I did screw up the numbers, and it is around half a year. I was working in kPa and did the conversion from 1mm Hg to kPa but forgot to double the result.
Re:Slow leak, still way above Denver pressure
on
ISS May Have A Leak
·
· Score: 4, Informative
A little more detail - Denver is the "mile-high city," so figure it's at around that height. Figure sea-level air pressure at 100 kPa (14.5 psi); air pressure at 1 mile is somewhere in the area of 50 kPa (7.25 psi). At a rate of 2 mm of mercury a day they only have about a year before they reach Denver pressure, though I assume they'd want to do something about it well before that....
Slow leak, still way above Denver pressure
on
ISS May Have A Leak
·
· Score: 1
Two quotes from the article:
"There's no action for you at this time and no immediate concerns," Mission Control assured the two men. "We'll continue to investigate this on the next shift and we may have some actions for you tomorrow."
and
Normal pressure inside the space station is 14.7 pounds per square inch, the same as at sea level.
So basically they have a slow leak but they have a fair amount of time to find it.
nobody has explained adequately how Ghost reinstalls your disk image without Windows, since even Windows can't do that
It uses the DOS bootable floppy that you created, or that you create on another system when you need to restore your backup. Other drive imaging packages use the DOS bootable CD that they ship with, or the Linux bootable CD that they ship with, or something similar.
Not only do you not need to have Windows running when you need to do data restoration to a hard drive, in some cases you *can't* have Windows running while you're doing your restoration.
Tablet PCs in the field will help keep us even more accurately up-to-date. Currently, our drivers can only transmit when in range of our larger towers (which are only in the cities), meaning that customers wanting updated tracking information on a package routed to a rural area just have to sit and wait
You don't need tablet PCs to fix that, you need a better communications infrastructure. If replacing your current system is the most reasonable way to get to what you really need that's great, but it's a means not the end goal.
It's perhaps not completely ideal, but it works reasonably well in the experimenting I did. Depending on the drive, it can write directly to CD-Rs with compression (along with other drive types, but I've never tried those). You can do bootable CDs from it for recovery, etc. One interesting thing about it is that it's Linux-based, I suspect based off a standard distribution. Really there's nothing there you couldn't do with what's available standard under most Linux installs, it's just a question of putting it together.
For example, will we learn eventually (probably from the rantings of Rush Limbaugh) that the FBI has ordered Robert Novak and other journalists contacted in the Valerie Plame affair to prepare to turn over all their materials?
I'm guessing not, seeing as how those materials would likely lead to the charging of one or two of Ashcroft's friends and/or colleagues with multiple felonies.
The problem with any WYSIWYG editor is that they promote fiddling with appearance when really the appearance is minor and the content is what matters. For a lot of things I'd actually rather have WordPerfect 5.1 than anything WYSIWYG, because it really doesn't matter whether the text ends halfway down the last page or has an extra 1 pt of spacing between each line to take it to the bottom of the page.
Put differently, it all goes back to the aphorism "Perfection is the enemy of (good/progress/etc.)" which is true not just because in trying to make things perfect you often either ruin them or never finish them - it's true because everyone's idea of perfection is different, but most of us can agree on "pretty good".
What's being talked about here is reverse engineering trust heirarchies, algorithmically, simply from a discussion corpus extracted from Usenet.
I don't remember where I saw it, but I seem to recall that at least one company selling a business workflow product (e.g. expense reports, etc.) would do pretty much the same thing with email - attempt to analyze it to determine the structure of the company. That may've been vapor or marketing fluff, but I know this sort of thing's been discussed in the past.
how often they originate threads (and how those threads go),
how often they terminate threads (particularly short ones - those 2-post threads mentioned in the article or 3-post threads where the third post is also from the originator),
how many different people participate in threads with them (particularly direct responses),
how many groups they participate in,
how diverse those groups are,
assorted other metrics
It should be possible to eventually build a fairly powerful scoring system for both threads and authors, based on scores calculated for those who participate in those threads. The system doesn't even have to know whether the threads in question have good content or not - once you determine who the top 10-15 people in a group are, you can take a look at a representative sample of their posts and classify them as "guru," "solid contributor," "joe everywhere," "clueless n00b," and "troll," or just rank by the general value of that person's posts perhaps on a scale of 1-5.
