He has a hole in his truck "probably from someone bumping the truck into a structure" which he calls his "skylight", which the birds peck at he says: http://frominsidethebox.com/vi...
I think that fits into the "animal dung" category.
A good way to fix a vulnerability where you can send data without being charged is by charging for that data. Its always something better to say "we fix a vulnerability pointed out by security researchers" than to say "we demand money for using our infrastructure even if it is not involved".
Amazon is a tech company, Walmart isn't. I guess somewhere at Walmart HQ there has been a conversation like that:
A) Damn! Our sales are dropping! Which company has taken our customers? B) I've heard of Amazon to be successful. Its an internet retail store. A) Can we buy some "Internet" for us too? B) Internet isn't bought, its a network for communication. A) Either way, can we roll it out? B) We could, but Amazon has a major head start and has much more experience in that field B) Also, they largely benefit from synergies from their cloud services. A) We have to get that experience too. And we must get those synergies! A) We must become better than them, and win them on their own game! We are longer in the business than those computer-kids! B) Ok, boss.
Rust considers it "safe" to:
Deadlock
Have a race condition (https://doc.rust-lang.org/nightly/nomicon/races.html)
Leak memory
Fail to call destructors
Overflow integers
Abort the program
Delete the production database
This. The two biggest rust projects rustc and servo don't even compile with stable rust. And both get developed by those who really know rust. Yes, it makes sense to have unstable intrinsics for a compiler, but not even a slower compat mode??
And the language itself may be guaranteed to be stable, but many many libraries aren't, and lots of the language's features are only available as unstable. It happens that crates break their API, because their devs don't care or don't know.
Rust does have search results, but rust had a really unstable past, if a search result is before stabilisation with 1.0, its usually deprecated. And most results are from that time. But since stabilisation things got better, at least I hope, and everybody says. Just ask your question anew, people are usually very helpful.
Its very simple to get a backtrace for a panic in rust. Just set the RUST_BACKTRACE environment variable to 1 before executing the program. This is something far more easier to explain to your users than to explain the long process how to download a version with debug symbols of the program and use gdb.
And black-to-black crime is bad, agreed, but a police officer who shots a black innocent is another category. In every job accidents happen, and the accidents in jobs where there is a gun involved might be lethal, but the public cases aren't accidents. I don't think that black-to-white or white-to-black crime should get more attention than black-to-black.
The common cold isn't as harmful for the people's lives who get it, as the AIDS syndrome.
But I do say that we should focus more on security on the streets than in the air, because there are much more deaths per kilometer on the streets than in the air.
Centuries ago, you got into serious problems if you ran around claiming you have a mission to fulfil the devil gave you. Now, this gets replaced by feminism, and everybody who doesn't join in the holy movement is an infidel who should be punished. The level of intolerance towards people who aren't "feminists" is raising and raising.
If they really wanted to help women, they should go to india, help solving its rape problem, or to saudi arabia, where women get stoned when they sleep with men who aren't their husband.
No, the difference between 2nd preimage and collision is that for your hash function HASH, you have for 2nd preimage sth, and HASH(sth) given, and want to get sth2 so that HASH(sth) == HASH(sth2). sth2 is choseable by you, either completely, or only in parts. Sometimes you only have HASH(sth), but you never can modify sth.
Now for collision, you only have HASH given, as function, but you can chose both sth and sth2, either completely, or in parts.
This means you never can rewrite history with git, if you only can do collision attacks. You can however have two "histories": first an "official" one, then an inofficial one. So if you can do collision attacks, and you publish releases as signed email, together with the git sha ID, then you are vulnerable to NSA NSLs where they ask you to create a backdoored version of your software with the same hash as the public non-backdoored one for the next release. They can't require that for previous releases, as that would be 2nd preimage.
Ignore how Chrome is still rising steadily despite already doing all of the things you dislike
In some points, mozilla are so idealistic, they hurt themselves. They didn't chose to implement h.264 for HTML5, because to win the codec war. This just led to users switch to chrome. I know, google also did something very mean, they agreed to allow any video to be possible to be played back with ogv and remove h.264, both promises they didn't keep, at least the youtube promise hasn't been kept for a long time. But mozilla could have reacted earlier.
If mozilla really wanted to develop the best browser, they should implement more web APIs. Just think of the still largely missing MSE, or the U2F (which isn't that of a priority, but is a good example of yet another API not implemented). They also could have implemented apple's HLS, in order to beat iphones on this. But rather they chose "no", and said "once we have MSE, this is all doable by js". This is vaporware speak, not how to make a browser that wants to be the most used one.
Then they should do things only for those operating systems, but not for all of them.
shitty addons are more important than letting them fix things
They aren't. Fixing things is important, but why do they announce to lock down their Add-on API _after_ they have implemented and 99% completed e10s support? I know, there is servo, and probably its right when they say that many of the functionalities of the old APIs are now possible with HTML5. But WebRTC simply doesn't replace an API to send raw udp data.
Firefox surely just had to stay the course, and it would have been just fine
They should be open to change, but they shouldn't abandon their old principles.
Slashdot Mirror
Time synchronisation is in systemd already:
http://lists.freedesktop.org/a...
And it uses SNTP, not NTP.
I don't trust a single thing they [microsoft] do
Welcome to slashdot. Your position is quite shared on this place.
not even animal dung.
He has a hole in his truck "probably from someone bumping the truck into a structure" which he calls his "skylight", which the birds peck at he says: http://frominsidethebox.com/vi...
I think that fits into the "animal dung" category.
Not everybody, but they would certainly relax the market, thus lowering the price for YOU who wants to live outside the google barracks.
In fact, what you describe has been done by multiple industrials, most prominently ford.
is it hosted at sourceforge?
