the engineers notice they would have to throw out any fabricated chips if Microsoft told them we ain't accepting that particular hard code any longer?
This is not what I answered. I answered a question about why one would put things in hardware.
Pointing out errors like this is not hard after something has been broken. Doing it up front is way harder. I suggest you try doing a 10Mgate+ design sometime, and try maintaining control over the possible tradeoffs of speed (whatever) against a possible specification change in the future. This time around, Nvidia lost the gamble. Shit happens.
This is a complete non-issue. There are no critical systems connected to the Internet.
I have to bow to such expertise. I haven't got enough knowledge to make statements like this. I can't even make assumptions that cyber attacks have to be started using the 'net since I don't know anything about the systems involved.
However, power companies in several countries are using network-based information gathering to bill and control delivery to their large-scale customers. This information is sent over power lines which cross open terrain. I can see several scenarios where changing this information would cause trouble.
I don't spend any significant amount of time in front of the fridge. I other words - why should my fridge be smart?
On the other hand, I have any amount of recipe books which get dirtier each time they are used. I want a webpad with my recipes (fetched from a cache on my stationary box), able to control the volume of MP3 playback on my stereo, which can take spashes of water, being dropped on the floor, with a barcode reader for handling those pesky EAN-128 codes which will probably be used for quality info fairly soon (see the EAN barcode FAQ for more info) and which doesn't have a clunky keyboard, thank you very much.
Slashdot Mirror
Just let the nasty things be outcompeted by things you can take. That means that you don't sterilize things when you don't need to, which for most people mean almost never. Use a moderate amount of regular soap when washing to remove any grease, and learn how to wash effectively. Most people just push the dirt around, probably to let it make new friends.
I would be willing to bet a beer that if you placed the sun in the same position, the plane would crash.
It all depends what you are after. I would go for the cheaper service if I was satisfied with them, especially since I have a knee-jerk reaction to salespeople who try to shuffle the last fifty years or so of traffic estimation techniques under the carpet to sell a more expensive solution.
Other people have told you to ask for the usage graphs of your present provider. That will tell you if his resources are maxed out for any reasonable amounts of time. If they are, take it up with him. If they aren't, give the salesdroid a kick on the shin from me.
To make any reasonably fast computer hardware, you need access to design tools. There are no free production-level ASIC design tools available. This will set you back about $100k, at least.
Then you need access to a production line. If you can't tell the manufacturers that you want >1M parts, you will only get access to older technology. Expect power and speed problems, and you will need to sign an NDA.
Then you need to design and test the bits you want to make. Good luck. This will probably take enough time that the design has been surpassed a number of times by commercial vendors. The result is a netlist, tested against your chosen process.
The next step is layout of your gate-level netlist. This tool will set you back at least $100k. You will also need to learn to use it. This is an even more non-trivial task than synthesizing the netlist.
Then you need to actually make the parts. If you haven't got access to one of the prototype houses who make wafers with a number of chips on them, you will need to pick up the tab for the masks yourself. In addition, there is the silicon bits you need, and you will have to have them packaged.
Five or six years ago "IP" was the big buzzword in the ASIC business. You couldn't turn around at a conference without some vendor stuffing loads of IP buzzwords in your ear. Apart from a few notable exceptions (ARM being one of them up to a few years back), this business has flopped. It is just too hard and expensive to make a general bit that can be used across processes, and it will run too slow.
My two cents worth.
However, current hardware can't cope with these, and label designers would have fits if they had to make room for a typical EAN 128 code. They are not small by any definition of the word if applied in the way they were designed to be applied, even if the codeset is compact.
However, if EAN 128 were to be widespread, the barcode could contain lots of other nifty data, like expiry dates, traceability info (who made the bit of meat you hold in your hand), the lot number of the delivery. The code is extensible, and would have no problem accomodating a new, n-numbered identification number.
However, 13 digits seem to be sufficient in Europe right now.
Pointing out errors like this is not hard after something has been broken. Doing it up front is way harder. I suggest you try doing a 10Mgate+ design sometime, and try maintaining control over the possible tradeoffs of speed (whatever) against a possible specification change in the future. This time around, Nvidia lost the gamble. Shit happens.
Speed.
Ease of programming.
The reasons for putting things in microcode or hardware are not hard to see if you are an ASIC designer.
However, power companies in several countries are using network-based information gathering to bill and control delivery to their large-scale customers. This information is sent over power lines which cross open terrain. I can see several scenarios where changing this information would cause trouble.
On the other hand, I have any amount of recipe books which get dirtier each time they are used. I want a webpad with my recipes (fetched from a cache on my stationary box), able to control the volume of MP3 playback on my stereo, which can take spashes of water, being dropped on the floor, with a barcode reader for handling those pesky EAN-128 codes which will probably be used for quality info fairly soon (see the EAN barcode FAQ for more info) and which doesn't have a clunky keyboard, thank you very much.
Oh, and it should be dirt cheap.