First of all, its great to see google being devoted to open source this much. Not many companies do release so much code. That's really cool! I guess this also helped convincing Microsoft to abandon their "Open source == cancer" strategy, proving that you can be the leader of a market *and* release open source software.
Most apps will use at least two of analytics, ads, and GCM, probably almost all will use at least one of those three.
Also its good for Google being able to have at least some influence on the manufacturers.
But still, its very hard for competing ROM and app market manufacturers to emerge if they have to replace all of google's cloud services, in order to have the faintest chance at competing against the world market leader.
One good way to enable competition would be to require google to give access for their online APIs without lock-in (but the same java client side API), for a fair price that allows google to make money with the service and allows competition to offer products using that service. Similar has been done with POTS providers in the EU and the US in the past.
The EU has a parliament by the way. And it has even a mechanism for direct democratic voting. But I admit that it could be more legitimized by the peoples, yes.
There are over 1,800 apps on the FDroid repository alone that work just fine without Google's apps.
And there are even more apps that are open source but not on F-Droid because they use one Google API or another. Think of Signal, for example. Also, often developers add a feature to an app where they integrate Google maps or something, and bam the new version of the app can't be included into F-Droid anymore. This puts lots of stress onto the app developers, as they now have to develop for two stores, not just one, and very often they are annoyed by the F-Droid crowd demanding to remove Google API usage while not being able to bring up alternatives. Signal is a very good example for this as well.
Its a wonder that Telegram and Osmand (probably the two most useful F-Droid apps) still remain on the store.
distros like Cyanogen offer FOSS replacements for them
Cyanogen OS is not CyanogenMod. It is developed by the same people, but CyanogenMod is open source, while Cyanogen OS is not. And CyanogenMod can only exist in an area of semi-legality, tolerated by google because it is so small. Most CM users just want it to work ^ tm and install gapps illegally from dubious sources onto the device. CM itself provides no FOSS replacements for the google APIs, only some open source apps, but nothing more. Don't know about whether Cyanogen OS provides replacements.
Low predatory pricing is *the* tool against destroying competition that's smaller than you. You are the big one, you have the capital to keep the price low (even if it's not profitable for you!), for as long as you want. You just wait until your small competition either gets no customers because their prices aren't competing with your's or they go bankrupt because they used up their much smaller capital much faster than you.
Then, after all small competition is ruled out, you can rise prices again and make much more money than you spent on the aggressive predatory pricing.
Google's lock in system bases not on the google-owned apps (they are just a few, and yes they are very much used by users, but I guess people can come up with an alternative). The main reason to be locked in to Google is their proprietary APIs they offer to app developers. You can't simply take an apk and publish it on an alternative market, if there are no gapps installed on the device, most of the apps won't work.
So even if a competitor managed to replace all the gapps that are exposed to the user (maps, search, etc), they still would have a very hard time at building a competing app store. Most of the app developers don't want to port the app if the user count is low and nobody would install it if they couldn't install all the apps.
Its the same issue linux is facing. People don't care about operating systems. They want to install an application, and if it doesn't work, its not the fault of the application developers, its the fault of the operating system (at least for them).
You know if the brits leave the EU, that's a good thing. They were the major stopgap hindering to get real work done against the banks in the late 00's banking crisis.
Its also market, maps, calendar, etc. Gapps include so many services, many of them made in a way that the app developer has to choose between google proprietary and competitors. If some startup proclaims to compete with google, they usually get bought up, and don't continue to offer their services.
Fractured little fiefdoms. This confuses me. It seems like a significant loss.
What's a loss for the community is a win for the founders and the investors. A big win. Silicon valley wouldn't be as successful as it is if they were opening their doors to competition.
What was wrong with that model, that we had to run as fast as possible away from it?
Its easier to build a business model around an app that's closed down and proprietary as much as possible.
It says that the "hydrogen envelope" (probably most of the star), would be required to be ejected before the merger, otherwise it would choke the whole gamma ray burst already, meaning that we wouldn't have detected detect a gamma ray burst the first place if there was a hydrogen envelope. Probably something would have happened what happens right now in the sun, the "envelope" would convert the gamma ray radiation to lower frequency radiation, like visible light.
So you might be right with most of the star being already gone when the event happened.
The paper also tries to find an explanation on the cause for the delay:
"For a progenitor star in the mass range [of 100 to 1000 solar masses], most of the observed 0.4 s delay can be accounted for by the neutrino cooling timescale or by the extra time it takes the GRB to jet to cross the star relative to GWs for a jet Lorentz factor of [gamma around 4-7]".
