The recent news that IPSec VPN vendor NetScreen is buying Secure Sockets Layer remote access vendor Neoteris has two key implications.

First, it's a pretty strong signal that SSL VPNs are here to stay, and the market is starting to take significant steps toward maturity.

Second, it is pretty apparent that not all the current SSL remote access companies, particularly the start-ups, will be around in a couple of years.

On the first point, this is the classic progression of a technology. Several start-ups have a good idea, develop it, start to sell it, grab attention and make larger companies with broader product lines want a piece of the action too. Then four or five of the start-ups get bought by the bigger players and most of the remaining start-ups wither.

Recall just a few years ago when IPSec VPNs were just coming into vogue. There were more than 30 VPN vendors then. Today, major networking vendors dominate the field and there are a few, NetScreen among them, that focus on VPNs. But even these relative specialists are adding other areas of expertise, such as SSL remote access.

So this is a predictable pattern.

On the second point - that many of these start-ups won't survive - it seems probable that the SSL VPN shakeout will be faster than it was with IPSec VPNs. Money is tighter now than it was in 1996. When it becomes apparent that there are too many SSL VPN companies for the number of customers, the next round of venture money will dry up.

They key here, is what does this mean for customers looking to buy some of this equipment? Several customers of these SSL start-ups had a pretty simple answer: if you need it buy it and don't worry too much about whether the company folds.

The emphasis is on whether you need it badly enough. If the equipment solves a knotty problem at a significantly lower cost than other options - and that includes the ongoing cost of administration - buy it. Your problem will be solved for now, the ongoing cost will be minimal and even if the vendor goes out of business, the equipment might continue to perform well for years.

When it dies, someone will still be selling SSL remote access gear that will be a fine replacement. The key is to buy a product now that is solid technologically.

Tim Greene is senior editor at Network World.