In addressing the question of access and the cable industry, it is critical to identify what we are talking about. The debate is not made simpler by some of the recent rhetoric about the Internet.

Prof. Lawrence Lessig, who has called for cable forced access requirements, recently commented on efforts to protect intellectual property and its effect on technology by asking "Is the information revolution murdered?" Commenting more specifically on the FCC's cable modem service classification decision and notice of rulemaking in March, one critic said the decision "if it stands, makes cable companies God of your Web browser, with the ability to exercise absolute control." The Center for Digital Democracy's director, ceaselessly unhappy with most of what cable does, concluded that the FCC had "struck a deadly blow to the future health of the Internet." And the research director for Consumer Federation of America concurred in the medical diagnosis: "It's bad policy," he said, "because it will destroy the Internet."

Cable, it was and is vigorously asserted by the foregoing speakers, is against "open access" -- two words that, used together, resonate with fairness and democratic values. Cable's formulation in this debate, "forced access," was a decent verbal counterpunch to "open access." But, "forced access" sounds like a spoonful of bad-tasting medicine or Soviet-era sharing of apartments. If I had to march under a banner, it'd be a lot easier to get cheers with "open access" over "forced access."

This policy battle hasn't helped much by the care, as just noted, with which some open access advocates make their case. Partly, that case wasn't made by true believers at all. I refer to GTE, one of the chief funders of "open access" campaigns.

Phone companies have their own platform, of course, and weren't really seeking access as a third-party ISP to cable plant. No matter; telcos provided funding and spokespersons for open access. This behavior reached a nadir of sorts in Broward County, Florida where GTE in 2000 actually paid the legal fees for the county to take on a cable company trying to renew its cable franchise without an open/forced access requirement.

Rhetoric aside, because much of the criticism about cable modem service is not very carefully drawn, it mixes up different players and issues. And, most significantly, the critics fail adequately to consider the consequences of seeking a regulatory solution.

This is a pity, and for a reason that cable's critics might agree with -- that long-term consequences aren't being fully considered. Indeed, this concern with consequences at the early stage of the Internet animates Prof. Lessig (and others) to their activism, even if his solution is the opposite of mine.

We are still early on in the development of Internet communications. The Silicon Valley philosophers were wrong about the Internet changing everything -- unless they were referring to a person's net worth, depending on how long one rode the investment bubble; in that sense, everything did change! But much will change and we'd be turning our backs on the history of every epochal media change -- radio, TV, computers -- to ignore the Internet, and ensuring it launches correctly.

So what do we mean by access in the cable modem context? In fact, the debate is narrower, and I submit, less influential to the big questions about the Internet's future, than it would seem. It means that a customer has a choice of Internet Service Provider (ISP) akin to the choice available in the narrowband world.

Some cable companies have already offered a version of this choice, and more of these business arrangements may be in the offing. But this limited ISP choice is not identical to the arrangements a customer can elect on a dial-up by narrowband connection (although it probably compares to the number of DSL choices in most markets where choice is offered there).

Of course, even with cable's more limited arrangements, a customer that wants a particular ISP can get to that ISP. Cable operators -- or their own third-party ISPs -- don't discount prices if you choose to also subscribe to a second ISP. But that's a cost, not an access-to-features issue.

As the FCC recently put it, "It bears repeating that cable modem service subscribers, by click-through access, may obtain many functions from companies with whom the cable operator has not even a contractual relationship. .... [A] subscriber to [a cable company's] modem service may bypass that company's Web browser, proprietary content and e-mail."

With a narrowband connection, national, regional, and at least in the early days, local ISPs offered service. While there has been consolidation and bankruptcies among ISPs, there are concededly many more ISPs that a narrowband customer can choose from than in the cable modem environment (without having to perform the aforementioned click-through).

There is nothing magnanimous on the phone companies' part about how the diversity of ISPs via narrowband developed. Because an ISP is just another phone number to the connecting local exchange carrier (LEC), the LEC couldn't prevent the springing up of ISPs any more than it could prevent people from calling a fax machine instead of the phone company controlling all fax transmissions. And since many narrowband customers order a second phone line, there's something in it for the LEC anyway, even if it doesn't also get the ISP business of its customer.

Perhaps there will be a social history written of narrowband ISP diversity in the late l990s. Surely among the most famous were the free ISP services, who believed they could survive on a revenue stream comprised of advertising and cross-promotion. But the brief reign of free love, or at least, free ISP love, didn't last too long.

There are other forms of ISP diversity: non-English ISPs, variety in fees, and variances in the degree of customer support. Some appeal to customers by witty names or a very local presence. Others offer more local phone numbers and more modem banks for more certain dial-up connections. And price competition is present too, although not quite so much as during the blissful days of free-everything.

It's those elements of customer choice -- "diversity" in some respects -- that arise from multiple ISPs. But let us be careful with that word, diversity. Diversity of service plans or customer care has nothing to do with content diversity. And this is the very most important point, the point that cable's harshest critics seem to slide over.

