It was impractical to go near to the speed of light, too much energy, and the acceleration to keep humans alive would be too slow. This kills those two birds.
franl is right (grandparent post), you would see the ship moving faster than light, but it's impossible to communicate information faster than the speed of light with that effect. It's like the search light across the moon effect.
OK, and the OMS nozzles are the nozzles who's size is in between the big SSME bells and the small RCS nozzles, which are mostly hidden away. (Do I have this right?)
Yeah, the main reason cited for choosing a J-2 over the Space Shuttle Main Engine (SSME) is that the J-2 was always air lit, as the 2nd and 3rd stages of the Saturn V (the first stage used 5 F-1 engines).
However, I believe that the SSME is also lit in orbit, for the de-orbit burn, and possibly for orbit changing maneuvers (like when a space junk collision is likely). When lit in space, it uses fuel from 3 onboard fuel tanks rather than the big orange tank used for take off.
I guess the difference there is that no testing of the SSME has been done lighting it in the atmosphere.
If they didn't filter search results to the Chinese govt's liking, there would be three outcomes:
1. China does it for them, blocking certain searches and sites, resulting in a slightly clunkier UI for chinese (broken links etc)
2. China blocks google altogether
3. Google finds a way around the Great Firewall of China
What Google have done is basically #1, preserving the user experience. If Google stood their ground and demanded unfiltered results, #2 would probably happen, which would probably be more moral, but brave and stupid. Google could refuse to back down, China could then strike a deal with another search company, and Google would lose out big time (no Chinese bling), with no better moral outcome (there would still be the same level of censorship in China).
If Google did #3, that would be a form of electronic, um, police action on China, which is probably amoral and illegal for a company to do.
According to the post it's 1 billion database entries for a total of 22TB, so it's about 22kB per entry. Quite what's in each entry I'm not sure, probably whole genes (I think they're about 10000 base pairs, and a base pair could be encoded as two bits (there are 4 different ones), but they're probably encoded as bytes (G, C, T or A).
And remember this: it was the solid rocket boosters, and later the external tank, that destroyed two shuttles. Air Force: our thanks...
Well, the design with the solids and external tank was the economically acceptable plan, and design/use problems with these components did lead to the destruction of two shuttles, but that doesn't make it the Air Force's fault. Maybe the original pre-Air Force would have had problems too, space vehicles complex designs after all.
Well, the impression that I got from the interview is that they're mostly concentrating on development, making development easier without worrying about affecting other areas of the OS and keeping compatibility.
The only other thing they pushed was M$ being innovative, but that's more clever software than ground-breaking academic type of initiatives.
Google cache only retrieves the page source from its page cache. It doesn't wait for the original server to serve it images, it just lets your browser get them. It probably does some link canonicalizing (is that a word?) on the source too, but it all happens at google's data centres.
I think you'll find that those 3152 Ars Technica subscribers are simply viewing Ars Technica through Bloglines, rather than handing over the equivalent of $0.5 mil in subscriptions to Ars Technica directly. Could be wrong though.
Yes, but is Privacy necessary for Freedom? Some might say that you don't have freedom if you don't have privacy, but this still doesn't make Privacy == Freedom.
SQLite databases are just a single binary file (although it supports building a database in memory too).
Thx for the info, I could have RTFM I guess.
Before you moan, that's one of the features/goals of SQLite. When you need that simplicity, it's great.
That's fine I think, I guess it just bogs down when the number of rows gets to a certain threshold. It sounds like the DB has to be loaded at least partially into memory each time a script is executed, whereas the advantage of a dedicated daemon/service is that the binary is always running. Does it do anything fancy like "stay resident" when PHP is used in CLI mode?
That Mysql 5.0.x not PHP 5.x.x right? I've had plenty of experience with PHP 5 tripping up big LAMP apps, mostly because of changes in array_merge(). Now 5.1.x trips up apps that pass unreferencable variables to functions that have parameters passed by reference,
function foo(@a) { a=1 }
bar(baz(1));
Previously PHP gracefully switched to pass by value.
