Unfortunately if you *do* have to fill in a tax return in the UK then they don't pre-fill anything. So rather than just declaring how much self-employed income I made, I have to go round chasing down a bunch of stuff to fill in the form.
What makes it appear "safe" in the developed world is the constant routine of aircraft maintenance and pilot training that keeps the accident rate very, very low. But in other countries, that isn't the case."
And yet major crashes are rare enough that when one happens, even to an airline from a third world shithole, it's world news. When *TWO* happen to the same nearly-new variant under similar circumstances that tends to point fingers towards a problem specific to that variant.
Not just one org, there are piles of orgs, especially in academia that have been on the net since before the eternal september and the IPv4 crunch and still run their networks in much the way they always have. Mostly open to the Internet with maybe some limited firewalling on ports particularly likely to be abused.
Also when ever you rent a server or VM or colocation slot from a hosting provider it comes with a public IP open to the Internet by default. I could easilly see someone fed up with overpriced cloud storage shoving an iScsi storage array into a colo rack and just hooking it up to the hosting provider's network (wide open to the Internet) without thinking things through properly.
As I understand it there are paper checklists but for the really critical scenarios there are also "memory items" that the pilots are expected to be able to execute from memory before they go looking for the paper checklists.
Routers, TVs, video recorders, CCTVs, and more will be non-functional in 2038. And there is nothing we can do to stop it.
We can't stop todays or yesterdays products from having problems in 2038.
But the sooner this is fixed in the main-line the sooner it will reach the production lines the less effected equipment there will be still in use in 2038.
y2038 is mostly an issue for 32-bit systems, not saying there won't be some issues on 64-bit systems where developers have stored times in a 32-bit types, but those issues should be relatively contained and manageable.
I presume the GP was thinking about sports where people compete as individuals. Some team sports do indeed have a broader base of well-paid players, though still miniscule compared to the total number of people who play the sport.
And application design is heavilly dependent on form-factor. Smartphone apps are optimised for a small screen with inaccurate pointing and a crappy soft-keyboard that makes the already small screen even smaller. Laptop/desktop apps are optimised for large screens with a proper keyboard and an accurate multi-button pointing device.
Companies are neither inherently good or inherently evil, they just do what they think will help their bottom line and/or strategic goals, and yes keeping up a good PR face can be part of that.
This fix probablly hurts google's competitors more than it hurts google. Google can probablly make a pretty damn good guess whether someone is in incognito mode without resorting to tricks (if a browser shows up with no google cookies it's a pretty good bet it's in incognito mode). Smaller sites will find it harder to guess.
Similarly with ad-blocking if google sets up the defaults such that they let google's ads through while blocking the more obviously obnoxious ads from competitors they reduce the risk that people will seek-out a third party ad-blocking soloution which may block more aggressively. The web giants nightmare is that the more obnoxious end of the internet ad-market drives the majority of Internet users to install an agressive ad-blocker.
Presumably for a song to be counted it has to actually be identified. I would guess that youtube is using the same or a similar content ID system for these charts as the one they use to redirect advertising royalties.
Most legit youtubers with large followings are careful to avoid anything that will hit a youtube content ID match because a content-id means at best losing the ad-revenue for the video and possibly strikes against the channel. For TV theme songs if the copyright holder of a TV show cares enough to put the theme song in a content-ID database I suspect they also care enough to remove pirate copies of their TV show from youtube.
You'll go through customs when you first arrive in a country, even if you have a connection to another city in that country.
Afaict this varies between countries and airports. Some places customs is handled at the final airport, some places customs is handed at the point of entry but without passengers normally having to collect their checked baggage. Some places are the way you describe with passengers forced to collect their baggage, drag it through customs and re-check it.
The reason they have become rare is because they are no longer needed.
The first part of the sale process is attracting the customers attention, then you have to help them decide what exactly to buy, and then finally you can make the transaction.
In the specialised print media world, before the rise of modern internet commerce the ad served both the purposes of attracting the customer's attention and helping them decided which product they wanted. Then the customer would make the actual transaction by mail of phone. As you say this took up a lot of space, I remember some ads that were a lot bigger than "one or two pages".
