Re:Most import prediction: construction standards
on
Anticipating Earthquakes
·
· Score: 2, Insightful
In the 1995 Kobe earthquake, many buildings collapsed at the 4th floor, because they followed the building regulations extant at that time: there was a step reduction in column strength at that point. The breaks were clean enough to be notable. For example, Kobe City Hall collapsed at the 4th floor, and the break was so clean, they cleaned away the debris and re-joined the building. So it looks the same as before, just 2 floors shorter. Building regulations don't always get it right.
Some people might look to the haste with which toppled sections of the Hanshin Expressway and damaged pillars on the Shinkansen were removed, to decide whether the regulations were followed in the first place.
It would be interesting to know whether tuned mass dampers and active mass dampers are performing to spec in the minor earthquakes leading up to the next "big one".
Since we're interested in deep, safe places whilst the ground trembles, I'll do Down Street (using the "Did The Earth Move For You, Too?"), adjunct to the original 1907-1930 rules.
People don't need to leave the city. Instead, they move to evacutation zones where nothing's likely to fall on them. People die in earthquakes because (a) things fall on them and they die, or (b) things fall on them, they survive and then burn in the fires. In the Kobe earthquake a few years ago, relatives had time to bid farewell to trapped relatives as the fires approached. Get people a short way away from buildings, and they survive.
The Japanese system envisages, ultimately, being able to give a few minutes' warning, and that should be enough to save the majority of deaths. (Of course, the warning consists of gathering a bunch of learned people who learnedly pool their various thoughts on the matter, and get it wrong, in a learned manner, but the concept's reasonable.)
Having been in a couple of interesting earthquakes in Japan and the US, if someone gives me a 3-minute warning, I'm happy to go stand in the local park for 20 minutes.
And not all cultures loot after an earthquake, but here I'll obviously agree that NASA, spending US taxpayer's money, should clearly focus on the US population.
You said "Now, we actually task one aircraft to destroy multiple targets". That's only sort of true, if you're talking about a very select group of aircraft in US inventory. For the average sortie, you send up a package of aircraft that ensure the sortie's success. You need to suppress enemy air defence (SEAD), block their radars, dominate the airspace and refuel the aircraft. You're looking at 20 or 30 aircraft in the air for each one that actually drops a munition.
The US basically ceded the low altitude route for evading enemy radar to the Brits, specifically Jaguar and Tornado, partly because crew losses during training are perceived as too high for peacetime. A TFR/TFN (terrain following radar/terrain following navigation) has a slightly higher chance of hitting a target than the current US battle package, but the crew risk is several times higher.
The US allies capable of low-altitude, high-speed autonomous sorties right now are the Brits (Jaguar & Tornado), the French (Jaguar) and Italians (Tornado). The US has lost that particular skill.
So, in the current battle order, it's not necessarily the US battle plan that would benefit from your item 3, the ability to carry more bombs on a rack. You might be able to hit 2 or 3 targets instead of one, rather than tens of targets. It's an integration thing.
Of course, the outstanding B-52 and Canberra can still do anything required of them, which is kinda neat, given their age.
I did this with a client who's Fujitsu drive died an ugly death: there was a soot mark next to an IC on the dead drive. Since he'd bought several computers at the same time, I cloned one of the other drives using PartitionMagic, then swapped the PCB on the now-spare drive. No problem. That's got to be considered a trivial repair.
I've also had good luck pulling data off 2.5" drives by pulling the covers and simply running them through a hardware cloning box (about $120 now). The fact that you're reducing their MTBF to something like 10 hours is irrelevant if you get the job done in 20 minutes.
Oh, act lawyerish: only charge for successful recoveries. That way, the clients even sympathise with you if you don't succeed.
WRT traffic compression, this is seen all the time in Japan, where bicycles are encouraged to use the sidewalk. Two or three bikes can totally gum up the crossing at a stop signal, simply because their acceleration curves are different to those of the foot-bound humans milling around them and trickling through the gaps, self-sorting into a kind of starting order when the signal changes.
In this context, I think the problem is not that the Segway is good or bad, but that it acts differently. I could imagine it being much more successful in a country like the Netherlands, where pedestrians, cyclists and autos are all segregated.
Theakston's were taken over by Scottish & Newcastle a few years back and the bulk of production was shifted to Newcastle. Paul Theakston set up Black Sheep Brewery in Masham, so you might want to check that out. 3.8% ABV on draft, 4.4% ABV in bottle.
I think there's probably a Boddington's god somewhere who will track you down and do you a little mischief.
My first ever beer was a Boddington's in the Old Vic, St. Annes. I was young enough not to realise that a bar with sawdust on the floor, a pile of assorted inconvenienced people laying in a somewhat disshevilled heap around the entrance and pool tables where there were more half cues than cues, was a bad sign. Ah, the nectar of Boddington's. The Cream of Manchester.
