There's one guy working on a project called Atom/Atomic/Atome or something, which is basically your app compiled in to an OS container, instead of being built on top of an OS container, but still responding similar to a docker container.
As mentioned in the article (I know...), the base for Neon will be Ubuntu LTS. So everything except KDE will be nice and stable, unlike with a rolling distro like Netrunner.
Now the best scenario I can come up with is that you might get apps intended for use in a particular locale outside Five Eyes that are available only in the official or majority language of that locale.
No idea what the Five Eyes are, but my situation is this: I live in Norway, I cannot stand having anything but english as the display language in my OS, but I want to be able to type my name, emails, documents etc using norwegian keyboard layout as the norwegian alphabet contains æÃÃ¥.
Have you reported it to the developer of each affected app? Or does it affect all apps system-wide?
It affects all the _apps_ that are installed by default, so I'll say it's Microsoft. Note that the desktop, and all desktop-related programs, correctly respects my language choice.
And I'm not alone, social.microsoft.com has several posts on the issue from other people.
Would you prefer that an app be made entirely unavailable for download if its (often small) developer has not yet translated the app to your "forced display language"?
My forced display language is English. And why is using my keyboard layout language somehow a better choice than the display language I've explicitly told it I want?
It's superior to 7, 8, and 8.1 in every way that I can think of. Why do people not want a FREE upgrade?
I've tried upgrading 3 times after Windows 10 was launched, and I reverted within a day each time.
For one it BSODs every 20 minutes, even when just sitting there. For comparison, I've only had 2-3 BSODs with Windows 8.1 since I installed it shortly after launch, all game related.
Secondly "apps" are a PITA because they do not follow my forced display language.
A clean install should fix the BSOD issue, but until they sort out the apps I won't be touching Windows 10.
Most databases allow you to do transactions (BEGIN TRAN/COMMIT TRAN) that force it to be atomic.
Sure, but in that case you might get errors when you try to commit if another connection has changed the value in the meantime, and you'll have to retry the whole thing. That's what UPSERT avoids.
Is this summary saying that, prior to now, you could not update data that was in the database?
No. In some cases you want to use a table like a key-value storage. If the key does not exist you need to insert a row with that key. Otherwise you'll want to keep only one row with that key, and just update the value.
One can always do a "if key exists then update else insert", but the problem is that this is not atomic, because the "key exists" check is a separate statement from the insert or update statement. This can lead to issues if you have multiple connections accessing the same keys at the same time.
The UPSERT allows you to do this as one atomic operation.
When I went to uni the first course in statistics required one book at around $200. However the profs had agreed to use the same book in the three follow-up statistics courses as well. So if you were going for a bachelor in statistics, you just bought that one book for all your statistics courses.
Similar story in physics, at least for the first two-three courses. The math courses were a bit more fractured for some reason, though in two of the courses the textbook was written by the prof and provided free as PDF. I got them printed and bound at the uni press for like $15 each.
However my book expenses was nothing compared to what they had to put up with in other areas, such as psychology or pedagogy. They had to buy $200 books just for a chapter or two.
My personal opinion is that life is really, really, really, REALLY rare.
Another factor is that high intelligence isn't necessarily a beneficial thing in terms of evolution. Cockroaches and rats will probably outlast the human species. Bacteria, algae and similar most certainly will.
If you can't provide a truly unlimited service, don't advertise it
They should just adopt a kind of Gabriel Horn's solution to the problem.
Yes you can upload and download as much as you want 24/7, hence unlimited access, but your bandwidth will be reduced in steps based on transferred data, so that in the end you'd only have a few kbps.
I just sent it back to the vendor (under warranty) because the latest OS update caused the speakers to melt.
While browsing the net I noticed the smell of burnt plastic, and quickly noticed the edges (where the speakers are) was too hot to touch. Turned the thing off but they were cooked.
If somebody steals your card and PIN, they can easily drain pretty much as much as they want from your bank account, up to whatever your bank's daily withdrawal limit is
In my (Norwegian) bank, I can define my own withdrawal limits on my cards.
