Not quite true. If you don't need more than the basic user-grantable capabilities, you can self-sign your application, distribute it and just tell the users to ignore the warnings they see on install.
If you need more capabilities and you want anyone to be able to install your software, _then_ you need to go through the testing process.
It's not possible to run _unsigned_ apps, but it is possible to run self-signed apps, if you enable this in the settings. Self-signing means signing with an untrusted key and certificate you can generate by yourself using the tools in the SDK.
Self-signed apps can access just a subset of capabilities, though.
You are right in that the API bindings aren't yet complete - and improving them is a task that will go on for a long while still. Those API bindings are the part that takes the most effort in the porting work. People are welcome to help out - the whole thing is open source and we do accept good patches.
BTW, the sys.path Unicode problem has been fixed since release 1.3.11 and we're up to 1.3.14 already (released on Monday). Are you still using version 1.3.8, released in July? I recommend you get that latest version, since it has a heap of bug fixes and some nice new features like support for OpenGL ES and the text-to-speech feature of S60 2.8 and later.
The Keyboard class that you see duplicated is just from some example code that demonstrates how to write a keyboard event handler. Making that feature available by default is pending a more extensive overhaul of the UI framework that will have to wait until next year. Porting the thing to S60 3rd edition is what has been taking most of our time.
The nice people at HIIT have made a wxWidgets implementation of the Python for Series 60 appuifw UI widget API, so now you run the same UI code on your desktop machine and on your phone, without needing the Symbian emulator. The code is not quite complete yet, but a lot of stuff _is_ supported. They also released their PyExpat port, which is a good example of how to extend Python for Series 60.
Yep, I've seen it before. When I first saw that pure red and blue appeared to be at different depths I thought the phosphors on the screen were actually placed that way. After a while I decided that made no sense and that it had to be just the chromatic aberration in my eyes.
Here is a simple
demo picture I made of the effect. Does everyone see blue as closer than the red?
Try setting the sample rate of your mouse to the same as the screen refresh rate. I use 80 Hz for both the mouse and the screen and pointer movement looks very smooth.
K6 processors may have accidentally obtained their reputation of instability. There probably is a correlation between a K6 processor and the overall system 'tweakedness' factor (overclocking, fiddling with memory timings, lots of cruel and unusual peripherals etc.), because the K6 processor is popular among computer hobbyists. Many processors in tweaked systems naturally leads to many reports of problems, and the correlation will make it appear that the processor is the component to blame, since it is practically always mentioned in the reports. The same phenomenon will probably affect the Athlon line of processors as well.
...about 1 microcurie of the stuff... ...criticality of Am241 occurs at about 2 curies... ...order 2,000 good and fresh smoke detectors...
That would be about 2 _million_ fresh smoke detectors.
Re:What we really need is...
on
Linux BIOS
·
· Score: 1
The fact that the original PC had a ROM BASIC is, by the way, still visible in modern PC's in the form of one of the most confusing error messages of all time. When you boot a PC and no partition on the hard drive is marked bootable (IIRC), the BIOS will say: "No ROM BASIC";)
I just spent quite a while searching the HelixCode site for a list of mirrors, but couldn't find any. Am I just being blind or is the list not there...?
After a calculation that was a bit too lengthy to fit in this comment box, I have some figures on that meteor shower... I tried to compute an upper limit on the amount of debris that could possibly end up on Earth after a nuclear detonation on the Moon as follows: Assume that the entire energy of the nuclear blast is converted to kinetic energy of lunar rock, and that the chunk (or cloud...) of rock will travel in a straight line to Earth. Optimally, all of the rocks will have exactly the energy to get to the point between the Earth and the Moon where their gravity is equal and opposite. The gravitational potential of this point is about 2.4 MJ/kg higher than the potential on the lunar surface. If we take a 1 MT nuke, detonate it and haul all the rock we can possibly get to Earth with that energy we can ship 1.7 million tons, or a rock cube with a side of 80 metres!
Fortunately a surface explosion would direct most of its energy towards the Moon and wouldn't send much to Earth.
To be honest, I have no idea. This article has a few interesting takes on the matter. The two major problems in terraforming Venus are
Heat. Venus receives about twice as much solar power than Earth.