Whatever the merits or shortcomings of limited ISP choice with cable modem service, there is no restriction on access to any content on the Internet. If anything, cable modem service permits faster, always-on access to all Internet sites.

As the FCC put it in its recent cable modem classification order, "We are unaware of a single allegation that a cable operator has" denied access to unaffiliated content or relegated unaffiliated content to the slow lane of service. (para.87)

Furthermore, the access cable debate is not, or should not be, about whether residential bandwidth hogs buy residential rate service but use it for business. Business content sellers who want to upload files to customers can contract with cable or telcos -- or FedEx, for that matter -- to transfer material. But there is no denial of access to the Internet, no blow to free flow of information, when a cable operator prevents someone from using residential bandwidth for strictly business purposes. Any kid who's been caught holding the exit door open for his friends to sneak into the movie theatre knows that his delinquency has nothing to do with freedom of speech.

Three other things that this access debate is not about. First, it's not about the cable operator using ownership of the last mile or backbone to protect copyrights that it may own or wish to protect. Cable is not peering into traffic to see whether somebody is sending an unauthorized copy of Lord of the Rings. This area of self-help may some day become a contentious area -- akin to the notice-and-take-down procedures developed with respect to posted infringing or defamatory material. But cable today doesn't operate as a cyberspace copyright sheriff roaming the broadband range. Whether it (or other ISPs) should have responsibilities to copyright owners may be the next big question, but it's by no means a cable-only question.

Second, the debate is not about denying service to so called video-on-demand streamers. In the heady days of 1999, the Internet-only streaming industry promised free original movies and animation, again all supported by advertising of one form or another. There are companies that produce for nonsubscription Internet distribution, but the business model there is still developing. As for those who follow the pay model -- dominated today by the adult sector, with more mainstream films and gaming to follow -- cable applies no specific source-based restriction to such services. If download limits are imposed on any form of on video streaming, it is to allow the shared cable platform actually to be shared - and that goes for any bit-heavy user, whether the streaming video competes with a cable product on the cable system line-up or not.

Finally, the access debate is not about -- has nothing to do with -- the traditional sort of access provided by cable on one-way cable systems. Cable operators agree under some franchises to offer public access, educational, and governmental channels. But at least one franchising authority, Vermont, has required a cable company to furnish bandwidth for PEG-like functions on its broadband service.

This reservation, in the age of browsers, is ludicrous. Any public access Web content can be reached by means of the browser, and fast, too, if reached on cable broadband plant. When it comes to Web content, there is no gatekeeper role of a newspaper editor, broadcast programmer, or cable network picker. Everyone has public access.

I have run through what the access debate is not about because it is important to see how narrow the rest of the discussion really is, in terms of the enduring values of diversity of expression on this medium of communications. Access to content, the end-to-end character of the Internet, Napster and post-Napster debates over copyright limitations -- none of these have to do with the cable access question.

Even the related content questions that do arise with cable modem service, about walled garden and site caching, don't amount to much to be concerned with. AOL sells itself hard and successfully on making the on-line experience fun and easy. Its email and instant messaging services are, of course, classic Internet functions. As to some of its program features, it directs the user to its own, preselected databases. Many customers like this feature of AOL; others prefer to surf the net, on their own or with the help of search engines, to reach their own databases. Nothing that AOL or other on-line ISP providers do prevents net surfing. If there is a wall to this garden, it is a very short wall, and for those who wish, it's easy to jump over it.

As to caching, it's been argued that the late (but not much lamented) @Home network cached pages on some basis other than their popularity to customers. Whether it did, it didn't turn out to be much of a secret to success, did it? Even if a company cached a favored site that is not among the most visited sites (the usual basis for caching), this advantage is unlikely to be significant, given the speed of broadband. There has been no evidence that non-usage based caching policies -- whether by cable, DSL, or portals like Yahoo -- are having a discernable effect on Web site visits.

On the subject of content, finding out how to aggregate audiences to Web pages goes far beyond how many ISPs exist on a platform. It's the Internet's version of the questions, How to Pick Up Girls, How to Pick Up Boys, How to Reverse Baldness, and How to Bake a Fat-Free Extra Cheese Pepperoni Pizza. Lots of approaches, none quite hitting the mark.

In those early days of browsing, when the exclusively on-line service world of Compuserve, Prodigy, and AOL, gave way to the browser, it became very difficult to find a Web site. You'd guess what a URL might be and hope you'd guessed right.

Three years ago, the view was that portals were the key to solving the confusion of hundred of millions of Web sites. Lock a subscriber in with My Yahoo or My Excite and that customer was yours, happy to have the Web organized by the portal.

Alongside portals was the heat among search engines. Alta Vista became the popular engine, only to find Google, MSN, and Ask Jeeves running close behind.