I admit I can't really understand what they said here. The "neutrino cooling" is most likely cooling by neutrino radiation. But all references to neutrino cooling I could find in the paper indicate that this cooling is happening before the two black holes form, as a requirement for their formation, and not as an event between the BH merger and the GRB emission. Also I'm not sure what this has to do with a jet. I mean jets are these matter streams that leave an object at relativistic speeds (therefore also the Lorentz factor mention). But it is very unlikely that for the first GW we detect (GWs get emitted into all directions) we also detect a GRB emission which only gets to us because of a jet being targeted at us (something that affects only a tiny fraction of the star's sky), isn't it?
They say that the two black holes formed inside a giant star, and collapsed. This created both the gamma ray burst and the gravitational wave. Then the light needed to travel through the star's matter. While the gravitational wave can travel with the speed of light in vacuum, the light requires more time, that's where the.4 seconds delay comes from.
They say also that in order for there being a gamma ray burst, there needs to be matter close to the colliding black holes. So this rules out a pair of black holes that orbited each other for a long time, because they would have cleared out the region.
There is certainly a lot more going on than described by Darwin.
Yes. There is also certainly a lot more going on than described by Einstein. There still is no theory that unifies quantum theory with relativity. That's what scientists do, they constantly improve the status quo, and amend and improve the theories of their predecessors. Sometimes they propose conflicting theories. Then the scientific community looks for ways to resolve the conflict in a scientific manner, as in by experimentation or similar.
No aspect of the theory of evolution predicted epigentics, which infers that the mechanism isn't well understood.
Did Darwin have the genetic methods of our time? No, he hasn't.
But the theory of evolution is obviously True, so why bother continuing to study it at all?
I'm okay that scientists study it and question it, like every scientific theory. If the theory is true, then it gains more strength and more detail by every failed attempt at proving it wrong. If it is false, it is even better that it gets proven wrong.
It's very possible that several traits never rise to that degree of prominence and are simply just there, which makes description of selection by environmental factors ambiguous.
Its a sum of traits that get selected after, not just one.
but a direct one to one causation isn't there.
True. This is a known physical phenomenon. While the genetically "best fit" member of the species can have bad luck and get eaten by a tiger, and some other "badly fit" member of the species somehow can have luck and is able to procreate, in the statistics, the more fit side wins, and the less fit side loses.
Its like playing at a machine in a casino: while the bank has sometimes losses and has to pay even sometimes very large prizes, they still win in the long run, as most of the people win nothing or less than they invested.
Or take nuclear decay. If you look at an individual unstable atom, you never know when it decays, whether its in the next milisecond or you will have to wait 10 million years. You just know that if you have a certain amount of unstable atoms, half of them has decayed after a specific time that's unique to the atom's element.
A certain segment of the population also condemns the scientific theory of evolution and that earth is older than five thousand years.
I mean criticising Hawking's theories about black holes makes total sense, as they never were confirmed in reality, hawking radiation is in fact so small for usual black holes that its lower than the CMB: the black hole gains more mass through CMB radiation that enters it than it loses to hawking radiation. Wake me up when a black hole is found that's so small that its Hawking radiation is actually measurable.
But criticising Einstein whose theories have been confirmed multiple times? I don't know. Yes, maybe he is wrong in one point or another, the way newton was wrong when Einstein amended his theories, but his description is pretty accurate for what we can observe.
Why has this been modded down? Climate change is expected to force billions of humans to leave their home in the next century. The answers for the refugee problem this creates are all not pleasing.
You forgot option 4: Not letting in refugees into your country. This means machine guns. And option 5: Letting in refugees into your country, see that they outnumber your population by a factor of ten or more. This means civil war.
I mean when Einstein's theory of relativity was in question, did they take a survey and consider the matter settled?
That was about a subject of not large importance back then. Einstein didn't say that all industrialized and industrializing countries should invest trillions to re-design their energy processing systems to not rely on climate harming material. He did not say that countries whose economy bases on export of oil and gas should stop extracting in order to save the world. All he said was some theory about stuff whose influence you need a telescope or have to build satellites with atomic clocks in order to tell a difference. That was pretty far away from reality on earth back then.
It doesn't matter how great the tech is, there isn't industry or developer support and interest.
