Apt has a concept of virtual packages which are implemented by several different packages. That way I can easily switch between free or non-free JVM, for example.
So? Google is a closed-source company, nothing wrong with it. However, they ARE friendly to OpenSource and open standards.
"I.e. money changed hands" ???? I meant that it was negotiated with satellite imagery providers. Do you expect them to work for free? Also, Google API Key does not restrict you or your users in any way. And it is not even used to track your users.
Also, Apple forces you to _pay_ for a privilege to access their SDK. MS also has code-signing restrictions for mobile applications, ditto for Symbian.
Actually, Google Android is about the ONLY mobile SDK which will allow user to override default call handlers, address books and audio handling with a simple installation of unsigned application.
I'm sorry, I don't really like Google (especially their data storage policy), but Android is probably going to be the most open mobile platform.
I can't remember a Google project which started as Open Source and then was closed-sourced.
iPhone can not be used as a navigator. It has some _similar_ functionality, but it's nowhere close to, say, TomTom. And yes, probably, it was negotiated.
I don't understand what is so restrictive in Google Maps. Yes, Google wants to make sure you do not abuse them. What is wrong with it? They don't require you to sign NDAs or provide your real name for Google Maps keys.
Google is going to open all parts that matter: OS, GUI environment, Java VM, codecs.
As for Google Maps API - they have to control it because they are legally obliged to do it by contracts with satellite imagery providers. For example, Google Maps can't be used in stand-alone applications or in car navigators.
It's pretty clear that Google NDAs are temporary. Google officials pledged (several times) that they are going to open most of the SDK. Personally, I believe them.
NDA-before-open-disclosure is a fairly common thing, there's nothing strange about it.
Uhm... GWT's JavaScript is not magical. It's pretty clean and understandable. For example, I was able to quickly hack-in support for undeclared runtime exceptions in GWT 1.5 (where they are not supported there anymore).
Let's calculate. The kinetic energy of 1 kg moving at 8km/s is 3.2*10^7J. The enthalpy of H2 combustion is 286kJ/mol but if we take O2 required for this reaction into account it'll be just 15.8kJ/g=1.58*10^7J/kg
So we'll need to burn at least: 1/2(m+Mf)*v^2=Mf*Jc where m is payload, Mf - mass of fuel and oxidizer, Jc - specific heat of combustion per 1 kg of fuel and oxidizer.
1/2(1+Mf)*6.4*10^7=Mf*1.58*10^7 (1+Mf)*3.2=Mf*1.58 Mf~=5.5kg of stochiometric fuel and oxidizer mix in the _ideal_ case.
That's more than $40/kg in real life (you can use the rocket equation for more precise estimation, I'm too lazy) and I did not took the weight of tanks and engines into account.
Orion is not feasible in the current political situation. Risk of hijacking of a space vessel fueled by _nuclear_ _bombs_ is too high.
Also, I don't really care about "important missions". I want a sustainable continuos space program.
There are projects of nuclear rockets (where a nuclear reactor heats gas to very high temperature), however. Maybe one day something will come out of them.
It's superb! Our embedded server just needs to serve static files with GWT app to client browsers. No need for any active content, so it's fast and painless.
We also have a separate C++ application which implements RPC services for GWT app. It's also small and fast, since it doesn't need to concern itself with serving HTML/js files.
Spontaneous fluctations in magnitude of more than several degrees are HIGHLY improbable (as in "unlikely to happen during the Universe's lifetime").
However, different equipment failures can happen. That's why cables are cooled slightly below the boiling point of helium. Which itself is well below the critical temperature for Nb-Ti and Nb superconductors.
No. One more time: there's NO resistance. In one experiment, for example, there were no measurable current decrease in a magnet after 20 years.
Low-TC superconductors are preferable because they have much higher critical current. Superconductors lose their superconductivity when a high enough magnetic field is applied. This magnetic field can be external or generated by the current passing through the superconductor itself.
Oh, and 1.9K temperature is used because it has a margin of safety for liquid helium (which has 4K boiling point).
Slashdot Mirror
Earthrise on Moon IS possible. Moon's orbit is not a perfect circle, that's why we have http://en.wikipedia.org/wiki/Librations
In the best case Earth rises at about 15 degrees.
Apt has a concept of virtual packages which are implemented by several different packages. That way I can easily switch between free or non-free JVM, for example.
A free Java implementation of MapReduce and GFS (Apache Hadoop) already works fine on 5000 computers cluster.
And there's no real reason why it can't scale further.
So? Google is a closed-source company, nothing wrong with it. However, they ARE friendly to OpenSource and open standards.
"I.e. money changed hands"
???? I meant that it was negotiated with satellite imagery providers. Do you expect them to work for free? Also, Google API Key does not restrict you or your users in any way. And it is not even used to track your users.
Also, Apple forces you to _pay_ for a privilege to access their SDK. MS also has code-signing restrictions for mobile applications, ditto for Symbian.
Actually, Google Android is about the ONLY mobile SDK which will allow user to override default call handlers, address books and audio handling with a simple installation of unsigned application.
I'm sorry, I don't really like Google (especially their data storage policy), but Android is probably going to be the most open mobile platform.
What "prior history"?
I can't remember a Google project which started as Open Source and then was closed-sourced.
iPhone can not be used as a navigator. It has some _similar_ functionality, but it's nowhere close to, say, TomTom. And yes, probably, it was negotiated.