Once you have rankings for the most significant posters you can actually generate very meaningful scores for threads, and you can most likely generate likely scores for other people as well - for example, the person who posts in every thread but whose posts are never followed is probably just being ignored; the person who posts a lot but mostly in response to someone classified as a troll is probably also going to get a low ranking.
That means that they're only monitoring newsgroups that reach msnews.microsoft.com. That's like saying Google Groups only monitors newsgroups that reach it - it's true, but so what?
Any globally-available group is or can be available on their servers with no significant difficulty. I poked around and came up with local groups (e.g. chi.general) and non-MS language groups (e.g. comp.lang.python). Perhaps you're confusing the msnews.microsoft.com domain with the microsoft.public hierarchy?
R: I'll pay you $6
D: I'll pay you $10, but I need $5 of that in taxes
Who helps the poor?
You left out part of it. It should be "I'll pay you $10, but I need $5 of that in taxes which will cover basic healthcare and help improve your local schools."
I always have issues if I don't eject the card first - the shutdown process "ejects" the card ("cardctl eject" or just cardmgr shutting down), but something else with then hang in the shutdown process and just report timeout errors on/dev/hdf because surprise! it's not there anymore.
I'm sure that if I want to dig into the scripts I can find the cause of this, but considering how often I use the CD it hasn't been worth it so far.
It was already slow as a three-legged turtle with just us regulars trying to post and vote. A direct link in Slashdot front page will surely keep it down at least until it (the comment) scrolls away...
I know, but I figured there might be a chance that in a link-heavy article people wouldn't click through all of them until they looked at the comments. Maybe this'll make the difference on how quickly it recovers.
How were you reading the file? One big chunk, one character at a time, or with an intermediate buffer? An intermediate-sized buffer with data being copied to a new buffer for output will be the most efficient.
Were you trying to make in-place changes to the string? Strings in Python can't change, so with larger inputs you might've been moving a lot more data around behind the scenes.
There are actually several ways to make standalone executables of Python programs; the one I use is Gordon McMillan's Installer, which basically packages up the needed DLLs, libraries, etc. and puts on an exectuable wrapper. It's cross-platform, though not unreasonably you have to compile on Windows to get a Windows exe and on your other platforms to get executables for those. There are others compiling options, the linked page includes a nice list of them with summaries.
Python can also be fairly well optimized; I have a strong suspicion that the slashdotting of trackers is more a bandwidth issue than a processing capacity issue. It's also quite possible (caveat: I haven't read the code yet) that the internal structure placed a higher priority on readability and maintainability than on processing efficiency - I know I've written server applications using that approach, because I'd rather spend an extra $500 on hardware upgrades if more power is needed than an extra $5000 later on additional maintenance, debugging and testing.
Finally, on the issue of speed: a lot of what a tracker does involves managing lists, hash tables, etc. - the type of thing that's built into Python and highly-optimized. It's like someone's discovery when VB5 or VB6 came out that a VB program was faster at some tasks than a corresponding C++ program, because all the processing was being done using very heavily optimized builtins - if you're running a relatively inefficient scripting language but using it to call highly-optimized native code, you may not see the performance degradation that you might expect.
Automatic takeoff and landings systems do not require GPS. The same thing already exists today using smaller-scale systems set up at every airport.
Not quite - it's at every major airport, but not at smaller ones. According to the FAA, the reason they want WAAS (which is effectively a Differential GPS broadcast from a non-military satellite) is to "[provide] precision guidance to aircraft at thousands of airports and airstrips where there is currently no precision landing capability."
First, having access to very accurate positional data is in fact important for modern warfare, particularly when precise targeting is desired. There's a reason why selective availability was used for so long, and a reason why non-military GPS receivers don't operate beyond certain speeds (they work at aircraft speeds but not at missile speeds, and they generally won't operate above a certain altitude - the signals are present, but the manufacturers have limits on speed and altitude in the firmware).
If you're trying to use GPS signals to navigate a missile through the observation dome on someone's palace or perhaps to destroy a TV station next to an apartment block accuracy to within 100 meters isn't acceptable, nor is high accuracy that can be degraded by some other power that disapproves of your moves.
Put differently, what would the US response have been if the GPS system was run by France and they'd degraded the signal quality over the Middle East to be accurate to within a kilometer? It would've made the invasion as it was handled impossible because of the difficulty of doing precision targeting without GPS use. Do you think the US wouldn't do similar things if our leadership disagreed that strongly with something another nation was doing? "Fine, go ahead despite our objections but don't expect our assistance (including the provision of accurate positioning signals)."
