The switch (ATM, Frame Relay, IP switch, etc.) market is like any technology equipment market - once the market sorts out how much it'll pay for product, the dollars per product pretty much level out. Things get better, but the actual dollars service providers will pay for the equipment pretty stays in the same range.
At present, ATM is more bark then bite. TCP/IP has won the war to the desktop, and frame relay seems to have become the data-link layer protocol of choice. Both UUNET, PSI, and a number of other backbone IP providers use Cascade 9000 switches to move data along their backbones (and a number of companies are now beginning to sell private frame services along the same infrastructure, thereby competing with companies like WilTel, MCI, AT&T WorldSpan, etc. in the frame market). This is probably a good thing, since the general feeling is that the IP market is not going to pay the bills for more then a year or so. ATM, despite the hype, is more a trunking protocol for intra-network transport, and although it promises to mux data, voice, videao, etc., few are really using it for this purpose, and TCP/IP is already doing that to the desktop. Afterall, what good is a technology that muxes these various data types, if it ultimately must be de-muxed and delivered to the office (and the desktop) on a bunch of dedicated strands of single-minded copper.
So how does PSI fit into all this? Well, they own physical network (in OSI speak, not really physical, but data-link layer, and that is about as 'physical' as it comes in the IP world). This is good. They have a bunch of switches deployed, so the potential is there to do a lot of cool things with them, and this is also good. Just like BBN, management is about as confused as Gil Amelio (and I hereby coin a new adverb: 'amilioly', as in, PSI manages their business amilioly). So, as a company, a good buy they might be (I'd need to look at the numbers a little closer), but as a long term investment, I'd wait and see who steps in to fill the management void and then decide. I am repeating myself here, but if an RBOC buys them, jump ship. Those guys still can't figure it all out. However, if a strong, tightly run telco or communication company steps in (read: MFSish), then hold tight.
ANyhow, just a thought... |
