Actually there is a lot of incentive, but the problems for BGPSec are: - you need to change the software on the router - the vendor has to add it - it will use a lot more CPU on your router, which can cause more problems than it solves - DNSSEC and BGPSec are brittle, many mistakes have been made, just look up
"However, because IPv6 is designed from the outset to be a hierarchical addressing scheme, address tables should end up being much smaller (even though each entry is longer) which in turn means that accidents should be less common."
The hierarchical addressing scheme for IPv6 never happend in practice. IPv6 is routed the same way as IPv4.
But less prefixes are needed, because with IPv4 much smaller networks are allocated to providers. With IPv6 every provider network basically just needs only a few prefixes because IPv6 is such a large address space.
Why would anything need to communicate with the service ? Yes, that would make it an attackvector, but I doubt it would need to.
The services just downloads from the Mozilla site (and checks certificate of the https-connection and/or downloaded binary) and installs that in the program files directory.
That's all it should be doing.
Maybe and only maybe, it should try to read somewhere on the system if the automatic updates should not be enabled.
But more likely any setting for that should just disable the update service.
Firefox isn't like IE, it is trivial to just unpack many different versions for testing (and you can also create multiple profiles). You can even run them side by side, if Firefox is your main browser you don't even have to use your every day browser as the browser you test with.
It would have been a good idea to have started the project earlier, Firefox 4 was the biggest memory hog of all Firefox versions, but it's been going down each and every version since they started a project to fix it:
I would have said the same, but on Wikipedia it says:
Since version 2.5.46 of the Linux kernel, the major parts of Clinux have been integrated with the main line kernel for a number of processor architectures.
The project continues to develop patches and supporting tools for using Linux on microcontrollers.
So I think Linux does not require an MMU. Ubuntu (including their ARM version) does.
Firefox 11 pretty much fixes most outstanding bugs, they are a few in Firefox 12. They've are now busy with the top 100 addons and over 50% of the leaky addons have been fixed.
Slashdot Mirror
Actually there is a lot of incentive, but the problems for BGPSec are:
- you need to change the software on the router
- the vendor has to add it
- it will use a lot more CPU on your router, which can cause more problems than it solves
- DNSSEC and BGPSec are brittle, many mistakes have been made, just look up
and so on.
"However, because IPv6 is designed from the outset to be a hierarchical addressing scheme, address tables should end up being much smaller (even though each entry is longer) which in turn means that accidents should be less common."
The hierarchical addressing scheme for IPv6 never happend in practice. IPv6 is routed the same way as IPv4.
But less prefixes are needed, because with IPv4 much smaller networks are allocated to providers. With IPv6 every provider network basically just needs only a few prefixes because IPv6 is such a large address space.
And on Android by default that would be ?: Google
Have you tried looking at about:memory or creating a new profile ? Or running in 'Safe Mode' ?
That might help find the cause and solve the problem, instead of using workarounds.
Have you tried to create a new profile ?
Why would anything need to communicate with the service ? Yes, that would make it an attackvector, but I doubt it would need to.
The services just downloads from the Mozilla site (and checks certificate of the https-connection and/or downloaded binary) and installs that in the program files directory.
That's all it should be doing.
Maybe and only maybe, it should try to read somewhere on the system if the automatic updates should not be enabled.
But more likely any setting for that should just disable the update service.
Like Google Chrome ?
I don't know, ask Google I guess.
I believe since Firefox 10 plugins should break anymore:
https://wiki.mozilla.org/Features/Add-ons/Add-ons_Default_to_Compatible
Wasn't that already solved in Firefox 10 ?:
https://wiki.mozilla.org/Features/Add-ons/Add-ons_Default_to_Compatible
Firefox isn't like IE, it is trivial to just unpack many different versions for testing (and you can also create multiple profiles). You can even run them side by side, if Firefox is your main browser you don't even have to use your every day browser as the browser you test with.
Wasn't that already solved in Firefox 10 ?:
https://wiki.mozilla.org/Features/Add-ons/Add-ons_Default_to_Compatible
It would have been a good idea to have started the project earlier, Firefox 4 was the biggest memory hog of all Firefox versions, but it's been going down each and every version since they started a project to fix it:
http://blog.mozilla.org/nnethercote/category/memshrink/
The biggest problem was a problem with JavaScript memory leaking, which was caused by the javascript engine replacement.
Actually March 2013:
https://wiki.mozilla.org/RapidRelease/Calendar
That policy was already changed in Firefox 10:
https://wiki.mozilla.org/Features/Add-ons/Add-ons_Default_to_Compatible
People can already change the brake system now (or do bad maintenance), I thought we were specifically talking about the engine.
Most people don't hack their Office software either. That doesn't mean it could be useful to be able to do that.
What is special about electric cars anyway ? Your statement about 'public roads' and all that.
That seems to me like the equivalent of radio.
I would have said the same, but on Wikipedia it says:
Since version 2.5.46 of the Linux kernel, the major parts of Clinux have been integrated with the main line kernel for a number of processor architectures.
The project continues to develop patches and supporting tools for using Linux on microcontrollers.
So I think Linux does not require an MMU. Ubuntu (including their ARM version) does.
I want to check if mine is on the list ;-)
I wouldn't know how they could patent it. The Microsoft proposal is based on combining SPDY and Websocket.
Encryption is pretty much useless for a communication protocol if you don't know who you are communicating with.
Opertunistic IPSEC does exist, it isn't widely deployed, it depends on DNSSEC AFAIK.
Let's talk about adoption rates.
Wrong again, it prevents transparent proxies.
If the user/administrator explicitly configures a proxy in the browser, it works just fine.
Firefox 11 pretty much fixes most outstanding bugs, they are a few in Firefox 12. They've are now busy with the top 100 addons and over 50% of the leaky addons have been fixed.
Not only on the network, but also in your OS networking stack and networkcard and wifi drivers.
That is a large part of what they are fixing in Linux.
It is obvious why wifi drivers might have large buffers, retransmission is much more common than on the wire.