Is it just a coincidence that some of the leading Internet-based application companies are pushing aggressively into network connectivity at exactly the same time the major telephone companies are pushing into content? Or are we witnessing the end of the Internet as we know it?

Think back to the online world fifteen years ago. There was AOL, there was Compuserve, there was Prodigy, and there was Apple’s eWorld. Sure, there were researchers and students posting to Usenet newsgroups and navigating through Gopher sites, but the Internet was a sideshow for individuals and business users. It was a time of great innovation and excitement. Yet the online world of those days was fragmented and small. Every online service was an island.

Are we going back to those days?

As Om Malik has been reporting, Google is putting together a next-generation network infrastructure platform including a national fiber backbone, WiFi wireless networks in cities, and, through its investment in Current Communications, powerline broadband connections to the home. And eBay spent billions to acquire Skype’s VOIP and presence platform. Meanwhile, the telcos are feverishly working to put together multi-channel video services to compete against cable TV operators. They are trying to move up the stack at the same time the other companies are moving down. While all this is happening, Yahoo! is positioning itself as a video arms dealer, and Microsoft is talking about an alliance with AOL.

I don’t quite see the big picture yet, but something tells me the model of the Internet and communications world we’ve followed since the early 1990s may be falling appart.

The concept is that connectivity, applications, and content are distinct technical, business, and regulatory spheres. I’m one of the culprits, as an advocate of the “layered model” for Internet policy. But if you go back to Mary Meeker’s seminal “Internet Report” in 1995 (written, she told me, because so many prospective investors she met on the Netscape IPO road show were completely clueless about the Internet), you’ll see the same pattern: infrastructure businesses (ISPs), application businesses (search, advertising, and e-commerce), and content businesses. The only company that seriously spans all those markets today is AOL Time Warner… and look where it got them.

Soon, though, most of the major Internet players are likely to be hybrids of two or more layers. Google and eBay will be infrastructure and applications; Yahoo! and News Corp. will be applications and content; telephone, wireless, and cable operators will be infrastructure nad content; Microsoft and Time Warner will span all three levels. And that’s just what we’ve seen announced so far. In this market, everyone is in play.

Is this good or bad? Honestly, I’m not entirely sure. I’m not so much of an open platforms bigot to think that rigorous formal separation of network layers is always desirable. GoogleNet and eBay-Skype are incredibly exciting, because they herald a frontal challenge to manifold inefficiencies of legal telecom. On the other hand, there are good reasons to think that the separation of networks, applications, and content helped catalyze the Internet’s extraordinary success over the past decade.

It does sound as though all of us excited about the “Web 2.0” vision of open standards built on top of open standards, facilitating mashups and lightweight innovations all around, might want to question our assumptions. Yes, that is where we have been heading, but it might not be where we ultimately go. If GoogleNet, SkypeBay, MSAOL, Telco Fiberia, and CableLand emerge as competing integrated fiefdoms, we’ll see something more like the early 90’s online services, albeit on a much bigger stage.

As noted, I’m entirely sure what’s happening out there, or what I think of it. I just have the sinking suspicion that it’s more than we appreciate. I welcome your comments.