Now, that i'm not in the ISP business --
*all* large ISPs have the same "small" problem, namely the complete lack of understanding of the current workforce market by the management. There are only few people in the world who really know how the global routing works, and some of them already worth more than average Joe Banana telco tech will make for his entire life. Needless to say that they don't work for telcos, so those who still do have something to think about.
I think the biggest problem for most ISPs today is not so much a dearth of capable engineers. It shouldn't really take long, or much effort to train people who are capable of learning and are capable in general. Some of the biggest problems many ISPs face have to do with a dearth of money resources, since most ISPs are small upstarts, then a dearth of capable managers and marketers. Capable managers find capable engineers and capable marketers. [As a point of reference, it only took me about four months to go from knowing nothing about Cisco's to running an ISP's network (multi-homed to different providers); I had a college student's background in IP, a few books, Cisco manuals, and the networking community at large to draw from. To be fair, one network engineer who has participated here has little respect for me, though I know some others do. My point is that people can be trained. Many important participants in this list are very young!] I suspect most ISPs were/are started, and, most importantly, run by folk who don't understand the rapidly evolving market for Internet access. If people praise UUNET's service, I would dare say that UNNET can satisfy them because it has been a provider for quite some time now; they should know their business pretty well. I also dare say that other ISPs who have been around for some time are good for nothing.
In my practice, leaving a large ISP by members of engineering or operational staff causes rather steep increases in their income. On the other hand, engineers having compensation bigger than their managers is something unheard of in the telco corporate culture, and so is trusting engineers to do planning, without the "benefit" of last-year-plus-five-percent beancounting.
A few months ago I was offered very lousy deals by two different ISPs to help them run their networks. One of them was reselling access out of a cheap T1 bought from a cheap ISP whose network capacity was oversold by about 4000%; the other one had a T3 on order and three existing T1s in place but only had sales for a handful of T1s' worth of bandwidth. Both were run by morons who expected me to work for little trying to get an enterprise to succeed but which had no chance to do so, and I had the impression that I would have little influence on management. At least the ISP at which I started my career payed me little but had good chances of succeeding, was fun and had at least one capable manager at most times. I now earn much more working for a larger corporation, which, while not an ISP, promises to have a lot of fun work in which I am or will be involved.
Those who stay are either young enough to work just for fun, or have other reasons (immigration status, etc) to stay. Of course, incompetents tend to stay till pensions.
Folk, the above is true of more people in this field than you can imagine.
--vadim
PS Although i post from a Sprint accound i do not work or speak Sprint.