I don't understand what is so restrictive in Google Maps. Yes, Google wants to make sure you do not abuse them. What is wrong with it? They don't require you to sign NDAs or provide your real name for Google Maps keys.
Google is going to open all parts that matter: OS, GUI environment, Java VM, codecs.
As for Google Maps API - they have to control it because they are legally obliged to do it by contracts with satellite imagery providers. For example, Google Maps can't be used in stand-alone applications or in car navigators.
It's pretty clear that Google NDAs are temporary. Google officials pledged (several times) that they are going to open most of the SDK. Personally, I believe them.
NDA-before-open-disclosure is a fairly common thing, there's nothing strange about it.
Nope. You can use GWT with PHP (or any other language) services using XMLHttpRequest/JSON interoperability system in GWT.
Uhm... GWT's JavaScript is not magical. It's pretty clean and understandable. For example, I was able to quickly hack-in support for undeclared runtime exceptions in GWT 1.5 (where they are not supported there anymore).
Yes, of course it's N2O4 (I need to to write less posts after 20h coding session...).
Yes, and UDMH/N2O4 has a much bit worse Isp - about 330 seconds against 350-380 for LOX/CH4.
A lot of Russian rockets actually use http://en.wikipedia.org/wiki/Unsymmetrical_dimethylhydrazine and NO3 as an oxidant. And Methane + O2 is a bit worse than UDMH+NO3 mix.
Personally, I like the idea of nuclear rockets. But they are just too far from reality.
"Spaceship" 1 is garbage. It's not even close to orbital speed (its maximum speed was 3518km/h while you need about 29000km/h to enter the LEO).
The whole "two stage" system is also mostly junk it just gives an extra 1000km/h which is totally lost when compared with the orbital speed.
Thanks for correction!
I really should have used the rocket equation rather than simple energy balance.
That's really interesting, if we can cut launch costs to about 3x fuel costs (as in airplanes) then we'll have a pretty viable space transport system.
$20 per pound is WAY too small.
Let's calculate. The kinetic energy of 1 kg moving at 8km/s is 3.2*10^7J. The enthalpy of H2 combustion is 286kJ/mol but if we take O2 required for this reaction into account it'll be just 15.8kJ/g=1.58*10^7J/kg
So we'll need to burn at least:
1/2(m+Mf)*v^2=Mf*Jc where m is payload, Mf - mass of fuel and oxidizer, Jc - specific heat of combustion per 1 kg of fuel and oxidizer.
1/2(1+Mf)*6.4*10^7=Mf*1.58*10^7
(1+Mf)*3.2=Mf*1.58
Mf~=5.5kg of stochiometric fuel and oxidizer mix in the _ideal_ case.
That's more than $40/kg in real life (you can use the rocket equation for more precise estimation, I'm too lazy) and I did not took the weight of tanks and engines into account.
Orion is not feasible in the current political situation. Risk of hijacking of a space vessel fueled by _nuclear_ _bombs_ is too high.
Also, I don't really care about "important missions". I want a sustainable continuos space program.
There are projects of nuclear rockets (where a nuclear reactor heats gas to very high temperature), however. Maybe one day something will come out of them.
The main problem is: chemical rockets suck.
There's just no way to cheaply lift payload to orbit using our current rockets. That's why there's no revolutions in spacecraft-building.
We need something like space-plane, launch loops or space elevator for new space revolution.
Actually, we use GWT in our embedded devices.
It's superb! Our embedded server just needs to serve static files with GWT app to client browsers. No need for any active content, so it's fast and painless.
We also have a separate C++ application which implements RPC services for GWT app. It's also small and fast, since it doesn't need to concern itself with serving HTML/js files.
There's no public stats, but I have personally worked on GWT projects for some big companies.
You can indirectly measure the level of activity by checking mailing lists of GWT and GWT-related projects. There's a fair lot of traffic.
There's a lot of corporative GWT-apps because it's, probably, the best toolkit for rich-client web applications.
However, it's not used much in the public web because most sites just don't need that kind of user interfaces.
Also, GWT is incompatible with web spiders.
Moonlight (Novell version of Silverlight) supports only parts of Silverlight 1.0.
Silverlight 2.0 is MUCH MUCH MUCH bigger (basically, it's a cut-down version of CLR) and it'll take a loooong time to be implemented on Mono.
And by the time it's implemented MS will have Silverlight 3.0
Quench is quite probable if, for example, some pipe ruptures - no need for some improbable event :)
Thanks for explanation!
(English is not my native language)
Spontaneous fluctations in magnitude of more than several degrees are HIGHLY improbable (as in "unlikely to happen during the Universe's lifetime").
However, different equipment failures can happen. That's why cables are cooled slightly below the boiling point of helium. Which itself is well below the critical temperature for Nb-Ti and Nb superconductors.
No. One more time: there's NO resistance. In one experiment, for example, there were no measurable current decrease in a magnet after 20 years.
Low-TC superconductors are preferable because they have much higher critical current. Superconductors lose their superconductivity when a high enough magnetic field is applied. This magnetic field can be external or generated by the current passing through the superconductor itself.
Oh, and 1.9K temperature is used because it has a margin of safety for liquid helium (which has 4K boiling point).
I did say "with some caveats in strong magnetic fields" :)