Forgot your password?
typodupeerror

Comment: Re:Thanks (Score 1) 393

TCP performance on the Internet is almost totally limited by latency (AKA RTT or round trip time for the ACKs), not the bandwidth.

Modern TCP stacks, including Windows 7, 8 and Linux these days have a feature called TCP Timestamping, where an RTT estimate is taken for the connection, and a feature called TCP Autotuning where the window size is automatically scaled up to fill a Long fat pipe.

So no... the days where TCP throughput of a session was totally limited by latency are long gone.

Comment: Re:I'll believe it when it actually happens. (Score 1) 116

by Bill, Shooter of Bul (#47550081) Attached to: eSports Starting To Go Mainstream
Yeah, I kind of doubt its going to be easy to capitalize on any kind of gaming tv, because of the existing streams .. unless the competitions start charging for broadcasting rights. In which case they'll be the inevitable backlash from the comuniity. Viewership will go from worldwide without restrictions, to nation based, depending on rights distributions.

Comment: Re:Thanks (Score 2) 393

You could even run a network monitoring app. But the browser is one highly visible one that most people already have installed.

Perhaps you could, but now essentially you are having "users that think they have problems" downloading an extra application and they start monitoring after there's a problem most likely.

This means your app cannot get the right data on what's normal for the user or for the world, because you have a sample of app users that are biased towards users that already are experiencing network issues of some sort, and you don't have a good baseline for the user that installed it either.

Comment: Re:Thanks (Score 4, Insightful) 393

It will probably end up pissing off ISPs to the point of either finding ways of faking the data, blocking the data, or just as policy telling customers to ignore the speed numbers.

If the data is blocked, the browser should figure out why and explain to the user that there seems to be an issue with their network; in other words "Blocking" should make it even worse for the ISP. a smarter browser UI could be a tremendous help to support technicians, which the ISPs should absolutely love ---- perhaps even tell the user exactly which entity to contact, even display their ISP's support number on the screen, to help accelerate the problem resolution process, and providing access to comments by other users of the same ISP, leading to happier customers, and customers who can share info with each other pertinent to troubleshooting or why this is happening, etc.

A lot of people won't be able to distinguish when something is their ISP's fault and when it might be the end servers fault.

I am suggesting the browser should also take some responsibility to the interpretation of the results here. There should be a highly visible "troubleshooting" button that causes some tests to be run. Explanations should be right there in a natural language that any English speaker could understand.

The browser should not show an alert if there is not enough data to make a conclusion with a fair measure of statistical confidence.

We can definitely make a strong distinguishment between a "web site performance issue" and a client connectivity issue, with data from a sufficient number of users.

The browser would also need to take into account geographic location and client connectivity, however.

e.g. Is the site slow because the visitor is half way around the world from the nearest mirror, or is it slow because they're connecting over congested WiFi or 3G networks, instead of a wired connection?

I realize it's not "easy", but the web browser is the only software component that is in a position to take the kinds of measurements that are required and help alert the user to the problem, tell the user which entity they should contact, and assist with troubleshooting.

Comment: Re:Thanks (Score 5, Interesting) 393

SO when you pay for that service it says something like "up to 75mbps" which in reality means that the speed test and google's home page could see that much speed and everyone else will look like dial up from the 1990's.

I have a suggestion.... Web browsers should take some measurements and display prominently in a visible status bar or other location.... average TCP throughput --- And Estimated average bandwidth;

Both a "this site" value, a "this browser session" value, and (Optionally) if the user decides to share their numbers, Community average bandwidth for this site, Community average bandwidth for this ISP, and Community average for this site on this ISP.

If Community average for this site on this ISP is more than a standard deviation below Community average for this site,

Then a little warning exclamation point should appear to the right of the browser bar. On mouseover, and for a few seconds after loading the page, a little warning bubble should appear for a few seconds. "Your internet service provider seems to have below average performance in loading this page."

Comment: Re:subdomain trust (Score 1) 90

by mysidia (#47536433) Attached to: New SSL Server Rules Go Into Effect Nov. 1

Or is this an option?

RFC 3280 #4.2.1.11

The name constraints extension, which MUST be used only in a CA certificate, indicates a name space within which all subject names in subsequent certificates in a certification path MUST be located. Restrictions apply to the subject distinguished name and apply to subject alternative names.

...

>

It is an option that was not forced on the root CAs. Essentially none of the public CAs are signing from intermediary CAs with name restrictions applied to their certificates.

Generally the restriction mechanism is only allowed to do something kind of "creepy"; where the root CA essentially "sells" this service to a smaller company for perhaps $50,000 or so and issues a restricted certificate --- that allows whoever bought this service to sign subcerts within certain constraints.

Comment: Re:Why? (Score 1) 90

by mysidia (#47536381) Attached to: New SSL Server Rules Go Into Effect Nov. 1

or at least just force via policy certain certificates onto each computer's browser as trusted?

That works fine for Internet Explorer on Windows via group policy.

It doesn't work for Firefox or Java (separate private trusted certificate storage databases).

More importantly: It doesn't work for iPhones, Androids, or macs accessing intranet resourses, or that require a valid certificate to setup Activesync connection.

Comment: Re: name and location tweeted... (Score 1) 877

This isn't really about privacy, though - it's about SouthWest's perogative to refuse service to someone they feel was being abusive.

Their perogative to arbitrarily refuse service ends when they accept your money and enter into an agreement to render service; they essentially can't back out without cause, or they risk being sued for breach of contract and discriminatory actions.

Also, there is this matter of coercing a customer to remove a public message under threat of arrest, that the customer had a right to post.

"Success covers a multitude of blunders." -- George Bernard Shaw

Working...