Chemical composition. 96% carbon dioxide in the atmosphere makes a hell of a greenhouse effect.
Being so close to the Sun as Venus is, it would need a lot more dust in the upper atmosphere than Earth to produce a nuclear winter -like conditions, if it is possible at all. In the end, warming up Mars is probably easier than cooling down Venus.
The mass of the Moon = 7.348E22 kg Average orbital velocity of the Moon = 1023 m/s Kinetic energy of the Moon: 1/2*mv^2 = 3.84E28 J Converted to megatons of TNT (4.187E15 J): 9.18E12 MT.
The entire US nuclear arsenal contains a bit over 2000 megatons of energy (1996, source) -- 4.6E9 times less. I couldn't find the entire world arsenal, but it can't be more than 10000 MT. In addition, if the weapon(s) were detonated on the surface, most of the energy would be wasted, because there would be very little mass for the weapons to hurl off the surface. (Conservation of momentum - to push the Moon in some direction you must push something else in the opposite direction.)
In conclusion: Even if we detonated all the nukes of the world up there, the Moon's orbit wouldn't change noticeably.
Two _decades_?!? Just out of curiosity, has anyone of Slashdot's readers ever got to play with a machine with >20 yrs uptime? What does it take to keep a machine up and running for that long? For example, a lot of parts surely break and are replaced during that period. Do the machines have any original parts left?
It would indeed be an uber-cool feature to have, but with both Intel and AMD concentrating fully on getting the highest MHz numbers stamped on the CPU's it isn't likely we will see this any time soon. Raw processing power is the thing that sells processors nowadays, and full virtualization is a niche feature. And it's been done in software already... </cynic mode>
On the other hand, increasing processor power means that the overhead from running the 'desperate' software virtualizer fortunately becomes more irrelevant each day. The x86 family has traditionally used sheer raw power to compensate for the backwards compatible design flaws.
By the way, does anyone know extensively the full virtualization capability is integrated into the operating system of those IBM mainframes? How easy is it to create new a VM? Is it like running a program in a chroot environment, or like VMWare, where you have to boot an entirely new copy of the operating system?
Well, actually chopping your thoughts up in 10ms time slices might be almost enough, since the fastest neuron firing rates in the brain are about 500 Hz, IIRC. There may be some issues related to the precise timing differences between firings of different neurons, but the bottom line is that the serial processing speed of the brain isn't much - less than 1000 operations per second at the neuron level.
Slashdot Mirror
Not quite true. If you don't need more than the basic user-grantable capabilities, you can self-sign your application, distribute it and just tell the users to ignore the warnings they see on install.
If you need more capabilities and you want anyone to be able to install your software, _then_ you need to go through the testing process.
It's not possible to run _unsigned_ apps, but it is possible to run self-signed apps, if you enable this in the settings. Self-signing means signing with an untrusted key and certificate you can generate by yourself using the tools in the SDK.
Self-signed apps can access just a subset of capabilities, though.
You are right in that the API bindings aren't yet complete - and improving them is a task that will go on for a long while still. Those API bindings are the part that takes the most effort in the porting work. People are welcome to help out - the whole thing is open source and we do accept good patches.
BTW, the sys.path Unicode problem has been fixed since release 1.3.11 and we're up to 1.3.14 already (released on Monday). Are you still using version 1.3.8, released in July? I recommend you get that latest version, since it has a heap of bug fixes and some nice new features like support for OpenGL ES and the text-to-speech feature of S60 2.8 and later.
The Keyboard class that you see duplicated is just from some example code that demonstrates how to write a keyboard event handler. Making that feature available by default is pending a more extensive overhaul of the UI framework that will have to wait until next year. Porting the thing to S60 3rd edition is what has been taking most of our time.
You mean something like this?
Not yet, since Python for S60 3rd edition hasn't been released yet. We're working on it.
The nice people at HIIT have made a wxWidgets implementation of the Python for Series 60 appuifw UI widget API, so now you run the same UI code on your desktop machine and on your phone, without needing the Symbian emulator. The code is not quite complete yet, but a lot of stuff _is_ supported. They also released their PyExpat port, which is a good example of how to extend Python for Series 60.
The probability of moving to Python 2.4 is high, but no definite schedule has been decided yet. Sometime next year, probably.
