More ambitious proposals for IBM's network division

If you're an avid reader of this column (and who isn't?), you'll recall that during the summer I urged IBM to spin off part of its Networking Hardware Division (NHD) with the goal of re-energizing NHD and making it a more powerful player. My logic went like this: While IBM's NHD brings in plenty of revenue - at least that of 3Com or Cisco - it doesn't have the market muscle or visibility of either of those companies. That's because NHD is serving two masters. It must support its huge base of SNA customers while at the same time competing against the 3Coms and Ciscos of the world in non-SNA - read IP-centric - networking. By spinning off the SNA products portion of NHD into a separate company - stealing a page from AT&T's spinoff of Lucent - IBM could achieve a couple of goals. First, the spinoff company would live and breathe SNA networking, offering customers new solutions for bringing mainframes into this intranet world. Second, NHD would be free to focus more intently on IP networking. Well, IBM CEO Lou Gerstner hasn't taken my advice: NHD stands undivested. Undaunted, however, I will plow ahead in offering recommendations.

If splitting up NHD isn't to IBM's taste, now would be a good time for the company to use some of its ample cash reserves to take advantage of M&A opportunities that would strengthen NHD.

With a market cap of just over $2 billion (based on a share price south of $14 as I write this), Cabletron is an appealing takeover candidate. The acquisition of Cabletron - which just bolstered its own product line and distribution with the buyout of Digital's network business - would give IBM additional strength in key areas such as switching and net management.

Cabletron's stock has been battered by some bad corporate news and turbulence in the technology sector. But the company is solid and IBM should consider the opportunity before Cabletron makes a big buyout of its own or a competitor such as acquisition-minded Lucent or current partner Nortel snaps up Cabletron.

If Gerstner wants to try something smaller, how about grabbing up one of the remaining Gigabit Ethernet companies - perhaps Foundry Networks, an early leader? That type of deal would put IBM in the forefront of this hot market and give it a competitive edge against Cisco, which has struggled to bring out its own Gigabit Ethernet offerings. Either deal would continue to flesh out Gerstner's vision of network computing - a vision that has already been firmly grounded on the software side by the acquisitions of Tivoli and Lotus. And here's another plus. I won't seek any advisory fees for my advice. Promise.