On the Internet companies have their own websites that are only a click away from the ad. So all the ad has to do is the first step, attracting the customers attention.
If "personalized ads" are for items that you've recently purchased, then whomever did the algorithm for "personalizing" those ads was a complete cretin.
Afaict the issue is that the advertising firms know what you have been looking for but not what you have actually bought.
So you start looking for a new vacum cleaner, the ad networks pick up on this from the pages you visit and the search keywords you use. They start showing you vacum cleaner ads because people who are searching for vacum cleaners have a higher probability of cliking through on vacum cleaner ads than generic ads.
At some point you go ahead and buy a vacum cleaner, but the ad networks don't know that. So the ads for vacum cleaners continue for a while until they have gathered enough information to belive you are no-longer interested.
Afaict plastics, are typically made from hydrocarbons that are not saturated with hydrogen. These hydrocarbons result when heavier alkanes are cracked to produce lighter alkanes.
So the availability of feedstocks for plastic production is likely to depend on the relative demand for heavy and light oil products. Gasoline is somewhere in the middle so it seems like reduced demand for gasoline could do either way in terms of availability of feedstocks for plastic products.
A prison systems job is or at least should be to reduce damage caused by criminals to regular society. There are three mechanisms through which it can do that.
1. Deterrent, most people don't want to be locked up, hopefully the threat of being locked up keeps most people in-line. 2. Rehabilitation. convincing prisoners that they have options other than a life of crime, that they have friends and family they would rather be with than in prison. 3. Removal, if the prisoner is locked up and communications are restricted then it's harder for them to commit crimes against anyone other than the other prisoners and the wardens.
A good prison system IMO balances the factors, not so cushy that it fails to be a deterrent, but not so isolating from society that prisoners become totally dysfunctional and feel they have no choice but to turn back to crime.
Afaict browser based java had become more of a liability than an asset. Relatively few people were still using it for legitimate applications but (like most browser plugins) it was a massive source of security drama, driving the perception that Java is insecure. I'm not surprised they decided to kill it.
The browser vendors have also been pushing to kill plugins for much the same reason. Afaict firefox and chrome still allow flash as a special case, but have blocked all other native-code plugins. I'm not sure what is going on over on the MS side.
As I understand it, Openjdk is the reference implementation, but it is only distributed in source code form.
Oracle has just changed the license on their binary JDK distributions to be far more restrictive of what can be done without paying.
So that basically leaves many windows users with the choice of building it themselves (which I understand is not exactly a pleasent process), paying oracle, or using binaries from some unknown third-party.
Linux users aren't really affected, they just get their openjdk builds from their distro.
Keeping a site where any random member of the public can create an account and upload videos alive requires walking a fine line. They have to respond sufficiently robustly to copyright infringements to avoid being sued out of existance and sufficiently robustly to advertiser complains to avoid the ad-revenue drying up. Unfortunately the economics of running a site where any random member of the public can watch and upload videos for free means that on average very little human attention can be spent on each user.
Yes there are a few users who attract enough eyeballs that real money is invovled and afaict at least some of those users do get special treatment from youtube but they are a tiny minority.
I think your suggestion of mass fraudulent copyright claims is far more likely to mean the end of youtube as we know it than to mean better protection for creators.
Radioactive half-lives are a measure of how "hot" (radioactive) a compound is per unit time.
Something of an oversimplification. It measures how long it takes for half the radioactive atoms to decay, but what comes out with each decay, whether the decay products are also radioactive and of course what proportion of the original material is actually radioactive atoms varies.
1 mole of 1hr half-life material will kill you much faster than 1 mole of 1 billion year material.
Yes.
But the one hour half-life material will rapidly decay, after a day over 99.9999% of it is gone. So while it's intially dangerous it is not in itself a long term problem (of course it's decay products might be).
Generally the worry from a radiological point of view (chemical toxicity is also a concern with heavy elements) are the isotopes with half-lives somewhere in the middle, short enough that the substance has noticable radioactivity, but long enough that the reactivity won't reduce significantly within a lifetime.
Nuking the enterprise cert is going to give Facebook some pain, making it harder to beta-test apps and harder to use iphones for internal applications, but it's likely to be manageable pain.
