randy@psg.com (Randy Bush) writes:
> Just out of curiosity, why do many (not all) of the large backbone providers
> establish their face to the web (their corporate webserver) on slow,
> badly positioned machines?
Because it is done by marketing, not engineering?
In my opinion, this alone should reduce the Value Number
of any given provider.
Personally, flaws or not, I welcome Boardwatch's attempts
to come up with a widely-published metric for the
Internet. This likely will lead other publications into
similar investigations, some of which may well bring the
writers and editors of various periodicals into contact
with the folks at CAIDA.
Also, flaws or not, Boardwatch did do something
fantastically clever, and that's examining things on an
end-to-end basis, rather than obsessing about details of
what's going on between the endpoints. People concerned
about the abysmal end-to-end throughput of even modern TCP
across much of the present Internet should be rejoicing
and helping other journalists develop better and more
scientific approaches to categorizing expected versus
observed end-to-end performance.
The combination of work aimed at measuring the internals
of one's network and work that measures the "(un)happiness"
of certain classes of applications should be of enormous
value to engineers willing to admit that they don't know
all the things that affect network performance observed by
end-users.
The reality, however, is that most American ISPs seem to
engage in knee-jerk denial or aggressive posturing
whenever there is a suggestion that their network is
anything but perfect. I certainly have been in the middle
of that kind of thing, so I can hardly claim innocence.
Such reactions are pure marketing: we can't admit that
maybe there is some way we could improve things or some
set of things our network doesn't do well because that
would hurt our product image.
People who react with marketing-think and who build with
marketing-think in the first place deserve to be torn
apart by analyses based in marketing-think. While the
article in question isn't in my hands yet, based on Mr
Rickard's and others' comments on the NANOG list, I think
it's safe to guess that this is precisely what the
Boardwatch study does.
Sean.