I also have a separate bank account with no cards "attached" where I keep most of my monthly spending cash. If I need more available on my card I just send an SMS with the amount I want to transfer from my "safe" account to my "card" account. Takes about 3 seconds to get the money there and only costs me a few cents.
So, I'm never worried about drained accounts.
That said, my bank will cover any fraud unless they have strong reason to believe I've been negligent protecting the PIN code.
In these 10 years since launch, they could have precomputed every possible picture, hash them, and then the probe could have simply sent the hashes instead of the full size pictures.
Just for fun, let's see what it would take for them to pull this off. The LORRI image sensor is 1024x1024 pixels with 12 bits per pixel.
So the number of distinct images divided by the timespan available gives 2^(12*20) / (10 years) = about 5.6 * 10^63 hashes per second.
Let's say you had a CPU capable of computing one such image hash per nanosecond (very optimistic), you'd need 2^(12*20) / (10 years) / (1 nanosecond) = about 5.6 * 10^54 CPUs to pull this off.
For comparison that's an order of magnitude or so more than the number of nucleons in our earth.
If those CPUs consumed 50W of power computing these hashes (again very optimistic), the entire project would consume 2^(12*20) / (1 nanosecond) * (50 watt) = 8.8 * 10^64 joules.
For reference that's two orders of magnitude more than the total mass-energy (including dark matter) of the Virgo supercluster, the supercluster which contains our Milky Way galaxy.
I've always seen IEEE 754 floats give NaN for division by zero, which is fine by me.
You get +/- infinity if a non-zero number is divided by zero, with the sign taken from the denominator. You only get NaN (not a number) if you divide zero by zero.
But I don't understand why people want to block regular banner ads. Coming up with content then hosting it on a website isn't free.
Because of drive-by downloads.
Last year the ad network of a non-trivial Norwegian site was hacked, and they started serving malware which targeted Java. If the user hadn't updated Java fairly recently, they'd get infected without any user interaction.
The malware was designed specifically to target the largest bank in Norway. This bank required Java for their login procedure (they no longer do, took them long enough).
So, if the user visited this site with a vulnerable Java runtime, and then logged in to this bank later to pay some bills, the malware would send the money elsewhere.
Since the malware was running on the local machine, it could bypass the two-factor authentication (password+token) required when transferring money.
Just look at paintings, in comparison van Gogh was a pixel artist while Rembrandt made proper high-definition 3D, yet both have made works that are considered great art.
When people complain about pixelation it's because nearly everyone cares about what looks good and not about good art.
The real mystery though is how the universe could be very nearly flat (without being exactly flat). Such "fine tuning" is clear evidence we're missing something quite fundamental.
The observable universe has to be sufficiently big for a planet like us to form, so that puts a lower bound on it.
But if the size of the whole universe really is random, then it seems likely that it's far larger than the observable, no?
Or are there any theoretical upper bounds I'm not familiar with?
What I remember is that it featured a rather eye-watering construction of two overlapping switch statements (?) which was syntactically legal, but perhaps shouldn't have been
Because it's not big enough to number all our hosts? I can reach the hosts that have v4 over v4, but not the ones that don't.
You said it wasn't a big issue that you cannot contact v4 from a v6 address, because one can simply use v4 to connect to v4. Yet you also say we need v6 because we don't have enough v4 left.
Oh, did you mean "NAT as it existed before we ran out of IP addresses"? Well, that's why we need IPv6, now when we are talking about NAT, it includes carrier-grade NAT.
If you're behind a carrier grade NAT then fiddling with your own router config won't help much will it. That's the part I quoted and objected to.
Slashdot Mirror
There's one guy working on a project called Atom/Atomic/Atome or something, which is basically your app compiled in to an OS container, instead of being built on top of an OS container, but still responding similar to a docker container.
Maybe several people doing the same, but this is one such project: http://www.includeos.org/
Simply add one include in you C++ project and compile it into a VM image.
As mentioned in the article (I know...), the base for Neon will be Ubuntu LTS. So everything except KDE will be nice and stable, unlike with a rolling distro like Netrunner.
Ãf¦ÃfÃfÂ¥
Gah forgot slashdot is in the technological stoneage... The norwegian alphabet contains three letters not found in the english alphabet.