Nuking the facebook app from the appstore would likely to significant damage to both Apple and Facebook.
Finally, and this cannot be stressed enough: "Market share" is not important for open-source projects.
I disagreee for a couple of reasons.
Firstly while some opensource projects are run by hobbyists in their spare time and some are run by collaborations between a bunch of stakeholders. Others are largely paid for by a single entity based on revenues they somehow extract from the project. Mozilla falls into the latter category. They make most of their revenue by selling the top search engine slot in their browser (currently to google). If their Market share drops it will be much harder for them to sell that slot for a good price.
Shrinking an organisation is hard, it is hard to avoid losing your best people first as you try and shrink. It's hard to figure out what is really important and what is make-work.
Secondly market share translates to mind share. As much as people may like standards neither the standards themselves or any of their implementations are perfect. If Firefox's market share drops too low then web developers may decide it is not worth bothering to debug their sites on it at which point using firefox may become increasingly painful.
At least this noname Chinese factory managed to not screw up with first production batch
Sure they did, they made an unauthorised substitution of the Ethernet port (fitting a plain jack instead of a jack with integrated magnetics) that meant the first production batch had to be re-worked.
Github's advantage is it is the largest and best known of the git collaboration sites, so it is where you and your potential collaborators are most likely to already be set-up rather than creating yet more accounts on yet more sites.
The main downside of github's new free repos seems to be the collaborator limit, gitlab offers unlimited collaborators. Bitbucket seems to count any user you give access to a private repo as part of your "team" for the purpose of account limits (free accounts are limited to 5 "team members").
All these companies have to strike a balance, giving stuff away for free is how they attract new users, but if they give too much away for free then those users will never upgrade to become paying customers. Getting the balance right is tricky, especially as charging for functionality that used to be free is likely to get you conversions in the short term, but also get you a bunch of bad PR, so any changes to what is included in the free tier need to be very carefully thought through.
Slashdot Mirror
Unfortunately if you *do* have to fill in a tax return in the UK then they don't pre-fill anything. So rather than just declaring how much self-employed income I made, I have to go round chasing down a bunch of stuff to fill in the form.
What makes it appear "safe" in the developed world is the constant routine of aircraft maintenance and pilot training that keeps the accident rate very, very low. But in other countries, that isn't the case."
And yet major crashes are rare enough that when one happens, even to an airline from a third world shithole, it's world news. When *TWO* happen to the same nearly-new variant under similar circumstances that tends to point fingers towards a problem specific to that variant.
Not just one org, there are piles of orgs, especially in academia that have been on the net since before the eternal september and the IPv4 crunch and still run their networks in much the way they always have. Mostly open to the Internet with maybe some limited firewalling on ports particularly likely to be abused.
Also when ever you rent a server or VM or colocation slot from a hosting provider it comes with a public IP open to the Internet by default. I could easilly see someone fed up with overpriced cloud storage shoving an iScsi storage array into a colo rack and just hooking it up to the hosting provider's network (wide open to the Internet) without thinking things through properly.
As I understand it there are paper checklists but for the really critical scenarios there are also "memory items" that the pilots are expected to be able to execute from memory before they go looking for the paper checklists.
You can search by submitter. For example:
https://bugs.debian.org/cgi-bi...
The op set up his slashdot profile to hide his email address though.
Routers, TVs, video recorders, CCTVs, and more will be non-functional in 2038. And there is nothing we can do to stop it.
We can't stop todays or yesterdays products from having problems in 2038.
But the sooner this is fixed in the main-line the sooner it will reach the production lines the less effected equipment there will be still in use in 2038.
64-bit linux *does* have a 64-bit time_t.
y2038 is mostly an issue for 32-bit systems, not saying there won't be some issues on 64-bit systems where developers have stored times in a 32-bit types, but those issues should be relatively contained and manageable.
I presume the GP was thinking about sports where people compete as individuals. Some team sports do indeed have a broader base of well-paid players, though still miniscule compared to the total number of people who play the sport.
And application design is heavilly dependent on form-factor. Smartphone apps are optimised for a small screen with inaccurate pointing and a crappy soft-keyboard that makes the already small screen even smaller. Laptop/desktop apps are optimised for large screens with a proper keyboard and an accurate multi-button pointing device.