Saying there is no interest at all is wrong. Rust is "most loved" language in stack overflow's developer survey 2016, before Swift. Yes, Swift's developer base is much larger than the base of Rust, probably because it is used for iOS app development, that's a big market. And it may become the language to write apps in for both Android and iOS. Swift has a successful future in front of it. That doesn't mean Rust has none.
Rust has areas it excels at, for example it does multithreading much better than go, a language previously known as "the" language to do mulithreading.
Also, Swift's memory model is limited to one model only: reference counting. In Rust you have much greater flexibility, and can use whatever model you want. For example, AFAIK, in swift the memory model is limited to a thread safe one. This means there are always locks invoked (or perhaps a lock-less strategy on platforms that support it). Rust has this too, but it also has a non-thread safe counterpart, which is of course much faster.
I could continue to list the advantages. Rust has deserved every bit of its hype.
Swift can replace Objective-C and Java as the language to program apps in. That's a giant market. But there is still much room for Rust to expand to.
Well I hope that open source packages won't switch to the new snap system, as of course it adds duplication, and now many application providers have to update one of their libraries only because of some badlock vuln or something. Some app store owners try to counter this by threatening app owners to take down their apps if they don't update the libraries. But this only gets the biggest libraries and those with most light shined on them, the small library might never get updated.
Yes, poettering has seen correctly that the duplication can be fought with fs level dedup like btrfs is doing it, but that still doesn't solve the update problem.
Where the snap system shines at is closed source applications and open source applications which both get shipped outside of the distro's packaging system: if adopted by all distros, you can ship cross-distro binaries without having to bother about some distro's settings for their libraries.
And closed source applications is one of the areas the linux desktop sucks at. There are almost no linux versions of many more or less popular closed sourced applications.
Slashdot Mirror
First of all, its great to see google being devoted to open source this much. Not many companies do release so much code. That's really cool! I guess this also helped convincing Microsoft to abandon their "Open source == cancer" strategy, proving that you can be the leader of a market *and* release open source software.
Most apps will use at least two of analytics, ads, and GCM, probably almost all will use at least one of those three.
Also its good for Google being able to have at least some influence on the manufacturers.
But still, its very hard for competing ROM and app market manufacturers to emerge if they have to replace all of google's cloud services, in order to have the faintest chance at competing against the world market leader.
One good way to enable competition would be to require google to give access for their online APIs without lock-in (but the same java client side API), for a fair price that allows google to make money with the service and allows competition to offer products using that service.
Similar has been done with POTS providers in the EU and the US in the past.
As I said, most of the apps won't work if there is no gapps installed on the device. Some of them will continue to work, I agree with that.
The EU has a parliament by the way. And it has even a mechanism for direct democratic voting. But I admit that it could be more legitimized by the peoples, yes.
There are over 1,800 apps on the FDroid repository alone that work just fine without Google's apps.
And there are even more apps that are open source but not on F-Droid because they use one Google API or another. Think of Signal, for example. Also, often developers add a feature to an app where they integrate Google maps or something, and bam the new version of the app can't be included into F-Droid anymore. This puts lots of stress onto the app developers, as they now have to develop for two stores, not just one, and very often they are annoyed by the F-Droid crowd demanding to remove Google API usage while not being able to bring up alternatives. Signal is a very good example for this as well.
Its a wonder that Telegram and Osmand (probably the two most useful F-Droid apps) still remain on the store.
distros like Cyanogen offer FOSS replacements for them
Cyanogen OS is not CyanogenMod. It is developed by the same people, but CyanogenMod is open source, while Cyanogen OS is not. And CyanogenMod can only exist in an area of semi-legality, tolerated by google because it is so small. Most CM users just want it to work ^ tm and install gapps illegally from dubious sources onto the device. CM itself provides no FOSS replacements for the google APIs, only some open source apps, but nothing more. Don't know about whether Cyanogen OS provides replacements.
claiming that low prices were predatory
Low predatory pricing is *the* tool against destroying competition that's smaller than you. You are the big one, you have the capital to keep the price low (even if it's not profitable for you!), for as long as you want. You just wait until your small competition either gets no customers because their prices aren't competing with your's or they go bankrupt because they used up their much smaller capital much faster than you.
Then, after all small competition is ruled out, you can rise prices again and make much more money than you spent on the aggressive predatory pricing.