I have a hardware disk copy unit: plug the old drive in one side, the new drive in the other, push the button, and the drive is cloned to the new one in 5-20 minutes. I upgraded 5 disks in an office from 10GB to 180GB in a relaxed 90 minutes. It will also clone from an internal drive over firewire or USB2.0, so I simply back up a disk whenever I work on a PC now. It also seems to copy drives that Partition Magic won't touch. Details in Japanese here but there's got to be a US equivalent somewhere. $120.
I was using such a device in the early 1980's. Two small boxes, one of which beeped noisily if it was separated from its partner by more than a meter or so. Then VDUs started arriving, and my fickle alarm decided it would remain contentedly silent if there was a monitor anywhere in the building.
My knowledge of Chinese is minimal, to say the least, but doesn't their document "(AVS-CFP-Video Coding) (1.0)" show the contact person to be one Feng Wu at the Microsoft Asia Research Institute's Network Multimedia Group? Author is Wang Ping of Microsoft.
I have 1GB SD with an SD-to-CF adapter and a CF-to-PC card adapter plus a PC card-to-USB adapter. It gets most things done. The adapter set cost about $30.The SD card spends its life rotating between my MP3 player, camera, cellphone & notebook.
Comparing SD & MMC, you might want to consider that an SD card has a write protect switch which MMC & CF do not (but I expect I'll be corrected on that), and is much faster than MMC (up to 10Mbps vs 300kbs or 1Mbps). Ignore MMC.
If you go with CF, a useful real-world speed comparison is at Rob Galbraith's site.
Kakaku. Japanese prices for a Japanese product in Japan. Dynamism is offering a bespoke customization service, and charges a premium for that (quite fairly).
Yes, many "traditional" auction houses charge both a buyer's and a seller's premium, plus quite large storage and shipping charges if you cannot immediately walk out with the item (which you can't, if you bid via phone or the internet. The buyer pays quite a lot more than expected, and the seller gets a lot less.
On many of the Japanese-originated PDAs, you'll notice CF and SD slots. In the US reviewers tend to mistakenly assume this is simply giving users a choice. In Japan, as I'm assuming you know given your location, typically the memory card goes in the SD slot and the feature card goes in the CF slot: wireless, Docomo P-In, Air H" or whatever.
I agree that PDA penetration in the market is less than in the US; you suggest this is because everyone has a laptop instead. Actually, for the casual user, I think it's because everyone has a cellphone instead: most newcomers to the internet in Japan enter via cellphone.
But I see a lot of PDAs in business and tech: visiting a large electronics company last year, everyone at the meeting had company-issued PDA swhich were kept updated over an internal PHS network.
The SL-C750 and C760 are selling well in Japan, especially the C760. 760 has 128MB flash RAM vs 64MB, 1700mAH battery vs 900mAH. Both support VPN over services such as YahooBB Mobile (Yahoo's hotspot adjunct to home ADSL service). With support for wireless and cellular data cards (up to 128k for cellular), it's probably much more of a connected mobile device in its home market that in the US.
Street price of theC750 is about $375, and $500 for the C760.
Depending on the design of the battery pack, it sounds like one of the internal cells may have failed and the charging chip in the pack has switched it out. So your initial potential isn't as high, and you drop to the power management's cutoff point that much faster.
In Windows, you may be able to find a battery icon somewhere, and have it tell you the pack's serial number, the initial power capacity of the cell (in mAH) and the current capacity. I actually monitor this across my notebooks.
I have two packs for my main machine: the original short-life pack which I have in the machine when it's home and always plugged in, and an extended-life pack which I only use when I'm out and about. That way, I know exactly how many times it's been charged: 229. Its capacity is down to 80% whilst my original pack is down to 40%.
... but shelf space in stores isn't. There's plenty of good boxed software that never gets to see the light of day because the company is outbid for shelfspace at PCWorld, etc. Online distribution has more benefits than simply reduced (hopefully) price.
This would be the perfect time and place to test new/unproven technologies...
In markets like Iraq, India and (especially) China, the "new" technologies are easier to roll out because there isn't a strong legacy technology to displace. Consider cellphones: in Iraq, cellphone networks seem to be automagically re-emerging because network damage is effectively point failure, since there is relatively little wired backbone to maintain. Whereas restoring a badly-damaged POTS network can take serious time and expense. In China, where there is little legacy technology, cellular networks are cost effective because they are not replacing a POTS network: if cellular isn't built, something else has to be. In Iraq, there is probably a substantial military data network infrastructure that can easily be converted to a public backbone. In other words, VOIP & cellular may be the only sensible options in emerging/recovering economies, and POTS is the expensive option.