Yeah, I've been aware of the date problem with MySQL, it only checks if the day is in the range 1..31. It kind of makes sense if you're doing a lot of copies withing DB's (copy one date field to another date field, etc), but you tend to do that sort of stuff with relations.
Basically it surprises me that it stores the date as text, I would expect it to use number of days since 1970 (Unix) or 1 AD.
Ah, shit, I'm feeding a troll, aren't I? I wasted fifteen minutes of my life and I can never have them back.....
Uh, yeah, that's an affirmative.
Parent post:
What we've found is that for lighter tasks (such as many webpages), SQLite often works quite well. And once/if you need a database system that's more capable, it's best just to move to PostgreSQL.
Yeah, I'm no DB monkey but I certainly administer them, and just installed Postgre and it's PHP management thingy (phpPGadmin), and it seems quite slick. Can you just do an SQL dump from MySQL and put it into Postgre, no problems? I'm talking phpBB and PHP-Nuke DB's, nothing fancy.
And, uh, where the hell does SQLite store it's tables? Never used it.
Back to the flat text file issue, they're great until you edit them, then you have to do locking if the site has anything more than a few concurrent visitors. Then as you get busier, you have to worry about race conditions with the locks... I don't know if this one has been thought through.
I believe the Asynchronous part refers to two methods of send()ing the httpRequest—Synchronous means that by the time your send() call returns, it's either succeeded or failed; Asynchronous means it returns immediately, and you have to add an event handler to use the data once it's been returned. The former is considered unusable on production servers because it will lock up (???) JavaScript until the XML has been read.
DHTML is a web page that dynamically changes it's appearance, but without a response (or extra data) from the server, in response to either user actions (menus appearing in response to a mouseover) or timed events (ubiquitous countdown clocks). DHTML is usually the former.
AJAX is dynamically updated a page with data from the server, in response to either user action (like clicking on one of your labels in gmail) or again, some timed event, like when Gmail updates your inbox every 5 mins. The latter is still really clunky, but there is no other way without persistant connections, which would probably put a new sort of strain on servers, and I don't think is possible with JavaScript.
Interesting paper, but doesn't really make a difference to the Moon/Earth compositional difference. It's just saying that the "C1 model" doesn't hold for Mars. If it didn't hold for the moon, then maybe that's because it lost a lot of it's native material during the Earth/Theia impact, and Theia did fit the C1 model--or not. Or maybe the moon is captured, and it does or it doesn't fit the C1 model.
Conservation of Angular Momentum would play an important role, so and the direction that the proto-moon blobbed out would be fairly aribitrary, so I would guess it just turned out that the earth ended up with an AM vector pointed at 23.5 degs to the ecliptic.
This is my problem. If angular momentum is being conserved, and if the tilt of the axis is even half of where it is now then one would expect the moon to follow an orbit near the celestian equator at the time it separated.
Yeah, I see what you're saying, but (a), what I said before, I don't know what tidal forces can do to these things, specifically the inclination of the proto-moons orbit, and (b), maybe Theia hit at the right trajectory to give the proto-Earth/Theia blob a tilt of about 23 degrees, and then a little bit (say 5%) into a low inclination orbit, maybe following the original trajectory of theia, of which 1.2% became the moon (yep, the moon is 1.2% of the mass of the Earth). This would make hardly any difference to the Earth-Theia blob's AM, so it may have nudged the vector up or down a bit. I think that's another of the successes of the Earth/Theia theory, it accounts for the axial tilt of the earth, and the Moon's orbital properties don't really come into that.
But this theory was mandated by a theory (that carbonaceous chondrites were the original planetessimals) that has suffered some serious setbacks in the last couple of years. The question is: If the moon was co-formed with the earth, why are the elemental makeup of the body so different? The answer that the impact theory postulates is that the moon was formed by material from the Earth's crust.
Interesting, what are the setbacks? As for the composition differences, there basically automatic in the theory. If the two bodies mixed well, they'd be the one body forever. The moon and the earth have the same Oxygen isotope ratios, which supports the impact theory.