Now the best scenario I can come up with is that you might get apps intended for use in a particular locale outside Five Eyes that are available only in the official or majority language of that locale.
No idea what the Five Eyes are, but my situation is this: I live in Norway, I cannot stand having anything but english as the display language in my OS, but I want to be able to type my name, emails, documents etc using norwegian keyboard layout as the norwegian alphabet contains æÃÃ¥.
Have you reported it to the developer of each affected app? Or does it affect all apps system-wide?
It affects all the _apps_ that are installed by default, so I'll say it's Microsoft. Note that the desktop, and all desktop-related programs, correctly respects my language choice.
And I'm not alone, social.microsoft.com has several posts on the issue from other people.
Would you prefer that an app be made entirely unavailable for download if its (often small) developer has not yet translated the app to your "forced display language"?
My forced display language is English. And why is using my keyboard layout language somehow a better choice than the display language I've explicitly told it I want?
Ever heard of SELECT .. FOR UPDATE?
How do you prevent two connections from inserting the same key at the same time using that?
It's superior to 7, 8, and 8.1 in every way that I can think of. Why do people not want a FREE upgrade?
I've tried upgrading 3 times after Windows 10 was launched, and I reverted within a day each time.
For one it BSODs every 20 minutes, even when just sitting there. For comparison, I've only had 2-3 BSODs with Windows 8.1 since I installed it shortly after launch, all game related.
Secondly "apps" are a PITA because they do not follow my forced display language.
A clean install should fix the BSOD issue, but until they sort out the apps I won't be touching Windows 10.
Most databases allow you to do transactions (BEGIN TRAN/COMMIT TRAN) that force it to be atomic.
Sure, but in that case you might get errors when you try to commit if another connection has changed the value in the meantime, and you'll have to retry the whole thing. That's what UPSERT avoids.
Is this summary saying that, prior to now, you could not update data that was in the database?
No. In some cases you want to use a table like a key-value storage. If the key does not exist you need to insert a row with that key. Otherwise you'll want to keep only one row with that key, and just update the value.
One can always do a "if key exists then update else insert", but the problem is that this is not atomic, because the "key exists" check is a separate statement from the insert or update statement. This can lead to issues if you have multiple connections accessing the same keys at the same time.
The UPSERT allows you to do this as one atomic operation.
Yes, that is a dilemma.
When I went to uni the first course in statistics required one book at around $200. However the profs had agreed to use the same book in the three follow-up statistics courses as well. So if you were going for a bachelor in statistics, you just bought that one book for all your statistics courses.
Similar story in physics, at least for the first two-three courses. The math courses were a bit more fractured for some reason, though in two of the courses the textbook was written by the prof and provided free as PDF. I got them printed and bound at the uni press for like $15 each.
However my book expenses was nothing compared to what they had to put up with in other areas, such as psychology or pedagogy. They had to buy $200 books just for a chapter or two.
My personal opinion is that life is really, really, really, REALLY rare.
Another factor is that high intelligence isn't necessarily a beneficial thing in terms of evolution. Cockroaches and rats will probably outlast the human species. Bacteria, algae and similar most certainly will.
If you can't provide a truly unlimited service, don't advertise it
They should just adopt a kind of Gabriel Horn's solution to the problem.
Yes you can upload and download as much as you want 24/7, hence unlimited access, but your bandwidth will be reduced in steps based on transferred data, so that in the end you'd only have a few kbps.
I just sent it back to the vendor (under warranty) because the latest OS update caused the speakers to melt.
While browsing the net I noticed the smell of burnt plastic, and quickly noticed the edges (where the speakers are) was too hot to touch. Turned the thing off but they were cooked.
Then I found I wasn't alone... https://forums.geforce.com/default/topic/834884/shield-tablet/speakers-damaged-after-3-0-update/
Apparently somehow the speakers got fed DC current while doing nothing in particular. Impressive if you ask me.
If somebody steals your card and PIN, they can easily drain pretty much as much as they want from your bank account, up to whatever your bank's daily withdrawal limit is
In my (Norwegian) bank, I can define my own withdrawal limits on my cards.