The article mentions that Hindi's a problem because...Hindi is written in a non-Latin script, which has numerous different digital encodings instead of one or two standard ones.... Yet cited successful translation pairs include Chinese (with Big5 and GB encodings), with a much larger character
set than Hindi, and Arabic (with at least ISO 8859-6 and CP-1256), which has a smaller character set. Does this mean the text is being romanized before use? If so, this itself can be a major task. For example, Japanese has various encodings of its character sets including utf-8, shift-JIS, iso-2022-jp, euc-jp and of course unicode, with romanization systems including Hepburn and Kunreeshiki.
I'd hazard a guess that this system will also have trouble with a high-context language such as Japanese.
If all the effort is expended at the point of accumulating the parallel texts, then that's simply lots of computer time but if the text has to be massaged by the user to suit the system at translation time, then that could still be a lot of work.
The approach sounds rather like Translation Memory (as used in Trados and other systems) on a grand scale: "here's a sentence I translated earlier", as Blue Peter would say.
Can't knock it if it works, of course, which it appears to do.
Babylon 5 was created by Straczynski. In homage to Blake's 7, the Drazi warship in season 3 was based on the Liberator. Ron Thornton worked on both series: modelling in B7 and CGI in B5.
The Excalibur ship in the Crusade successor to B5 also looks rather Liberator-like.
The team designing the original Liberator and its variants didn't have experience of designing props for wire work, so the Liberator was difficult to hang and manipulate. It was slipups like this that made some effects seem cheaper than they were.
Since one is probably using fast film at a sports event, grain becomes a problem as the image is enlarged. There's no grain on a digital image, and enhancement tools with pixel interpolation can keep the jaggies at bay. In the context of a sporting or other event where I'm not in control of the pace of the action, I'd rather work with digital (even at 4MP) than film.
Slashdot Mirror
Some people might look to the haste with which toppled sections of the Hanshin Expressway and damaged pillars on the Shinkansen were removed, to decide whether the regulations were followed in the first place.
It would be interesting to know whether tuned mass dampers and active mass dampers are performing to spec in the minor earthquakes leading up to the next "big one".
Since we're interested in deep, safe places whilst the ground trembles, I'll do Down Street (using the "Did The Earth Move For You, Too?"), adjunct to the original 1907-1930 rules.
Japanese cows have to give nose prints. I don't know what happens if they refuse.
The Japanese system envisages, ultimately, being able to give a few minutes' warning, and that should be enough to save the majority of deaths. (Of course, the warning consists of gathering a bunch of learned people who learnedly pool their various thoughts on the matter, and get it wrong, in a learned manner, but the concept's reasonable.)
Having been in a couple of interesting earthquakes in Japan and the US, if someone gives me a 3-minute warning, I'm happy to go stand in the local park for 20 minutes.
And not all cultures loot after an earthquake, but here I'll obviously agree that NASA, spending US taxpayer's money, should clearly focus on the US population.
The US basically ceded the low altitude route for evading enemy radar to the Brits, specifically Jaguar and Tornado, partly because crew losses during training are perceived as too high for peacetime. A TFR/TFN (terrain following radar/terrain following navigation) has a slightly higher chance of hitting a target than the current US battle package, but the crew risk is several times higher.
The US allies capable of low-altitude, high-speed autonomous sorties right now are the Brits (Jaguar & Tornado), the French (Jaguar) and Italians (Tornado). The US has lost that particular skill.
So, in the current battle order, it's not necessarily the US battle plan that would benefit from your item 3, the ability to carry more bombs on a rack. You might be able to hit 2 or 3 targets instead of one, rather than tens of targets. It's an integration thing.
Of course, the outstanding B-52 and Canberra can still do anything required of them, which is kinda neat, given their age.
I've also had good luck pulling data off 2.5" drives by pulling the covers and simply running them through a hardware cloning box (about $120 now). The fact that you're reducing their MTBF to something like 10 hours is irrelevant if you get the job done in 20 minutes.
Oh, act lawyerish: only charge for successful recoveries. That way, the clients even sympathise with you if you don't succeed.
Well, all the British bits will be made in Taiwan, and all the US bits will be made in China.
In this context, I think the problem is not that the Segway is good or bad, but that it acts differently. I could imagine it being much more successful in a country like the Netherlands, where pedestrians, cyclists and autos are all segregated.
Theakston's were taken over by Scottish & Newcastle a few years back and the bulk of production was shifted to Newcastle. Paul Theakston set up Black Sheep Brewery in Masham, so you might want to check that out. 3.8% ABV on draft, 4.4% ABV in bottle.