For this theory to work, Earth must have had a very low axis tilt (so that the angular momentum of the ejection mass was close to the ecliptic), and the object must have hit close to the equator but also had managed to knock the earth off its axis (to the angle it is in today). I just don't see this happening in any scenario I can run through quickly in my head.
I had assumed that either the original study had taken care of this, but maybe not. As for running it through your head, I thought it basically has to be done through supercomputer, because it's not just a 2 body collision. Conservation of Angular Momentum would play an important role, so and the direction that the proto-moon blobbed out would be fairly aribitrary, so I would guess it just turned out that the earth ended up with an AM vector pointed at 23.5 degs to the ecliptic. Even then, I've read that the earth's axis tilt has shifted significantly since that time, not just through precession.
In any case, a quick google shows that they haven't run through all the different impact angles and resulting angular momenta for this anyway.
Captured objects are generally far from the ecliptic (Phobos, Deimos, Pluto). Phobos and Deimos are not terribly spherical and look more like captured asteroids than like co-formed moons.
The question is did/could tidal forces either from the earth or other planets cause the moon's orbit to migrate to the ecliptic? Phobos and Deimos would be immune because they could easily be recent captures.
But if this theory goes, then it seems to me that there are alternatives that seem more likely for the question of the formation of the moon.
This is where you trot out your goofy theory...;-)
... I suspect that our moon was formed along with the earth in the same band-- the fact that the moon is a near perfect sphere, and that it is within a couple degrees of the ecliptic support this hypothesis I think better than the idea that either the moon was ejected from the earth or that it was captured
The most plausible theory for the formation of the Moon is that the earth was hit by a Mars-sized body, and a bit proto-earth mantle blobbed up, found its way into orbit, and became the moon.
The capture hypothesis is the default, but basically doen't work because of gravitational mechanics. The chances of the moon coming in at the right speed and angle are too low.
The fact that the moon is a good sphere doesn't really rule out any hypothesis. It's big enough to so that it must be round, because it would have to have become molten when it formed (all the gravitational energy from all the mass collapsing in).
I suspect the fact that it's orbit is close to the ecliptic doesn't tell much about its formation either, because of post-formation tidal forces and nudges from other solar system objects, but I'm not sure about this one.
How would useful links be lost forever? Yeah, they wouldn't have their PageRank increased, whereas by rights they should be counted if they are a truly useful and relevant link entered by a blog responder, but at least the URL is out there, and the link will find its way into non-blog reply pages (without the rel="nofollow" attribute).
Slashdot Mirror
It was impractical to go near to the speed of light, too much energy, and the acceleration to keep humans alive would be too slow. This kills those two birds.
franl is right (grandparent post), you would see the ship moving faster than light, but it's impossible to communicate information faster than the speed of light with that effect. It's like the search light across the moon effect.
OK, and the OMS nozzles are the nozzles who's size is in between the big SSME bells and the small RCS nozzles, which are mostly hidden away. (Do I have this right?)
Yeah, the main reason cited for choosing a J-2 over the Space Shuttle Main Engine (SSME) is that the J-2 was always air lit, as the 2nd and 3rd stages of the Saturn V (the first stage used 5 F-1 engines).
However, I believe that the SSME is also lit in orbit, for the de-orbit burn, and possibly for orbit changing maneuvers (like when a space junk collision is likely). When lit in space, it uses fuel from 3 onboard fuel tanks rather than the big orange tank used for take off.
I guess the difference there is that no testing of the SSME has been done lighting it in the atmosphere.
... they make a new Firefox theme that looks like IE7? The interface looks great.
Just a thought.
If they didn't filter search results to the Chinese govt's liking, there would be three outcomes:
1. China does it for them, blocking certain searches and sites, resulting in a slightly clunkier UI for chinese (broken links etc)
2. China blocks google altogether
3. Google finds a way around the Great Firewall of China
What Google have done is basically #1, preserving the user experience. If Google stood their ground and demanded unfiltered results, #2 would probably happen, which would probably be more moral, but brave and stupid. Google could refuse to back down, China could then strike a deal with another search company, and Google would lose out big time (no Chinese bling), with no better moral outcome (there would still be the same level of censorship in China).