I also have a separate bank account with no cards "attached" where I keep most of my monthly spending cash. If I need more available on my card I just send an SMS with the amount I want to transfer from my "safe" account to my "card" account. Takes about 3 seconds to get the money there and only costs me a few cents.
So, I'm never worried about drained accounts.
That said, my bank will cover any fraud unless they have strong reason to believe I've been negligent protecting the PIN code.
In these 10 years since launch, they could have precomputed every possible picture, hash them, and then the probe could have simply sent the hashes instead of the full size pictures.
Just for fun, let's see what it would take for them to pull this off. The LORRI image sensor is 1024x1024 pixels with 12 bits per pixel.
So the number of distinct images divided by the timespan available gives 2^(12*20) / (10 years) = about 5.6 * 10^63 hashes per second.
Let's say you had a CPU capable of computing one such image hash per nanosecond (very optimistic), you'd need 2^(12*20) / (10 years) / (1 nanosecond) = about 5.6 * 10^54 CPUs to pull this off.
For comparison that's an order of magnitude or so more than the number of nucleons in our earth.
If those CPUs consumed 50W of power computing these hashes (again very optimistic), the entire project would consume 2^(12*20) / (1 nanosecond) * (50 watt) = 8.8 * 10^64 joules.
For reference that's two orders of magnitude more than the total mass-energy (including dark matter) of the Virgo supercluster, the supercluster which contains our Milky Way galaxy.
Unless I messed up the calculations that is...
with the sign taken from the denominator
Should of course be "with the sign taken from the numerator".
I'll go to bed now...
I've always seen IEEE 754 floats give NaN for division by zero, which is fine by me.
You get +/- infinity if a non-zero number is divided by zero, with the sign taken from the denominator. You only get NaN (not a number) if you divide zero by zero.
But I don't understand why people want to block regular banner ads. Coming up with content then hosting it on a website isn't free.
Because of drive-by downloads.
Last year the ad network of a non-trivial Norwegian site was hacked, and they started serving malware which targeted Java. If the user hadn't updated Java fairly recently, they'd get infected without any user interaction.
The malware was designed specifically to target the largest bank in Norway. This bank required Java for their login procedure (they no longer do, took them long enough).
So, if the user visited this site with a vulnerable Java runtime, and then logged in to this bank later to pay some bills, the malware would send the money elsewhere.
Since the malware was running on the local machine, it could bypass the two-factor authentication (password+token) required when transferring money.
Serious question: what's with the medium.com hate? I really don't get it.
The guy is confused. Art != looks good.
Just look at paintings, in comparison van Gogh was a pixel artist while Rembrandt made proper high-definition 3D, yet both have made works that are considered great art.
When people complain about pixelation it's because nearly everyone cares about what looks good and not about good art.
The real mystery though is how the universe could be very nearly flat (without being exactly flat). Such "fine tuning" is clear evidence we're missing something quite fundamental.
The observable universe has to be sufficiently big for a planet like us to form, so that puts a lower bound on it.
But if the size of the whole universe really is random, then it seems likely that it's far larger than the observable, no?
Or are there any theoretical upper bounds I'm not familiar with?
What I remember is that it featured a rather eye-watering construction of two overlapping switch statements (?) which was syntactically legal, but perhaps shouldn't have been
Are you think of Duff's Device? It overlaps a switch with a do-while loop: http://en.wikipedia.org/wiki/Duff's_device
Because it's not big enough to number all our hosts?
I can reach the hosts that have v4 over v4, but not the ones that don't.
You said it wasn't a big issue that you cannot contact v4 from a v6 address, because one can simply use v4 to connect to v4. Yet you also say we need v6 because we don't have enough v4 left.
See the issue now?
Why? Because I also run v4 everywhere and use that to reach v4 hosts.
So why are we even bothering with v6 again when all we need is just to keep our v4?
Oh, did you mean "NAT as it existed before we ran out of IP addresses"? Well, that's why we need IPv6, now when we are talking about NAT, it includes carrier-grade NAT.
If you're behind a carrier grade NAT then fiddling with your own router config won't help much will it. That's the part I quoted and objected to.