Followup: All-I-can-eat

The “All-I-can-eat-mentality” article has triggered (as expected) numerous responses. Some of them provided useful data, links to more information or informative perspectives – many thanks to those readers. A few others were unfortunately following the “I-am-right” line without considering facts. Most of the readers from the Service Provider community decided to stay anonymous (when you read all the comments, it becomes obvious they made a wise decision) or respond off-line.

Whatever your position in this issue, I would like to ask you to keep your comments focused on the topic. Although you were all infinitely more polite than the usual forum/blog crowd and provided some really good arguments, writing angry replies does not help. What’s happening with Internet is (like it or not) our common problem … or you could take the blue pill and continue bashing the other side.

I particularly liked the summary of our discussion posted on Slashdot (where someone included the link to my blog):

Whoa, whoa, whoa, that article seems to be promoting a balanced viewpoint that denies a) that telcos are totally evil and b) that we should all be allowed to have as much bandwidth as we want and not have to pay for it. We'll have none of that nonsense on /.

It’s also painfully obvious that the current handling of ISP service offerings is often dismal. For example:

  • Publicly accessible service definitions don’t exist. What most large ISPs offer on their web sites is the X Mbit @ Y $/€/whatever per month. They “forget” to mention that the offered rate is PIR, with expected CIR being only a few percent of PIR.
  • When introducing the traffic caps, some providers started with ridiculously low and probably a bit overpriced figures.
  • A while ago some people decided to keep P2P traffic (which is the worst bandwidth hog of all) in line by resetting TCP sessions (a link to the recent state of affairs would be highly appreciated).

In the next few weeks, I’ll try to cover a few of the topics raised in the comments, including:

  • Why do we have to live with oversubscriptions?
  • Why are the Service Providers forced to use traffic management?
  • Is it possible to have a fair and consumer-friendly service definition?
  • Why will everyone have to invest in deep packet inspection (DPI)?

9 comments:

  1. hi

    One just has to look at commercial costs. Checkout backhaul costs and if only some users are eating up all the bandwidth then the business model is gone. get rid of these top x users by email or incentives or DPI (There is also Alacoya one could consider vs Cisco SCE but even here is legal conisderations) then the model is back on again.

    The rest is surely irrelevent. Its a business and if it doesnt make money then it shouldnt be done.

    At the end of the day these business are owned by shareholders and as such they have empowered the mgmt with the right to do what they want. Consumers can vote with their wallet.

    I agreed with comments suggesting people pay tiered amounts. So if you wanna download all month then you should be able to but you should also pay a fair economic cost for it. This has to happen eventually or fixed line aint ever gonna make money. Check out the graph for mobile ARPU vs bandwidth costs (and as such increaed fixed costs). Since iphone ARPU goes up linearly while badwidth goes up exponetially. In the end people should pay what they use and allow the market to work out the appropriate margins for profit but there must be proft at the end of the day.

    ReplyDelete
  2. Mark Paquette06 July, 2009 16:38

    I think subway and other transportation riders should also pay by the pound.

    ReplyDelete
  3. @Bad Designer: Would you be willing to contact me off-blog? Just use the "Contact me" link (top of the page).

    ReplyDelete
  4. I agree with Bad Designer completely. Bandwidth isn't free or even cheep, especially for the small guys. One $30-$40 a month customer can use bandwidth the ISP is paying 10 to 30 times that amount for.

    One a side note it would be great to see some articles about Cisco SCE and other SPI solutions. I have some SCE experience and that product is less than user friendly.

    ReplyDelete
  5. Agreed, it's very wise to comment anonymously when you offer no evidence to substantiate your claims.