Second, Europe is doing this for the same reason the FAA wants WAAS (Wide Area Augmentation System) - so it's possible to do precise navigation for all sorts of purposes, including automated takeoff and landing. Systems like that are useless unless they can actually be relied on, and with SA as an option the reliability of the systems is a political issue not a technical one.
"and they all say that it can't work and isn't a vulnerability. Therefore, here's my incorrect research into security of what I wrongly thought were security holes in smart cards. Since the work itself is useless, what do people like and not like about how I have the paper formatted?"
I don't know if Palm has come out with any b&w units with 320x320, but my Clie T415 is one from Sony and it's not the newest.
Looking at their site, it looks like the SJ20 is their only current high-res monochrome option, but they've had a few in the past, including the SL10 that they currently have available as a $99 refurb.
11: Can the Ambient Orb break?
Well, the electronics are pretty sturdy, but the Orb itself is made of frosted glass. So if you drop it you might still get your information, but the glass probably won't make it.
I'd say the point is to have a conveniently viewable status display. I've seen reports about setups where different systems have different bird sounds associated with various statuses; apparently after a fairly short time changes in the background noise jump right out at you.
I could see using locally-connected glowing globes this for all sorts of monitoring; stock market tracking isn't really near the top of the list. I can see having a row of stuff like this visible in or near a server room for example, showing network latency or traffic load, system load, any of a variety of things.
What the difference between this and assorted other status/alarm LED displays? These are in a translucent block and are more easily visible from a distance. Not such a bad thing.
Heck, I use something similar as a shower timer - it dims over 15 minutes, and if I glance over and it's gone out then I'm probably running late.
Slashdot Mirror
I did screw up the numbers, and it is around half a year. I was working in kPa and did the conversion from 1mm Hg to kPa but forgot to double the result.
A little more detail - Denver is the "mile-high city," so figure it's at around that height. Figure sea-level air pressure at 100 kPa (14.5 psi); air pressure at 1 mile is somewhere in the area of 50 kPa (7.25 psi). At a rate of 2 mm of mercury a day they only have about a year before they reach Denver pressure, though I assume they'd want to do something about it well before that....
It uses the DOS bootable floppy that you created, or that you create on another system when you need to restore your backup. Other drive imaging packages use the DOS bootable CD that they ship with, or the Linux bootable CD that they ship with, or something similar.
Not only do you not need to have Windows running when you need to do data restoration to a hard drive, in some cases you *can't* have Windows running while you're doing your restoration.
You don't need tablet PCs to fix that, you need a better communications infrastructure. If replacing your current system is the most reasonable way to get to what you really need that's great, but it's a means not the end goal.
It's perhaps not completely ideal, but it works reasonably well in the experimenting I did. Depending on the drive, it can write directly to CD-Rs with compression (along with other drive types, but I've never tried those). You can do bootable CDs from it for recovery, etc. One interesting thing about it is that it's Linux-based, I suspect based off a standard distribution. Really there's nothing there you couldn't do with what's available standard under most Linux installs, it's just a question of putting it together.
Instant Recovery
I'm guessing not, seeing as how those materials would likely lead to the charging of one or two of Ashcroft's friends and/or colleagues with multiple felonies.
Aren't double standards wonderful?
Put differently, it all goes back to the aphorism "Perfection is the enemy of (good/progress/etc.)" which is true not just because in trying to make things perfect you often either ruin them or never finish them - it's true because everyone's idea of perfection is different, but most of us can agree on "pretty good".
I don't remember where I saw it, but I seem to recall that at least one company selling a business workflow product (e.g. expense reports, etc.) would do pretty much the same thing with email - attempt to analyze it to determine the structure of the company. That may've been vapor or marketing fluff, but I know this sort of thing's been discussed in the past.
