Friday, 4 April 2008

A response to LKO on Net Neutrality - Part 2

I'm sorry, but I laughed out loud when I read this post.

Particularly the hand-wringing over this statement:
"In order to throttle the Internet traffic originating from/or destined for end-user customers of independent ISPs, Bell is using measures to first, open each data packet, examine the packet data and header information, and then apply certain rules to the content in question. This aspect of Bell’s wholesale throttling activities give rise to concerns that Bell’s actions violate the privacy of the communications of its wholesale customers (as well as that of their own end-user customers). It also gives rise to concerns that Bell has violated its duty under section 36 of the Act not to control the content or influence the meaning or purpose of telecommunications carried by it for the public."

Now, I may not be a total techy, but that says to me that Bell is looking at the content people are sending across the internet, and then determining what content goes fast, and what content goes slow.

Are you freaking kidding me?!?!

"by examining the packet data and packet header information of GAS customer traffic, Bell can identify, inter alia, the type of data being transferred, the ISP upon whose network the data is being transferred, an end-user’s intention to acquire certain types of Internet content and the IP address and, hence, the identity of the end-user customer who is sending/receiving the data. The collection and use of such information by Bell, which in this case would have clearly been done without the prior consent of the end-user customers so affected, violates the privacy of such individuals."

As I pointed out in Part 1, there is a world of difference between the application used to transmit data, and the actual content of the data itself. What Bell is doing is inspecting the traffic to see the type of application being used, and applying rules to that application to ensure that it is not consuming too much bandwidth. I remind you, it is estimated that 70 to 90% of online bandwidth is consumed by 5 to 10% of all users, leaving very little bandwidth for the rest of us.

We don't know 100% of the story here. As I pointed out earlier, online bandwidth is oversubscribed. If we allow true "net neutrality", where no application throttling is allowed to take place, it is entirely possible that 10% of the user community will consume 100% of the available bandwidth, leaving nothing for everyone else. Which would contradict the stated purpose of open access for all.


Lord Kitchener's Own said...

Well Catelli, you clearly know more about the technology than I, but your contention that this is all just about determining "the type application being used to transmit data" does seem to be at odds with the CAIP's contention that "Bell can identify, inter alia, the type of data being transferred, the ISP upon whose network the data is being transferred, (and) an end-user’s intention to acquire certain types of Internet content" so are the CAIP misrepresenting what Bell is doing in your eyes?

Also, and again I'm no expert, but I've heard plenty of people say that that whole "70-90% of bandwidth being consumed by 5-10% of users" is hogwash, and that those stats come right from Rogers and Bell (or more acurately, firms hired by Rogers and Bell to crucnch the numbers). I've heard estimates that up to 10% of bandwidth is being consumed by YouTube videos alone. Is that crazy? 'Cause that would seem to belie this notion that 90% of the traffic is being taken up by 10% of users (being that that would leave 10% for YouTube and 0% for the other 90% of users).

Anyway, thanks for your posts on this. I'm certainly interested to learn more about all of these issues!

Catelli said...

It depends on the use of the word data. In Tech speak, there are two kinds of systems, data and voice. Voice systems are old-fashioned telephone systems that use analogue signals to transmit sound.

Data networks use digital signals to transmit "data". Because the two systems historically stayed separate, data become synonymous with digital.

Keep in mind that it is automated systems that parse and inspect the digital data packet. It isn't humans doing it, and the information is not logged for any extended period of time. Many corporations, educational institutes, not for profit orgs etc. are deploying packet inspection technologies to manage their networks. Its becoming accepted that using packet inspection technologies to manage bandwidth are a required cost of business to keep networks functional.

It is everywhere, even on your computer. If you have Windows XP or better on your computer, it automatically installs a QoS manager against your network connection. Most people don't use it, but it is there, and its purpose is to ensure Quality of Service (hence the QoS acronym).

I would tend to believe the bandwidth numbers based on what I've seen on our corporate network and in discussions with other IT Network managers. Normal web traffic and e-mail can be done on a 256k connection with very fast results. At home, I use Bell's DSL lite service because that's 99% of our Internet Activity. It isn't until you view streaming video (YouTube, etc.) or download through P2P (BitTorrent, Kazaa, eDonkey) that you need a faster home Internet Service. Everyone I know that has the "superfast" Internet service uses it for downloading.

The fact that they have 5 Megabit connections for their use and I (and many of my family members) have a 512k connection for my use speaks to the different data needs of the different types of traffic.

(For example, take the recent NCAA championship, it had many corporations nervous about what viewing it online could do to their networks.)

Oh, and the stats have to come from Rogers and Bell as it is their networks that are the backbone of the Canadian Internet experience. Only they have access to those numbers, no one else does.

Take the independent ISPs claims with a grain of salt too. It is in their best interest to inflate claims and provoke fears about the big corps inspecting data packets. it gets the public on their side. Just as it is in the big corps interest to inflate their story too. Each side has grains of truth in their claims, so as in all fights where truth is the victim, there are 3 sides to the story. I'm trying to show that there is way more to this than meets the eye.