Australians can't use Amazon or any of those similar competitors. iTunes is the only choice for most mainstream music here. There are competitors but they offer tiny catalogues.
There's also CD stores or piracy of course, but online there's no real competitor to iTunes.
Once you look beyond the US and UK, I'll think you'll find iTunes has quite a monopoly. And I'm sure it's partly the fault of the record companies.
Why is this a reason to use Firefox 3.6? 3.6 used software-only rendering on all platforms.
What's new here is Firefox 4's Direct2D acceleration on Vista and above, and Direct3D or OpenGL acceleration on all other systems. Remember that there is no such thing as Direct2D on Windows XP.
Remember that there is no such thing as Direct2D on Windows XP, so no version of Firefox (or any other browser) has had or will have Direct2D acceleration on Windows XP.
Firefox 4 does however add Direct3D compositing acceleration on certain video adapters - which is used at a higher level such as when elements overlay other elements.
The suggestion that Firefox 4 uses less acceleration than previous, or other, browsers on XP is mythical.
Only if the client is running IE on Windows XP (or older). Every other system supports SNI
That's wrong. The following won't support support SNI.
* Internet Explorer (any version, even IE8) on Windows XP * Safari on Windows XP * Safari on Mac OS X 10.5.6 * Any browser on Android * The browser on iOS 4 * The browser on Windows Mobile 7 * The browser on Blackberry * wget * Konqueror/KDE
It's gonna be a long time before we can consider using SNI. Wouldn't be surprised if IE8/WinXP hangs around for many years.
To protect against a man in the middle attack. The certificate ensures you are connecting to the server at the end and not some intermediate or impostor that's relaying the messages or posing as the server.
Why is there no mode with encryption, but without the bother of an SSL certificate at all?
Because that would provide no assurance that you are communicating with the server at the end and not some evil computer in between you and that server. It would make the encryption completely useless if you didn't know if the encrypted connection you've negotiated is with the server you want or is with an impostor.
Why was it ever possible to send a password in clear text to begin with?
Because people trusted the connection between their computer and their ISP and onward not to be easily to eavesdrop or intercept. With open wireless networks, this is not the case. As an aside, I've long thought that the existence of open wireless networks was the main problem here and nobody should be using those, but that's my opinion.
Trusted certs don't all cost $$$. StartSSL certs are trusted in all major desktop and mobile browsers and their lowest grade certs, where they verify you are the domain owner only and this verification is automatic, are free.
They may not be guarding access, but they are guarding against an eavesdropper discovering the password. Sure they can session hijack after switching back to HTTP, but they don't have the password. This fact could be leveraged further by requiring the user enter the password (and transition to HTTPS again) whenever something important like changing password, email, or making a monetary transaction is done.
That said, I do agree that switching back to HTTP while still logged in is still a bit silly.
- For HTTPS to work seamlessly it requires a certificate signed by a trusted 3rd party, which usually (but not always) costs money. This would disadvantage smaller websites/businesses.
- HTTPS also (currently) requires a separate IP address per certificate. This would mean the current practice of hosting large numbers of domains on one IP using name-based virtual hosting would not be viable, which is something the shared hosting industry really relies on. IP addresses are also a very limited resource at the moment.
- HTTPS cannot be cached between the user's browser and the server. Even the user's browser uses more conservative caching which will further reduce performance.
I think that if you hold a public office you don't have so much of an "expectation of privacy".
I think it's the same for celebrities too, though there is still obviously a line to cross.
And I don't think this applies for private citizens who are thrust into the public eye through no fault of their own (victims of crime, lottery winners, etc), the usual rules about expectation of privacy would apply.
Also, 12 jurors agreed on this and anybody now reading this, only has some of the facts.
Absolutely right, and this is why this whole thing, including the reason for the University terminating the researcher's contract, is purely speculation. Interesting and throught-provoking speculation though:)
The summary really approaches this from a certain point of view.
Let's look at it from a different point of view...
1. University hires a researcher who has done something in the past which would make him look bad, but is not technically illegal. The researcher has technically never committed any crime, just gotten involved with some bad people at one stage. 2. Blogger starts a campaign of negative publicity against the researcher focused on this aspect of his history. Everything said is true, but is inflammatory and of a nature intended to defame the researcher. 3. The university, for all we know, may even know about the researcher's controversial past already and be cool with it. Either way, the university finds it increasingly difficult to support the researcher in the face of this negative publicity, and eventually lets go of him in order to save face. 4. The researcher sues not for libel, because all the statements made about him were true, but for the running of a negative campaign about him which eventually lead to him losing his job and reputation.