Google's lock in system bases not on the google-owned apps (they are just a few, and yes they are very much used by users, but I guess people can come up with an alternative). The main reason to be locked in to Google is their proprietary APIs they offer to app developers. You can't simply take an apk and publish it on an alternative market, if there are no gapps installed on the device, most of the apps won't work.
So even if a competitor managed to replace all the gapps that are exposed to the user (maps, search, etc), they still would have a very hard time at building a competing app store. Most of the app developers don't want to port the app if the user count is low and nobody would install it if they couldn't install all the apps.
Its the same issue linux is facing. People don't care about operating systems. They want to install an application, and if it doesn't work, its not the fault of the application developers, its the fault of the operating system (at least for them).
You know if the brits leave the EU, that's a good thing. They were the major stopgap hindering to get real work done against the banks in the late 00's banking crisis.
Its also market, maps, calendar, etc. Gapps include so many services, many of them made in a way that the app developer has to choose between google proprietary and competitors. If some startup proclaims to compete with google, they usually get bought up, and don't continue to offer their services.
If there were thought reading devices (and I'm sure there will be one day), they'd want them to be used as well.
Fractured little fiefdoms. This confuses me. It seems like a significant loss.
What's a loss for the community is a win for the founders and the investors. A big win. Silicon valley wouldn't be as successful as it is if they were opening their doors to competition.
What was wrong with that model, that we had to run as fast as possible away from it?
Its easier to build a business model around an app that's closed down and proprietary as much as possible.
Disclaimer: I am no physicist, I just read the papers and quote them :)
There is this paper which does some speculation for what could have been the cause: http://arxiv.org/pdf/1602.0473...
It says that the "hydrogen envelope" (probably most of the star), would be required to be ejected before the merger, otherwise it would choke the whole gamma ray burst already, meaning that we wouldn't have detected detect a gamma ray burst the first place if there was a hydrogen envelope. Probably something would have happened what happens right now in the sun, the "envelope" would convert the gamma ray radiation to lower frequency radiation, like visible light.
So you might be right with most of the star being already gone when the event happened.
The paper also tries to find an explanation on the cause for the delay:
"For a progenitor star in the mass range [of 100 to 1000 solar masses], most of the observed 0.4 s delay can be accounted for by the neutrino cooling timescale or by the extra time it takes the GRB to jet to cross the star relative to GWs for a jet Lorentz factor of [gamma around 4-7]".
I admit I can't really understand what they said here. The "neutrino cooling" is most likely cooling by neutrino radiation.
But all references to neutrino cooling I could find in the paper indicate that this cooling is happening before the two black holes form, as a requirement for their formation, and not as an event between the BH merger and the GRB emission.
Also I'm not sure what this has to do with a jet. I mean jets are these matter streams that leave an object at relativistic speeds (therefore also the Lorentz factor mention). But it is very unlikely that for the first GW we detect (GWs get emitted into all directions) we also detect a GRB emission which only gets to us because of a jet being targeted at us (something that affects only a tiny fraction of the star's sky), isn't it?
This page has a possible explanation: http://www.universetoday.com/1...
They say that the two black holes formed inside a giant star, and collapsed. This created both the gamma ray burst and the gravitational wave. Then the light needed to travel through the star's matter. While the gravitational wave can travel with the speed of light in vacuum, the light requires more time, that's where the .4 seconds delay comes from.
They say also that in order for there being a gamma ray burst, there needs to be matter close to the colliding black holes. So this rules out a pair of black holes that orbited each other for a long time, because they would have cleared out the region.
There is certainly a lot more going on than described by Darwin.
Yes. There is also certainly a lot more going on than described by Einstein. There still is no theory that unifies quantum theory with relativity. That's what scientists do, they constantly improve the status quo, and amend and improve the theories of their predecessors. Sometimes they propose conflicting theories. Then the scientific community looks for ways to resolve the conflict in a scientific manner, as in by experimentation or similar.
No aspect of the theory of evolution predicted epigentics, which infers that the mechanism isn't well understood.
Did Darwin have the genetic methods of our time? No, he hasn't.
But the theory of evolution is obviously True, so why bother continuing to study it at all?
I'm okay that scientists study it and question it, like every scientific theory. If the theory is true, then it gains more strength and more detail by every failed attempt at proving it wrong. If it is false, it is even better that it gets proven wrong.
It's very possible that several traits never rise to that degree of prominence and are simply just there, which makes description of selection by environmental factors ambiguous.
Its a sum of traits that get selected after, not just one.
but a direct one to one causation isn't there.