If Google did #3, that would be a form of electronic, um, police action on China, which is probably amoral and illegal for a company to do.
According to the post it's 1 billion database entries for a total of 22TB, so it's about 22kB per entry. Quite what's in each entry I'm not sure, probably whole genes (I think they're about 10000 base pairs, and a base pair could be encoded as two bits (there are 4 different ones), but they're probably encoded as bytes (G, C, T or A).
Informative comment, but...
Well, the design with the solids and external tank was the economically acceptable plan, and design/use problems with these components did lead to the destruction of two shuttles, but that doesn't make it the Air Force's fault. Maybe the original pre-Air Force would have had problems too, space vehicles complex designs after all.
Maybe that's what the author said, I don't know.
Well, the impression that I got from the interview is that they're mostly concentrating on development, making development easier without worrying about affecting other areas of the OS and keeping compatibility.
The only other thing they pushed was M$ being innovative, but that's more clever software than ground-breaking academic type of initiatives.
Google cache only retrieves the page source from its page cache. It doesn't wait for the original server to serve it images, it just lets your browser get them. It probably does some link canonicalizing (is that a word?) on the source too, but it all happens at google's data centres.
Dude, instead of
Try
How many global variables do you need?!?
Yeah, I guess it could lead to a DLL-hell type of situation sometimes, but it's pretty minor.
My reaction exactly. Which one is the F#!@^ing story?
I think you'll find that those 3152 Ars Technica subscribers are simply viewing Ars Technica through Bloglines, rather than handing over the equivalent of $0.5 mil in subscriptions to Ars Technica directly. Could be wrong though.
Yes, but is Privacy necessary for Freedom? Some might say that you don't have freedom if you don't have privacy, but this still doesn't make Privacy == Freedom.
Thx for the info, I could have RTFM I guess.
That's fine I think, I guess it just bogs down when the number of rows gets to a certain threshold. It sounds like the DB has to be loaded at least partially into memory each time a script is executed, whereas the advantage of a dedicated daemon/service is that the binary is always running. Does it do anything fancy like "stay resident" when PHP is used in CLI mode?
That Mysql 5.0.x not PHP 5.x.x right? I've had plenty of experience with PHP 5 tripping up big LAMP apps, mostly because of changes in array_merge(). Now 5.1.x trips up apps that pass unreferencable variables to functions that have parameters passed by reference,
Previously PHP gracefully switched to pass by value.
Yeah, I've been aware of the date problem with MySQL, it only checks if the day is in the range 1..31. It kind of makes sense if you're doing a lot of copies withing DB's (copy one date field to another date field, etc), but you tend to do that sort of stuff with relations.
Basically it surprises me that it stores the date as text, I would expect it to use number of days since 1970 (Unix) or 1 AD.
From grandparent post
Uh, yeah, that's an affirmative.
Parent post:
Yeah, I'm no DB monkey but I certainly administer them, and just installed Postgre and it's PHP management thingy (phpPGadmin), and it seems quite slick. Can you just do an SQL dump from MySQL and put it into Postgre, no problems? I'm talking phpBB and PHP-Nuke DB's, nothing fancy.
And, uh, where the hell does SQLite store it's tables? Never used it.
Back to the flat text file issue, they're great until you edit them, then you have to do locking if the site has anything more than a few concurrent visitors. Then as you get busier, you have to worry about race conditions with the locks... I don't know if this one has been thought through.
I believe the Asynchronous part refers to two methods of send()ing the httpRequest—Synchronous means that by the time your send() call returns, it's either succeeded or failed; Asynchronous means it returns immediately, and you have to add an event handler to use the data once it's been returned. The former is considered unusable on production servers because it will lock up (???) JavaScript until the XML has been read.
AJAX is dynamically updated a page with data from the server, in response to either user action (like clicking on one of your labels in gmail) or again, some timed event, like when Gmail updates your inbox every 5 mins. The latter is still really clunky, but there is no other way without persistant connections, which would probably put a new sort of strain on servers, and I don't think is possible with JavaScript.