A lot of people have something in their past that makes them look bad, but which is not actually illegal. If someone starts a negative campaign about you based on something like this, and it ends up with you losing your reputation or job, who is in the wrong? In this case, 12 jurors thought it was the blogger.
Does anyone else think that it should be possible to install an app but deny it some of the abilities it requests?
At the moment, if an app wants to send text messages, you can either - install the app and give it permission to send text messages, or - not install the app. There's no way to control what it can do. It is a "foot in the door" type effect where if you really want the app, you have to let it do what it wants, even the things you wouldn't approve of, or you can't use the app.
I'd like to be able to see that an app wants to send texts, deny it that permission, but install it anyway and use the other features of the app normally.
Got a work computer with some McAfee suite on it once. It would not update its virus definitions no matter what I did. After a while on tech support with McAfee it turns out it's because I had my default browser set to Firefox.
Even though they knew their software used ActiveX stuff to update itself, and that would only work on IE, they programmed it to open the default browser to do that updating.
It basically came down to the guy on the other end confirming to me that yes, I not only need to have IE installed, I need to set my default browser to IE to use that McAfee software.
This was about 4 years ago.
TL;DR mcafee antivirus software would only work when IE was set as default browser
When you refer to memory leaks, are you actually just referring to "using" memory?
Memory is cheap - I'd rather use a browser which makes the most of my *unused* memory and gives me a responsive interface, thanks.
As long as an app is not using up all unused memory and then needing even more, forcing the OS to swap or worse (OOM). But a smart app need not do this.
Slashdot Mirror
Australians can't use Amazon or any of those similar competitors. iTunes is the only choice for most mainstream music here. There are competitors but they offer tiny catalogues.
There's also CD stores or piracy of course, but online there's no real competitor to iTunes.
Once you look beyond the US and UK, I'll think you'll find iTunes has quite a monopoly. And I'm sure it's partly the fault of the record companies.
You mean Katey Sagal?
Why is this a reason to use Firefox 3.6? 3.6 used software-only rendering on all platforms.
What's new here is Firefox 4's Direct2D acceleration on Vista and above, and Direct3D or OpenGL acceleration on all other systems. Remember that there is no such thing as Direct2D on Windows XP.
You're not losing anything.
Remember that there is no such thing as Direct2D on Windows XP, so no version of Firefox (or any other browser) has had or will have Direct2D acceleration on Windows XP.
Firefox 4 does however add Direct3D compositing acceleration on certain video adapters - which is used at a higher level such as when elements overlay other elements.
The suggestion that Firefox 4 uses less acceleration than previous, or other, browsers on XP is mythical.
Slashdot ate my angle brackets - see my other reply next to yours.
Stoopid angle brackets. That should read:
"on Mac OS X *less than* 10.5.6",
"iOS *less than* 4", and
"Windows Mobile *less than* 7".
That's wrong. The following won't support support SNI.
* Internet Explorer (any version, even IE8) on Windows XP
* Safari on Windows XP
* Safari on Mac OS X 10.5.6
* Any browser on Android
* The browser on iOS 4
* The browser on Windows Mobile 7
* The browser on Blackberry
* wget
* Konqueror/KDE
It's gonna be a long time before we can consider using SNI. Wouldn't be surprised if IE8/WinXP hangs around for many years.
To protect against a man in the middle attack. The certificate ensures you are connecting to the server at the end and not some intermediate or impostor that's relaying the messages or posing as the server.
Because that would provide no assurance that you are communicating with the server at the end and not some evil computer in between you and that server. It would make the encryption completely useless if you didn't know if the encrypted connection you've negotiated is with the server you want or is with an impostor.
Because people trusted the connection between their computer and their ISP and onward not to be easily to eavesdrop or intercept. With open wireless networks, this is not the case. As an aside, I've long thought that the existence of open wireless networks was the main problem here and nobody should be using those, but that's my opinion.
Trusted certs don't all cost $$$. StartSSL certs are trusted in all major desktop and mobile browsers and their lowest grade certs, where they verify you are the domain owner only and this verification is automatic, are free.
Is the percentage of Windows users still on XP less than 50% yet?
