Документ взят из кэша поисковой машины. Адрес оригинального документа : http://hea-www.harvard.edu/~fine/opinions/flatrate.html
Дата изменения: Unknown
Дата индексирования: Tue Oct 2 01:10:09 2012
Кодировка:

Поисковые слова: star trail
Flat Rates for Internet Service August 1996

Flat Rates for Internet Service

I saw on C-SPAN the other day, the keynote of an Internet Service Provider's conference in San Francisco, ISPCon, and the title of the talk was "The State of the Internet". The inventor of Ethernet, Bob Metcalfe, was talking about (among other things) the need for better metering of service, so that accurate charging per packet or per byte could be implemented. While he seemed to be discussing only charges between ISPs to deal with their costs of routing traffic for each other, the discussion took on a more general tone, discussing whether or not flat rate was good or bad.

Arguments in favor of flat rates were bandied about. The major argument given in favor of flat rates was that they are responsible for the major developments on the Internet, such as FTP, and World Wide Web. Mr. Metcalfe was dismissive of these arguments - I don't remember the reasons he gave for dismissing them, so they must not have been very compelling reasons.

I tend to agree that flat rate encouraged these services, and that in general that flat rate encourages use (and therefore, innovation), whereas metered rate discourages use.

There is an argument in favor of flat rate that is related to use, but somewhat separate from the above arguments, and that's what I'd like to talk about here. It's a suprising statement to make, and not entirely intuitive, nevertheless I believe it to be the case. Here goes: Flat rate is better able to adapt to the changing world of networking.

Wow, that does sound stupid doesn't it? A rate that's fixed is more flexible than a rate that varies? How could that be? Well, partly because I'm phrasing it badly. Let me put it this way. As a consumer, I have an expectation that networking will continue to become both faster, and cheaper, with time. I also have an expectation that my data needs and desires will only increase steadily with time (the images in WWW were the first big jump in usage; audio and video will only add to that).

If we look only at the first part of my expectations, then every month, my ISP should be reducing my metered rate. I wasn't born yesterday, they're not going to do that. If their profit margin increases, they'll dance a jig for many months before competition gets around to convincing them to lower their rates. I argue that this is the case because most of us are too lazy to switch over to the cheapest ISP every month, especially considering the hassles of a constantly changing address.

Looking only at the second part of my expectations, it would be very difficult for me to justify ever increasing data needs, when I knew this would result in ever increasing costs. Especially since each new hot data type may give me an order of magnitude jump in data consumption.

With a flat rate, I'm encouraged to use more and more and more. With the same flat rate, the ISPs can provide me with more and more and more, without having to adjust their prices to reflect the latest technology. I win, they win, and all the world is a wonderful place.


I have an ISDN line, for which I theoretically pay a flat rate of 32.50 per month. Of cource, since Ameritech has never billed me, it has thus far been free. See how good flat rates are?
Back to Tom's home page.