Companies are neither inherently good or inherently evil, they just do what they think will help their bottom line and/or strategic goals, and yes keeping up a good PR face can be part of that.
This fix probablly hurts google's competitors more than it hurts google. Google can probablly make a pretty damn good guess whether someone is in incognito mode without resorting to tricks (if a browser shows up with no google cookies it's a pretty good bet it's in incognito mode). Smaller sites will find it harder to guess.
Similarly with ad-blocking if google sets up the defaults such that they let google's ads through while blocking the more obviously obnoxious ads from competitors they reduce the risk that people will seek-out a third party ad-blocking soloution which may block more aggressively. The web giants nightmare is that the more obnoxious end of the internet ad-market drives the majority of Internet users to install an agressive ad-blocker.
Presumably for a song to be counted it has to actually be identified. I would guess that youtube is using the same or a similar content ID system for these charts as the one they use to redirect advertising royalties.
Most legit youtubers with large followings are careful to avoid anything that will hit a youtube content ID match because a content-id means at best losing the ad-revenue for the video and possibly strikes against the channel. For TV theme songs if the copyright holder of a TV show cares enough to put the theme song in a content-ID database I suspect they also care enough to remove pirate copies of their TV show from youtube.
You'll go through customs when you first arrive in a country, even if you have a connection to another city in that country.
Afaict this varies between countries and airports. Some places customs is handled at the final airport, some places customs is handed at the point of entry but without passengers normally having to collect their checked baggage. Some places are the way you describe with passengers forced to collect their baggage, drag it through customs and re-check it.
The reason they have become rare is because they are no longer needed.
The first part of the sale process is attracting the customers attention, then you have to help them decide what exactly to buy, and then finally you can make the transaction.
In the specialised print media world, before the rise of modern internet commerce the ad served both the purposes of attracting the customer's attention and helping them decided which product they wanted. Then the customer would make the actual transaction by mail of phone. As you say this took up a lot of space, I remember some ads that were a lot bigger than "one or two pages".
On the Internet companies have their own websites that are only a click away from the ad. So all the ad has to do is the first step, attracting the customers attention.
If "personalized ads" are for items that you've recently purchased, then whomever did the algorithm for "personalizing" those ads was a complete cretin.
Afaict the issue is that the advertising firms know what you have been looking for but not what you have actually bought.
So you start looking for a new vacum cleaner, the ad networks pick up on this from the pages you visit and the search keywords you use. They start showing you vacum cleaner ads because people who are searching for vacum cleaners have a higher probability of cliking through on vacum cleaner ads than generic ads.
At some point you go ahead and buy a vacum cleaner, but the ad networks don't know that. So the ads for vacum cleaners continue for a while until they have gathered enough information to belive you are no-longer interested.
Afaict plastics, are typically made from hydrocarbons that are not saturated with hydrogen. These hydrocarbons result when heavier alkanes are cracked to produce lighter alkanes.
So the availability of feedstocks for plastic production is likely to depend on the relative demand for heavy and light oil products. Gasoline is somewhere in the middle so it seems like reduced demand for gasoline could do either way in terms of availability of feedstocks for plastic products.
A prison systems job is or at least should be to reduce damage caused by criminals to regular society. There are three mechanisms through which it can do that.
1. Deterrent, most people don't want to be locked up, hopefully the threat of being locked up keeps most people in-line.
2. Rehabilitation. convincing prisoners that they have options other than a life of crime, that they have friends and family they would rather be with than in prison.
3. Removal, if the prisoner is locked up and communications are restricted then it's harder for them to commit crimes against anyone other than the other prisoners and the wardens.
A good prison system IMO balances the factors, not so cushy that it fails to be a deterrent, but not so isolating from society that prisoners become totally dysfunctional and feel they have no choice but to turn back to crime.
Oracle basically murdered browser-based Java
Afaict browser based java had become more of a liability than an asset. Relatively few people were still using it for legitimate applications but (like most browser plugins) it was a massive source of security drama, driving the perception that Java is insecure. I'm not surprised they decided to kill it.