- information on who participates in a group,
- how much they participate,
- how often they originate threads (and how those threads go),
- how often they terminate threads (particularly short ones - those 2-post threads mentioned in the article or 3-post threads where the third post is also from the originator),
- how many different people participate in threads with them (particularly direct responses),
- how many groups they participate in,
- how diverse those groups are,
- assorted other metrics
It should be possible to eventually build a fairly powerful scoring system for both threads and authors, based on scores calculated for those who participate in those threads. The system doesn't even have to know whether the threads in question have good content or not - once you determine who the top 10-15 people in a group are, you can take a look at a representative sample of their posts and classify them as "guru," "solid contributor," "joe everywhere," "clueless n00b," and "troll," or just rank by the general value of that person's posts perhaps on a scale of 1-5.Once you have rankings for the most significant posters you can actually generate very meaningful scores for threads, and you can most likely generate likely scores for other people as well - for example, the person who posts in every thread but whose posts are never followed is probably just being ignored; the person who posts a lot but mostly in response to someone classified as a troll is probably also going to get a low ranking.
Any globally-available group is or can be available on their servers with no significant difficulty. I poked around and came up with local groups (e.g. chi.general) and non-MS language groups (e.g. comp.lang.python). Perhaps you're confusing the msnews.microsoft.com domain with the microsoft.public hierarchy?
D: I'll pay you $10, but I need $5 of that in taxes
Who helps the poor?
You left out part of it. It should be "I'll pay you $10, but I need $5 of that in taxes which will cover basic healthcare and help improve your local schools."
I'm sure that if I want to dig into the scripts I can find the cause of this, but considering how often I use the CD it hasn't been worth it so far.
I know, but I figured there might be a chance that in a link-heavy article people wouldn't click through all of them until they looked at the comments. Maybe this'll make the difference on how quickly it recovers.
Python can also be fairly well optimized; I have a strong suspicion that the slashdotting of trackers is more a bandwidth issue than a processing capacity issue. It's also quite possible (caveat: I haven't read the code yet) that the internal structure placed a higher priority on readability and maintainability than on processing efficiency - I know I've written server applications using that approach, because I'd rather spend an extra $500 on hardware upgrades if more power is needed than an extra $5000 later on additional maintenance, debugging and testing.
Finally, on the issue of speed: a lot of what a tracker does involves managing lists, hash tables, etc. - the type of thing that's built into Python and highly-optimized. It's like someone's discovery when VB5 or VB6 came out that a VB program was faster at some tasks than a corresponding C++ program, because all the processing was being done using very heavily optimized builtins - if you're running a relatively inefficient scripting language but using it to call highly-optimized native code, you may not see the performance degradation that you might expect.
Not quite - it's at every major airport, but not at smaller ones. According to the FAA, the reason they want WAAS (which is effectively a Differential GPS broadcast from a non-military satellite) is to "[provide] precision guidance to aircraft at thousands of airports and airstrips where there is currently no precision landing capability."
Put differently, what would the US response have been if the GPS system was run by France and they'd degraded the signal quality over the Middle East to be accurate to within a kilometer? It would've made the invasion as it was handled impossible because of the difficulty of doing precision targeting without GPS use. Do you think the US wouldn't do similar things if our leadership disagreed that strongly with something another nation was doing? "Fine, go ahead despite our objections but don't expect our assistance (including the provision of accurate positioning signals)."
Second, Europe is doing this for the same reason the FAA wants WAAS (Wide Area Augmentation System) - so it's possible to do precise navigation for all sorts of purposes, including automated takeoff and landing. Systems like that are useless unless they can actually be relied on, and with SA as an option the reliability of the systems is a political issue not a technical one.
Not all of them have banned the possession of more than a few of these licensed Pooh "massagers", but some have.
"and they all say that it can't work and isn't a vulnerability. Therefore, here's my incorrect research into security of what I wrongly thought were security holes in smart cards. Since the work itself is useless, what do people like and not like about how I have the paper formatted?"
Looking at their site, it looks like the SJ20 is their only current high-res monochrome option, but they've had a few in the past, including the SL10 that they currently have available as a $99 refurb.
And since I got about 6-8 inches and all the split ends whacked off of it last Friday, my showers are noticeably shorter.
I could see using locally-connected glowing globes this for all sorts of monitoring; stock market tracking isn't really near the top of the list. I can see having a row of stuff like this visible in or near a server room for example, showing network latency or traffic load, system load, any of a variety of things.
What the difference between this and assorted other status/alarm LED displays? These are in a translucent block and are more easily visible from a distance. Not such a bad thing.
Heck, I use something similar as a shower timer - it dims over 15 minutes, and if I glance over and it's gone out then I'm probably running late.