    Ivan, I look forward to your future articles on this matter. Specifically, it will be interesting to see how you:

    * Substantiate your earlier claim that Internet service provision is "far away from being a lucrative business" (examining quarterly returns from a few of the leading residential ISPs would be a good place to start);

    * Justify the almost linear inverse relation between subscription charges and consumer choice (data which I've referenced here before);

    * Acknowledge the obvious conflict of interest in cable television providers implementing broadband transfer caps to artificially restrict the availability of streaming video from third parties;

    * Explain how so many service providers from Sweden to South Korea to the US are able to offer cheap residential service without explicit transfer caps even as you claim consumption-based billing is inevitable;

    * Discuss the economic and technological stagnation effected by consumption-based billing (perhaps one of the anonymous commenters could point out how Vint Cerf is uninformed).

    ReplyDelete
  6. Hi,

    Bad Designer is 100% right, it's still business only - nothing else, so if I or You were running such a service we wouldn't ruin ourselves for customer pleasure only ... ;-), but there's a "war" on the market on the otherhand and the service is sold how it's being sold.

    ReplyDelete
  7. I agree that in the bandwidth world, there's no such thing as a free lunch.

    My primary concern with this is three-fold.

    First one is the development of the internet as a backbone for other services which have traditionally been "internet-offline", such as telephone services, cable-tv, security-systems (alarms) and so on. We all know that the bandwidth curve is growing in a steady rate.

    Take the latest offerings for Xbox-360 users. Video on demand. This will take up 8Mbps to provide an enjoyable experience. I sincerly doubt that most ISP's expect the average traffic rate being 8Mbps even if the customer bought a 20Mbps line.

    ISP's are forced to make sure that they can provide some sort of CIR that will match the majority of consumer's needs.

    The second point is the sort of advertising/contracts thats in place from the service providers to the consumers. At least around here, its common for ISP's to sell you a "20Mbps Fast an amazing internet connection", which in reality will max out around 6Mbps in daily usage. To use an analogy, if you bought a car that you were told could do 20km/litre, and it only did 6km/litre, you would certainly be at the dealer within a few days, complaining your head of. So at some point this will happen in the provider space as well.

    Third, how do you define users? Do you define them at all? In the last case, where you dont differentiate them, you will need to provide the same level of service to ALL of your customers, which is a simple calculation of the CIR you, as a service provider, has agreed on, multiplied by your customer base on each type of service (20Mbps, 10Mbps, 5Mbps and so on). This will give you the bandwidth requirement that you HAVE to support.

    The second case is alot more interesting. Do you profile your users? how do you profile them? As I see it, there are two ways of doing this, if you decide to do it at all. First is the easiest, and the most "fair" imho. You make it completely obvious that there's a cap on how much bandwidth you can use with this sort of "normal-consumer-connectivity-line", and what sort of service you can expect. You have to put this out to people in advance to avoid people showing up at your doorstep complaining that they can not watch the newest James-Bond movie, because you have oversubscribed your lines.

    The second option is the technical approach. You profile consumers based on their usage pattern. Traffic statistics can provide the backdrop for this sort of approach. When you have found an "out-of-profile" consumer, you are again faced with two paths.. 1) Tell the customer, that he's out of profile and if he/she wishes to continue this profile, he/she should be upgraded to another sort of profile to ensure this rate (basically increasing the CIR of this line). 2) Be a jerk about it, and do DPI and start dropping traffic. This will really piss alot of people of, because their expectation is a stable line. (To this end, remember that most customers that bought a 20Mbps and it only being 8Mbps doesnt complain, but if they get intermittent service, they DO complain).

    Basically what it comes down to, is matching expectations with technical limits (based on economics).

    My two cents worth.

    ReplyDelete
  8. I just thought to share with you: in Lithuania there is a practice, which limits user's traffic per day - i.e. you have 10Mbps line for 4Gb of data, if you're above that cap, you get 512Kbps (numbers depend on your data plan). IMO it's way cheapier for the provider and it is also more customer-friendly than DPI.

    ReplyDelete

You don't have to log in to post a comment, but please do provide your real name/URL. Anonymous comments might get deleted.

Ivan Pepelnjak, CCIE#1354, is the chief technology advisor for NIL Data Communications. He has been designing and implementing large-scale data communications networks as well as teaching and writing books about advanced technologies since 1990. See his full profile, contact him or follow @ioshints on Twitter.