My first ever beer was a Boddington's in the Old Vic, St. Annes. I was young enough not to realise that a bar with sawdust on the floor, a pile of assorted inconvenienced people laying in a somewhat disshevilled heap around the entrance and pool tables where there were more half cues than cues, was a bad sign. Ah, the nectar of Boddington's. The Cream of Manchester.
I have a hardware disk copy unit: plug the old drive in one side, the new drive in the other, push the button, and the drive is cloned to the new one in 5-20 minutes. I upgraded 5 disks in an office from 10GB to 180GB in a relaxed 90 minutes. It will also clone from an internal drive over firewire or USB2.0, so I simply back up a disk whenever I work on a PC now. It also seems to copy drives that Partition Magic won't touch. Details in Japanese here but there's got to be a US equivalent somewhere. $120.
I was using such a device in the early 1980's. Two small boxes, one of which beeped noisily if it was separated from its partner by more than a meter or so. Then VDUs started arriving, and my fickle alarm decided it would remain contentedly silent if there was a monitor anywhere in the building.
My knowledge of Chinese is minimal, to say the least, but doesn't their document "(AVS-CFP-Video Coding) (1.0)" show the contact person to be one Feng Wu at the Microsoft Asia Research Institute's Network Multimedia Group? Author is Wang Ping of Microsoft.
Comparing SD & MMC, you might want to consider that an SD card has a write protect switch which MMC & CF do not (but I expect I'll be corrected on that), and is much faster than MMC (up to 10Mbps vs 300kbs or 1Mbps). Ignore MMC.
If you go with CF, a useful real-world speed comparison is at Rob Galbraith's site.
SD, MMC
Kakaku. Japanese prices for a Japanese product in Japan. Dynamism is offering a bespoke customization service, and charges a premium for that (quite fairly).
Yes, many "traditional" auction houses charge both a buyer's and a seller's premium, plus quite large storage and shipping charges if you cannot immediately walk out with the item (which you can't, if you bid via phone or the internet. The buyer pays quite a lot more than expected, and the seller gets a lot less.
I agree that PDA penetration in the market is less than in the US; you suggest this is because everyone has a laptop instead. Actually, for the casual user, I think it's because everyone has a cellphone instead: most newcomers to the internet in Japan enter via cellphone.
But I see a lot of PDAs in business and tech: visiting a large electronics company last year, everyone at the meeting had company-issued PDA swhich were kept updated over an internal PHS network.
Street price of theC750 is about $375, and $500 for the C760.
In Windows, you may be able to find a battery icon somewhere, and have it tell you the pack's serial number, the initial power capacity of the cell (in mAH) and the current capacity. I actually monitor this across my notebooks.
I have two packs for my main machine: the original short-life pack which I have in the machine when it's home and always plugged in, and an extended-life pack which I only use when I'm out and about. That way, I know exactly how many times it's been charged: 229. Its capacity is down to 80% whilst my original pack is down to 40%.
... but shelf space in stores isn't. There's plenty of good boxed software that never gets to see the light of day because the company is outbid for shelfspace at PCWorld, etc. Online distribution has more benefits than simply reduced (hopefully) price.
Maybe with this or this (the text is Japanese but the pictures are English).
In markets like Iraq, India and (especially) China, the "new" technologies are easier to roll out because there isn't a strong legacy technology to displace. Consider cellphones: in Iraq, cellphone networks seem to be automagically re-emerging because network damage is effectively point failure, since there is relatively little wired backbone to maintain. Whereas restoring a badly-damaged POTS network can take serious time and expense. In China, where there is little legacy technology, cellular networks are cost effective because they are not replacing a POTS network: if cellular isn't built, something else has to be. In Iraq, there is probably a substantial military data network infrastructure that can easily be converted to a public backbone. In other words, VOIP & cellular may be the only sensible options in emerging/recovering economies, and POTS is the expensive option.
I'd hazard a guess that this system will also have trouble with a high-context language such as Japanese.
If all the effort is expended at the point of accumulating the parallel texts, then that's simply lots of computer time but if the text has to be massaged by the user to suit the system at translation time, then that could still be a lot of work.
The approach sounds rather like Translation Memory (as used in Trados and other systems) on a grand scale: "here's a sentence I translated earlier", as Blue Peter would say.
Can't knock it if it works, of course, which it appears to do.
The Excalibur ship in the Crusade successor to B5 also looks rather Liberator-like.
The team designing the original Liberator and its variants didn't have experience of designing props for wire work, so the Liberator was difficult to hang and manipulate. It was slipups like this that made some effects seem cheaper than they were.
Since one is probably using fast film at a sports event, grain becomes a problem as the image is enlarged. There's no grain on a digital image, and enhancement tools with pixel interpolation can keep the jaggies at bay. In the context of a sporting or other event where I'm not in control of the pace of the action, I'd rather work with digital (even at 4MP) than film.