A good way to fix a vulnerability where you can send data without being charged is by charging for that data. Its always something better to say "we fix a vulnerability pointed out by security researchers" than to say "we demand money for using our infrastructure even if it is not involved".
Real programmers use netcat.
Have you ran "sudo apt-get update"?
Here's the video URL returning the mp4 file. No script stuff needed, only curl, wget, or just a browser which can play media files:
http://cf.c.ooyala.com/dzMWU3e...
Amazon is a tech company, Walmart isn't. I guess somewhere at Walmart HQ there has been a conversation like that:
A) Damn! Our sales are dropping! Which company has taken our customers?
B) I've heard of Amazon to be successful. Its an internet retail store.
A) Can we buy some "Internet" for us too?
B) Internet isn't bought, its a network for communication.
A) Either way, can we roll it out?
B) We could, but Amazon has a major head start and has much more experience in that field
B) Also, they largely benefit from synergies from their cloud services.
A) We have to get that experience too. And we must get those synergies!
A) We must become better than them, and win them on their own game! We are longer in the business than those computer-kids!
B) Ok, boss.
Some things rust guarantees, others it doesn't:
https://doc.rust-lang.org/nigh...
Rust considers it "safe" to:
Deadlock
Have a race condition (https://doc.rust-lang.org/nightly/nomicon/races.html)
Leak memory
Fail to call destructors
Overflow integers
Abort the program
Delete the production database
This. The two biggest rust projects rustc and servo don't even compile with stable rust. And both get developed by those who really know rust. Yes, it makes sense to have unstable intrinsics for a compiler, but not even a slower compat mode??
And the language itself may be guaranteed to be stable, but many many libraries aren't, and lots of the language's features are only available as unstable. It happens that crates break their API, because their devs don't care or don't know.
Rust does have search results, but rust had a really unstable past, if a search result is before stabilisation with 1.0, its usually deprecated. And most results are from that time. But since stabilisation things got better, at least I hope, and everybody says. Just ask your question anew, people are usually very helpful.
I think this is somebody trolling. FFI and gem are common fanboy words.
Its very simple to get a backtrace for a panic in rust. Just set the RUST_BACKTRACE environment variable to 1 before executing the program. This is something far more easier to explain to your users than to explain the long process how to download a version with debug symbols of the program and use gdb.
Even if templates are already present in C++, they are a PITA to use. Rust is really really easy to use for those.
Ten ways how to generate more clicks.
And black-to-black crime is bad, agreed, but a police officer who shots a black innocent is another category. In every job accidents happen, and the accidents in jobs where there is a gun involved might be lethal, but the public cases aren't accidents. I don't think that black-to-white or white-to-black crime should get more attention than black-to-black.
The common cold isn't as harmful for the people's lives who get it, as the AIDS syndrome.
But I do say that we should focus more on security on the streets than in the air, because there are much more deaths per kilometer on the streets than in the air.
Agreed, these two points are better.
Marital rape is unpunished in saudi arabia as well.
It's not just america. Parts of Europe have it, too.
Centuries ago, you got into serious problems if you ran around claiming you have a mission to fulfil the devil gave you. Now, this gets replaced by feminism, and everybody who doesn't join in the holy movement is an infidel who should be punished. The level of intolerance towards people who aren't "feminists" is raising and raising.
If they really wanted to help women, they should go to india, help solving its rape problem, or to saudi arabia, where women get stoned when they sleep with men who aren't their husband.
No, the difference between 2nd preimage and collision is that for your hash function HASH, you have for 2nd preimage sth, and HASH(sth) given, and want to get sth2 so that HASH(sth) == HASH(sth2). sth2 is choseable by you, either completely, or only in parts. Sometimes you only have HASH(sth), but you never can modify sth.
Now for collision, you only have HASH given, as function, but you can chose both sth and sth2, either completely, or in parts.
This means you never can rewrite history with git, if you only can do collision attacks. You can however have two "histories": first an "official" one, then an inofficial one. So if you can do collision attacks, and you publish releases as signed email, together with the git sha ID, then you are vulnerable to NSA NSLs where they ask you to create a backdoored version of your software with the same hash as the public non-backdoored one for the next release. They can't require that for previous releases, as that would be 2nd preimage.
Ignore how Chrome is still rising steadily despite already doing all of the things you dislike
In some points, mozilla are so idealistic, they hurt themselves. They didn't chose to implement h.264 for HTML5, because to win the codec war. This just led to users switch to chrome. I know, google also did something very mean, they agreed to allow any video to be possible to be played back with ogv and remove h.264, both promises they didn't keep, at least the youtube promise hasn't been kept for a long time. But mozilla could have reacted earlier.
If mozilla really wanted to develop the best browser, they should implement more web APIs. Just think of the still largely missing MSE, or the U2F (which isn't that of a priority, but is a good example of yet another API not implemented). They also could have implemented apple's HLS, in order to beat iphones on this. But rather they chose "no", and said "once we have MSE, this is all doable by js". This is vaporware speak, not how to make a browser that wants to be the most used one.
No. That's second preimage attack. Collision is if you can chose multiple versions to map to the same hash.
Firefox being locked out of major mobile OSes
Then they should do things only for those operating systems, but not for all of them.
shitty addons are more important than letting them fix things
They aren't. Fixing things is important, but why do they announce to lock down their Add-on API _after_ they have implemented and 99% completed e10s support? I know, there is servo, and probably its right when they say that many of the functionalities of the old APIs are now possible with HTML5. But WebRTC simply doesn't replace an API to send raw udp data.
Firefox surely just had to stay the course, and it would have been just fine
They should be open to change, but they shouldn't abandon their old principles.