Take a look at http://carnegieinstitution.org/news_980917.html
Interesting paper, but doesn't really make a difference to the Moon/Earth compositional difference. It's just saying that the "C1 model" doesn't hold for Mars. If it didn't hold for the moon, then maybe that's because it lost a lot of it's native material during the Earth/Theia impact, and Theia did fit the C1 model--or not. Or maybe the moon is captured, and it does or it doesn't fit the C1 model.
This is my problem. If angular momentum is being conserved, and if the tilt of the axis is even half of where it is now then one would expect the moon to follow an orbit near the celestian equator at the time it separated.
Yeah, I see what you're saying, but (a), what I said before, I don't know what tidal forces can do to these things, specifically the inclination of the proto-moons orbit, and (b), maybe Theia hit at the right trajectory to give the proto-Earth/Theia blob a tilt of about 23 degrees, and then a little bit (say 5%) into a low inclination orbit, maybe following the original trajectory of theia, of which 1.2% became the moon (yep, the moon is 1.2% of the mass of the Earth). This would make hardly any difference to the Earth-Theia blob's AM, so it may have nudged the vector up or down a bit. I think that's another of the successes of the Earth/Theia theory, it accounts for the axial tilt of the earth, and the Moon's orbital properties don't really come into that.
Anyway, good chatting, I think I've said enough.
But this theory was mandated by a theory (that carbonaceous chondrites were the original planetessimals) that has suffered some serious setbacks in the last couple of years. The question is: If the moon was co-formed with the earth, why are the elemental makeup of the body so different? The answer that the impact theory postulates is that the moon was formed by material from the Earth's crust.
Interesting, what are the setbacks? As for the composition differences, there basically automatic in the theory. If the two bodies mixed well, they'd be the one body forever. The moon and the earth have the same Oxygen isotope ratios, which supports the impact theory.
For this theory to work, Earth must have had a very low axis tilt (so that the angular momentum of the ejection mass was close to the ecliptic), and the object must have hit close to the equator but also had managed to knock the earth off its axis (to the angle it is in today). I just don't see this happening in any scenario I can run through quickly in my head.
I had assumed that either the original study had taken care of this, but maybe not. As for running it through your head, I thought it basically has to be done through supercomputer, because it's not just a 2 body collision. Conservation of Angular Momentum would play an important role, so and the direction that the proto-moon blobbed out would be fairly aribitrary, so I would guess it just turned out that the earth ended up with an AM vector pointed at 23.5 degs to the ecliptic. Even then, I've read that the earth's axis tilt has shifted significantly since that time, not just through precession.
In any case, a quick google shows that they haven't run through all the different impact angles and resulting angular momenta for this anyway.
Captured objects are generally far from the ecliptic (Phobos, Deimos, Pluto). Phobos and Deimos are not terribly spherical and look more like captured asteroids than like co-formed moons.
The question is did/could tidal forces either from the earth or other planets cause the moon's orbit to migrate to the ecliptic? Phobos and Deimos would be immune because they could easily be recent captures.
But if this theory goes, then it seems to me that there are alternatives that seem more likely for the question of the formation of the moon.
This is where you trot out your goofy theory... ;-)
I'm not a pro astronomer either.
The most plausible theory for the formation of the Moon is that the earth was hit by a Mars-sized body, and a bit proto-earth mantle blobbed up, found its way into orbit, and became the moon.
The capture hypothesis is the default, but basically doen't work because of gravitational mechanics. The chances of the moon coming in at the right speed and angle are too low.
The fact that the moon is a good sphere doesn't really rule out any hypothesis. It's big enough to so that it must be round, because it would have to have become molten when it formed (all the gravitational energy from all the mass collapsing in).
I suspect the fact that it's orbit is close to the ecliptic doesn't tell much about its formation either, because of post-formation tidal forces and nudges from other solar system objects, but I'm not sure about this one.
Don't bullshit.
How would useful links be lost forever? Yeah, they wouldn't have their PageRank increased, whereas by rights they should be counted if they are a truly useful and relevant link entered by a blog responder, but at least the URL is out there, and the link will find its way into non-blog reply pages (without the rel="nofollow" attribute).