True. This is a known physical phenomenon. While the genetically "best fit" member of the species can have bad luck and get eaten by a tiger, and some other "badly fit" member of the species somehow can have luck and is able to procreate, in the statistics, the more fit side wins, and the less fit side loses.
Its like playing at a machine in a casino: while the bank has sometimes losses and has to pay even sometimes very large prizes, they still win in the long run, as most of the people win nothing or less than they invested.
Or take nuclear decay. If you look at an individual unstable atom, you never know when it decays, whether its in the next milisecond or you will have to wait 10 million years. You just know that if you have a certain amount of unstable atoms, half of them has decayed after a specific time that's unique to the atom's element.
You can't get them upset anymore. After all, they have vista.
Dude, this is already going to happen with all this "smart cars" revolution.
I didn't know that Myanmar or Liberia have walked on the moon. Very interesting.
A certain segment of the population also condemns the scientific theory of evolution and that earth is older than five thousand years.
I mean criticising Hawking's theories about black holes makes total sense, as they never were confirmed in reality, hawking radiation is in fact so small for usual black holes that its lower than the CMB: the black hole gains more mass through CMB radiation that enters it than it loses to hawking radiation. Wake me up when a black hole is found that's so small that its Hawking radiation is actually measurable.
But criticising Einstein whose theories have been confirmed multiple times? I don't know. Yes, maybe he is wrong in one point or another, the way newton was wrong when Einstein amended his theories, but his description is pretty accurate for what we can observe.
Why has this been modded down? Climate change is expected to force billions of humans to leave their home in the next century. The answers for the refugee problem this creates are all not pleasing.
You forgot option 4: Not letting in refugees into your country. This means machine guns.
And option 5: Letting in refugees into your country, see that they outnumber your population by a factor of ten or more. This means civil war.
I mean when Einstein's theory of relativity was in question, did they take a survey and consider the matter settled?
That was about a subject of not large importance back then. Einstein didn't say that all industrialized and industrializing countries should invest trillions to re-design their energy processing systems to not rely on climate harming material. He did not say that countries whose economy bases on export of oil and gas should stop extracting in order to save the world. All he said was some theory about stuff whose influence you need a telescope or have to build satellites with atomic clocks in order to tell a difference. That was pretty far away from reality on earth back then.
Oh so you say tax payers pay for the cleaning up? Great how the public hand funds this. What a wonderful job the politicians did.
Well perhaps all the assets are auctioned and you can buy a cheap coal mine or two, then you have more than enough coal for your stoves.
Yeah these are just stupid kids who feel too cool to just paint graffiti on a wall, they need to do something "leet".
It doesn't matter how great the tech is, there isn't industry or developer support and interest.
Saying there is no interest at all is wrong. Rust is "most loved" language in stack overflow's developer survey 2016, before Swift.
Yes, Swift's developer base is much larger than the base of Rust, probably because it is used for iOS app development, that's a big market. And it may become the language to write apps in for both Android and iOS. Swift has a successful future in front of it. That doesn't mean Rust has none.
Rust has areas it excels at, for example it does multithreading much better than go, a language previously known as "the" language to do mulithreading.
Also, Swift's memory model is limited to one model only: reference counting. In Rust you have much greater flexibility, and can use whatever model you want.
For example, AFAIK, in swift the memory model is limited to a thread safe one. This means there are always locks invoked (or perhaps a lock-less strategy on platforms that support it). Rust has this too, but it also has a non-thread safe counterpart, which is of course much faster.
I could continue to list the advantages. Rust has deserved every bit of its hype.
Swift can replace Objective-C and Java as the language to program apps in. That's a giant market. But there is still much room for Rust to expand to.
Well I hope that open source packages won't switch to the new snap system, as of course it adds duplication, and now many application providers have to update one of their libraries only because of some badlock vuln or something. Some app store owners try to counter this by threatening app owners to take down their apps if they don't update the libraries. But this only gets the biggest libraries and those with most light shined on them, the small library might never get updated.
Yes, poettering has seen correctly that the duplication can be fought with fs level dedup like btrfs is doing it, but that still doesn't solve the update problem.
Where the snap system shines at is closed source applications and open source applications which both get shipped outside of the distro's packaging system: if adopted by all distros, you can ship cross-distro binaries without having to bother about some distro's settings for their libraries.
And closed source applications is one of the areas the linux desktop sucks at. There are almost no linux versions of many more or less popular closed sourced applications.