But anyway, I'm one of the developers in this project, so if you have any questions or comments then post away and I'll try to answer.
Here is a simple demo picture I made of the effect. Does everyone see blue as closer than the red?
322 thousand litres, not 322 litres! (Or, with slightly more accurate figures, 327 thousand litres). That would fit in a cube with 7 m sides.
Try setting the sample rate of your mouse to the same as the screen refresh rate. I use 80 Hz for both the mouse and the screen and pointer movement looks very smooth.
K6 processors may have accidentally obtained their reputation of instability. There probably is a correlation between a K6 processor and the overall system 'tweakedness' factor (overclocking, fiddling with memory timings, lots of cruel and unusual peripherals etc.), because the K6 processor is popular among computer hobbyists. Many processors in tweaked systems naturally leads to many reports of problems, and the correlation will make it appear that the processor is the component to blame, since it is practically always mentioned in the reports. The same phenomenon will probably affect the Athlon line of processors as well.
...about 1 microcurie of the stuff...
...criticality of Am241 occurs at about 2 curies...
...order 2,000 good and fresh smoke detectors...
That would be about 2 _million_ fresh smoke detectors.
The fact that the original PC had a ROM BASIC is, by the way, still visible in modern PC's in the form of one of the most confusing error messages of all time. When you boot a PC and no partition on the hard drive is marked bootable (IIRC), the BIOS will say: "No ROM BASIC" ;)
You really should've made that a real link to see ;)
how many actually click on it
I just spent quite a while searching the HelixCode site for a list of mirrors, but couldn't find any. Am I just being blind or is the list not there...?
Fortunately a surface explosion would direct most of its energy towards the Moon and wouldn't send much to Earth.
- Heat. Venus receives about twice as much solar power than Earth.
- Chemical composition. 96% carbon dioxide in the atmosphere makes a hell of a greenhouse effect.
Being so close to the Sun as Venus is, it would need a lot more dust in the upper atmosphere than Earth to produce a nuclear winter -like conditions, if it is possible at all. In the end, warming up Mars is probably easier than cooling down Venus.Average orbital velocity of the Moon = 1023 m/s
Kinetic energy of the Moon: 1/2*mv^2 = 3.84E28 J
Converted to megatons of TNT (4.187E15 J): 9.18E12 MT.
The entire US nuclear arsenal contains a bit over 2000 megatons of energy (1996, source) -- 4.6E9 times less. I couldn't find the entire world arsenal, but it can't be more than 10000 MT. In addition, if the weapon(s) were detonated on the surface, most of the energy would be wasted, because there would be very little mass for the weapons to hurl off the surface. (Conservation of momentum - to push the Moon in some direction you must push something else in the opposite direction.)
In conclusion: Even if we detonated all the nukes of the world up there, the Moon's orbit wouldn't change noticeably.
Two _decades_?!? Just out of curiosity, has anyone of Slashdot's readers ever got to play with a machine with >20 yrs uptime? What does it take to keep a machine up and running for that long? For example, a lot of parts surely break and are replaced during that period. Do the machines have any original parts left?
It would indeed be an uber-cool feature to have, but with both Intel and AMD concentrating fully on getting the highest MHz numbers stamped on the CPU's it isn't likely we will see this any time soon. Raw processing power is the thing that sells processors nowadays, and full virtualization is a niche feature. And it's been done in software already...
</cynic mode>
On the other hand, increasing processor power means that the overhead from running the 'desperate' software virtualizer fortunately becomes more irrelevant each day. The x86 family has traditionally used sheer raw power to compensate for the backwards compatible design flaws.
By the way, does anyone know extensively the full virtualization capability is integrated into the operating system of those IBM mainframes? How easy is it to create new a VM? Is it like running a program in a chroot environment, or like VMWare, where you have to boot an entirely new copy of the operating system?
The 'autoexpand' functionality appears to work out of the box at least on Debian Potato. 'autolist' works right after saying 'setopt nolistbeep'.
Well, actually chopping your thoughts up in 10ms time slices might be almost enough, since the fastest neuron firing rates in the brain are about 500 Hz, IIRC. There may be some issues related to the precise timing differences between firings of different neurons, but the bottom line is that the serial processing speed of the brain isn't much - less than 1000 operations per second at the neuron level.