Not knowing anything about the case other than what I read in the
article, my hang up is that a transit provider can make a phone call
and destroy a customer's business with 30 minutes notice. On a DS3
that has actual real lead time to replace, that's a business killer.
The argument of "should be multi homed" holds some water, but I've
never considered multihoming as a typical remedy for a 30-90 day
outage. And then it only works if lines are underutilized to the
point of loosing one will constantly have zero affect on network
performance, even during peak use, and if there's still some level of
extra redundancy remaining. (Multiple contingency situations aside)
My opinion is probably somewhat influenced by the fact that I'm a
small ISP with customers that want the internet to NOT be slow,
facing that same DS3 lead time problem. I ordered a DS3 in early
December, (who's local loop was to ride on a preexisting OC3, sounds
easy, right?) and with dates slipping over and over again, and with
no firm install date in site provided from the company last week, I
had to finally cancel and order with a different company last week.
For the last month, the last thing I wanted to do was "punt" and
start the process over again, but at some point, one starts to feel
"choiceless."
"Do you think I placed that order in December just for fun?"
I see talk over and over again on NANOG about "Maybe some provider
will come in with [insert new technology here] and compete with the
cable/DSL providers" but as a small provider doing fix broadband
wireless, I just don't see how even an army of small providers can
compete against the likes of TENS OF BILLIONS OF DOLLARS of
cable/telco market capitalization.
After fighting Qwest for ten years, maybe I'm starting to feel a
little hopeless.