They may not be guarding access, but they are guarding against an eavesdropper discovering the password. Sure they can session hijack after switching back to HTTP, but they don't have the password. This fact could be leveraged further by requiring the user enter the password (and transition to HTTPS again) whenever something important like changing password, email, or making a monetary transaction is done.
That said, I do agree that switching back to HTTP while still logged in is still a bit silly.
- For HTTPS to work seamlessly it requires a certificate signed by a trusted 3rd party, which usually (but not always) costs money. This would disadvantage smaller websites/businesses.
- HTTPS also (currently) requires a separate IP address per certificate. This would mean the current practice of hosting large numbers of domains on one IP using name-based virtual hosting would not be viable, which is something the shared hosting industry really relies on. IP addresses are also a very limited resource at the moment.
- HTTPS cannot be cached between the user's browser and the server. Even the user's browser uses more conservative caching which will further reduce performance.
Actually, the original is here: http://rasteri.blogspot.com/2011/03/we-are-ie-comparing-every-version-of.html
It would be cool if they displayed not only signal strength, but the WPA passphrase too.
I think that if you hold a public office you don't have so much of an "expectation of privacy".
I think it's the same for celebrities too, though there is still obviously a line to cross.
And I don't think this applies for private citizens who are thrust into the public eye through no fault of their own (victims of crime, lottery winners, etc), the usual rules about expectation of privacy would apply.
Absolutely right, and this is why this whole thing, including the reason for the University terminating the researcher's contract, is purely speculation. Interesting and throught-provoking speculation though :)
The summary really approaches this from a certain point of view.
Let's look at it from a different point of view ...
1. University hires a researcher who has done something in the past which would make him look bad, but is not technically illegal. The researcher has technically never committed any crime, just gotten involved with some bad people at one stage.
2. Blogger starts a campaign of negative publicity against the researcher focused on this aspect of his history. Everything said is true, but is inflammatory and of a nature intended to defame the researcher.
3. The university, for all we know, may even know about the researcher's controversial past already and be cool with it. Either way, the university finds it increasingly difficult to support the researcher in the face of this negative publicity, and eventually lets go of him in order to save face.
4. The researcher sues not for libel, because all the statements made about him were true, but for the running of a negative campaign about him which eventually lead to him losing his job and reputation.
A lot of people have something in their past that makes them look bad, but which is not actually illegal. If someone starts a negative campaign about you based on something like this, and it ends up with you losing your reputation or job, who is in the wrong? In this case, 12 jurors thought it was the blogger.
I concur.
Where I come from, a speed bump is something designed to slow you down.
When you say "back-of-the-envelope" you mean Windows Calculator don't you ...
If this were a JPEG it'd be around 40 MB.
Does anyone else think that it should be possible to install an app but deny it some of the abilities it requests?
At the moment, if an app wants to send text messages, you can either
- install the app and give it permission to send text messages, or
- not install the app.
There's no way to control what it can do. It is a "foot in the door" type effect where if you really want the app, you have to let it do what it wants, even the things you wouldn't approve of, or you can't use the app.
I'd like to be able to see that an app wants to send texts, deny it that permission, but install it anyway and use the other features of the app normally.
Got a work computer with some McAfee suite on it once. It would not update its virus definitions no matter what I did. After a while on tech support with McAfee it turns out it's because I had my default browser set to Firefox.
Even though they knew their software used ActiveX stuff to update itself, and that would only work on IE, they programmed it to open the default browser to do that updating.
It basically came down to the guy on the other end confirming to me that yes, I not only need to have IE installed, I need to set my default browser to IE to use that McAfee software.
This was about 4 years ago.
TL;DR mcafee antivirus software would only work when IE was set as default browser
When you refer to memory leaks, are you actually just referring to "using" memory?
Memory is cheap - I'd rather use a browser which makes the most of my *unused* memory and gives me a responsive interface, thanks.
As long as an app is not using up all unused memory and then needing even more, forcing the OS to swap or worse (OOM). But a smart app need not do this.
A large collection requires lots of floor space which costs a lot.
You can never have as big a collection as you could with electronic materials.
Lots of cost in acquisition and (sometimes) cataloguing.
Lots of manual labour involved in processing, shelving etc requiring large staff and costs.
Only one person can read them at a time.
Books get returned late, and they get lost.
But the toppings contain potassium benzoate.