Both portals and search engines favor some sites over others and their rankings of sites likely have the greatest impact on where a customer goes on the Net. They're closest to the issues of content availability and foreclosure. But these aggregators are not, should not, be targets of regulation. At this point, a variety of other aggregation techniques apply, and the business models keep changing. And in the meantime, e-mail Spam and the eternal offers for the miniature digital camera are adding their own bit of mischief and misery. These techniques, tools, and gimmicks may have as much to do with reaching content as anything else. And these influential Internet elements have nothing to do with ISP choice.

To return to an understanding of what the access debate is, and is not about, in the cable context: how should we measure the benefits of trying to duplicate the narrowband ISP choice on cable? As I mentioned, unlimited ISP access could possibly lead to more competition in both customer service and price, although, given the available technology, entry as a narrowband ISP is a far less costly business than providing broadband. Then, too, privacy laws may apply differently to cable versus unaffiliated ISPs, so there might be a different privacy regime available. And for non-English speakers, a foreign language ISP might make a difference in deciding to use broadband. (But this would seem to be easily fixed with a reset to a new, non-English start home page.)

There are pluses, then. But what of the costs? For starters, let's recognize the competition that forced third-party access generates is nearly all nonfacilities based. There is no new last mile being constructed by the third-party ISP. Bear in mind that, after six years of non-facilities based competition in local phone service -- ushered in by the 1996 Telecommunications Act -- there seems to be a consensus building that nonfacilities competition is not the way build a sustainable marketplace. So multiple ISPs provide competition and choice. But it's not true facilities-based competition and choice.

And the costs of a government-imposed scheme would be considerable. Our experience with the history of carrier regulation tells us this. Regulation under Title II of the Communications Act -- the repository of federal carrier law for over three-fourths of a century -- has been characterized in the last 30 years as a march away from access and price regulation.

The 1980s saw deregulation of nondominant carriers under forbearance and moves away from price regulation under social contract and price caps. It would be strange indeed to jettison that learning, switch courses, and start to impose common carrier regulation. This is especially true given that cable has never been subject to common regulation and entered the federal law (both in the 1984 Cable Act and in 2002 under the FCC's information services ruling) as explicitly not a common carrier service. Further, it lacks the historical cost records or separations mechanisms to divide regulated from nonregulated facilities.

Some might say there is a version of carrier "regulation lite" that could be bloodlessly imposed. This is folly. The yoke of access regulation must be teamed with a yoke for price regulation. If parties cannot agree on price, the government must step in and decide it for them. And when it tries, it enters the long, long tunnel of cost-of-service regulation. That's a tunnel that has nearly all the existing common carriers marching, no, stampeding in the other direction.

What are some of the other costs? A new regulatory regime means a recalculation of the business plan for cable. Most of the rebuilds to provide service are completed. But those rebuilds are not set up for mandated third-party access schemes. Implementing a government approved third-party scheme would slow down re-rebuilds and add cost, cost that the third-party ISP would have to bear in some substantial part. Canada has had a regulated third-party access scheme. But the costs to ISPs in part have led to few takers beyond limited experiments.

And a mandated scheme might also impact the effectiveness of DOCSIS 1.1, just as this next generation of modems hits the market. DOCSIS 1.1 allows a cable operator to provide variable speeds for different customers. It will allow the operator to charge a video download with a price that reflects the higher bandwidth it occupies than, say, a routine email and browser account.

The cable industry is on the cusp of implementing this new scheme. A mandated access plan would make DOCSIS 1.1 rollout and debugging that much more difficult in systems. Where operators and ISP affiliates reach consensus in the market, DOCSIS 1.1 implementation issues can be part of the business discussion. Not so when regulation subordinates business arrangements. The result would be the slowdown of the video broadband applications that some say are so necessary to bring the service to its next level of consumer demand.

Finally, a forced rather than market approach to access may complicate customer service, rather than make it better. When parties mutually agree to work together, lines of responsibility of customer care can be the subject of consensus. When partnering is forced, there's a greater likelihood that a party will say a problem resides in the other party's part of the offering. So customer service may decline in a forced access world, even if it appears to be more competitive at first glance.

The FCC's cable broadband rulemaking notice raises this balance between the benefits and costs of forced access. I think a forced access regime imposes far greater costs than benefits. But as this question goes about being decided I think that one determination -- made in the FCC's March decision -- may be the most important of all at this point in the Internet's existence. And that is that this is a national policy and should be made at a national level. Leaving questions of access to a global service like the Internet to local authorities is a mistake. Given the national character of broadband policy, the U.S. is far better off with the federal treatment of this question.

To conclude, as we rethink access, as the Quello Center has asked us to do, let us be sure we are talking about the right issues. It would contrary to the values of the Internet if cable broadband providers started to limit a customer's access to reach the wide content world of the Internet. Vigilance on that issue is important.

But it is also important to separate that issue from the much different issue surrounding government-mandated access for an industry where none has existed before. Having nursed over 200 TV networks that didn't exist 25 years ago into existence, cable has, I submit, a defensible record in creating content diversity in American culture. It led the broadband development and pushed the phone companies to offer their own DSL service. I'd conclude, so far, so good. So for now, leave it be.