The browser vendors have also been pushing to kill plugins for much the same reason. Afaict firefox and chrome still allow flash as a special case, but have blocked all other native-code plugins. I'm not sure what is going on over on the MS side.
As I understand it, Openjdk is the reference implementation, but it is only distributed in source code form.
Oracle has just changed the license on their binary JDK distributions to be far more restrictive of what can be done without paying.
So that basically leaves many windows users with the choice of building it themselves (which I understand is not exactly a pleasent process), paying oracle, or using binaries from some unknown third-party.
Linux users aren't really affected, they just get their openjdk builds from their distro.
Keeping a site where any random member of the public can create an account and upload videos alive requires walking a fine line. They have to respond sufficiently robustly to copyright infringements to avoid being sued out of existance and sufficiently robustly to advertiser complains to avoid the ad-revenue drying up. Unfortunately the economics of running a site where any random member of the public can watch and upload videos for free means that on average very little human attention can be spent on each user.
Yes there are a few users who attract enough eyeballs that real money is invovled and afaict at least some of those users do get special treatment from youtube but they are a tiny minority.
I think your suggestion of mass fraudulent copyright claims is far more likely to mean the end of youtube as we know it than to mean better protection for creators.
Radioactive half-lives are a measure of how "hot" (radioactive) a compound is per unit time.
Something of an oversimplification. It measures how long it takes for half the radioactive atoms to decay, but what comes out with each decay, whether the decay products are also radioactive and of course what proportion of the original material is actually radioactive atoms varies.
1 mole of 1hr half-life material will kill you much faster than 1 mole of 1 billion year material.
Yes.
But the one hour half-life material will rapidly decay, after a day over 99.9999% of it is gone. So while it's intially dangerous it is not in itself a long term problem (of course it's decay products might be).
Generally the worry from a radiological point of view (chemical toxicity is also a concern with heavy elements) are the isotopes with half-lives somewhere in the middle, short enough that the substance has noticable radioactivity, but long enough that the reactivity won't reduce significantly within a lifetime.
Nuking the enterprise cert is going to give Facebook some pain, making it harder to beta-test apps and harder to use iphones for internal applications, but it's likely to be manageable pain.
Nuking the facebook app from the appstore would likely to significant damage to both Apple and Facebook.
Finally, and this cannot be stressed enough: "Market share" is not important for open-source projects.
I disagreee for a couple of reasons.
Firstly while some opensource projects are run by hobbyists in their spare time and some are run by collaborations between a bunch of stakeholders. Others are largely paid for by a single entity based on revenues they somehow extract from the project. Mozilla falls into the latter category. They make most of their revenue by selling the top search engine slot in their browser (currently to google). If their Market share drops it will be much harder for them to sell that slot for a good price.
Shrinking an organisation is hard, it is hard to avoid losing your best people first as you try and shrink. It's hard to figure out what is really important and what is make-work.
Secondly market share translates to mind share. As much as people may like standards neither the standards themselves or any of their implementations are perfect. If Firefox's market share drops too low then web developers may decide it is not worth bothering to debug their sites on it at which point using firefox may become increasingly painful.
At least this noname Chinese factory managed to not screw up with first production batch
Sure they did, they made an unauthorised substitution of the Ethernet port (fitting a plain jack instead of a jack with integrated magnetics) that meant the first production batch had to be re-worked.
Pretty much all M.2 SSDs from major brands are M.2 2280 , so slightly narrower than a SD card but over twice as long.
Github's advantage is it is the largest and best known of the git collaboration sites, so it is where you and your potential collaborators are most likely to already be set-up rather than creating yet more accounts on yet more sites.
The main downside of github's new free repos seems to be the collaborator limit, gitlab offers unlimited collaborators. Bitbucket seems to count any user you give access to a private repo as part of your "team" for the purpose of account limits (free accounts are limited to 5 "team members").
All these companies have to strike a balance, giving stuff away for free is how they attract new users, but if they give too much away for free then those users will never upgrade to become paying customers. Getting the balance right is tricky, especially as charging for functionality that used to be free is likely to get you conversions in the short term, but also get you a bunch of bad PR, so any changes to what is included in the free tier need to be very carefully thought through.