First, a look at the numbers.  The ranks of potential online workers are incredibly vast.  Taking a broad view of things: the BlogHerald estimates the number of US blogs at 15 - 30 million; about 4 million users have signed up for Second Life accounts; and eBay claims around 100 million US users.

Of course, these numbers include a very large number of people who do not blog, use Second Life, or sell things on eBay as their primary source of income; and there is doubtless a good bit of overlap among bloggers, Second Life users, and eBay sellers.  Overall, full-time online workers are probably a very small part of the overall labor force.  I think their numbers will grow steadily in the next few years, however.  If we whittle down the numbers of web application users to focus solely on people whose source of income is primarily drawn from online work, we're still left with a pretty sizable group.  60 Minutes claims that 150,000 people have quit their jobs to become full-time eBay sellers.  Karl Rove pegs this number at 700,000.  If Rove is right, then a hypothetical union representing all of the eBay sellers in the country would be among the largest unions in the US.

Organizing online workers is not a trivial task, of course.  There are a number of obstacles:

• Online workers aren't centralized around a small number of physical work locations, so it's not easy to reach out to them in a face-to-face organizing drive.

• Online workers aren't actually employees; if they've taken the trouble to organize a legal status for their work, they are probably owners of small limited liability companies or S-corporations.  Especially among eBay sellers and independent bloggers, they are more akin to small retail store owners than factory workers or janitors.

• You can't count online workers.  There are pretty good ways to estimate the number of active eBay and Second Life users, but the number of bloggers is an extremely slippery number.  All bets are off when counting the number of such people who are doing full-time work online.
  

Taken together, these challenges pose a major problem to organizing a bargaining unit under the National Labor Relations Act.  The process of forming a bargaining unit relies on the number of potential workers in the bargaining unit.  Currently, for a group of workers to force their employer to hold an NLRB election, one-third of the workers must sign union cards; and for that group to be successful in forming a bargaining unit, a majority of the workers must vote yes in the election.  But if you don't know how many workers are in the group to begin with, and they don't actually work for anyone but themselves, the law is more or less meaningless.  I'd have to guess this is the major reason why most traditional unions have largely ignored online workers: there is simply no way of organizing them, in the same way that you'd organize janitors, nurses, or factory workers.

On the other hand, there are a variety of tools which unions could use to help them organize workers online.  First and most obviously, email and contact forms.  Online workers generally have a vested interest in providing customers with a visible, easy method of contacting them, and prospective union organizers can use those tools in the same way customers can.  More than that, web-based applications tend to offer all kinds of functionality which facilitate organizing groups of people.  To take a simple example, consider an organizer wishing to put together a union of MySpace community builders - people like Joe Anthony, who helped build Obama's 160,000-friend page on MySpace.  That organizer could easily put together a MySpace page, add professional community builders as friends on the MySpace page, and post relevant news and updates on the page.  To take a much more elaborate example, consider Zack Exley's fascinating idea about using the tools in World of Warcraft to help organize World of Warcraft workers (h/t to Mike Connery).

Moreover, there are a wide variety of services which online workers need, and which a union could offer.  Most obviously, because online workers tend to be self-employed. a union of online workers could provide the kinds of fringe benefits which a larger employer usually offers: health care, a pension, etc.  The Freelancer's Union has had significant success offering these kinds of benefits to freelance workers in New York City, and I imagine there is plenty of latent demand for similar services in plenty of other cities.

Furthermore, an online workers union would be a powerful advocate against negative "workplace" policies.  Online workers (e.g., eBay sellers) and their workplace (e.g., eBay) have a very strange relationship.  The sellers don't exactly work for eBay, but they do work on eBay's platform.  eBay's policies have a direct and significant impact on their work, but they have very little formal role in counteracting those policies.  Already, there are many examples of user revolts against policies of online communities - perhaps most famously, the anti-Feeds revolt on Facebook, and less caustically, the ongoing discussion on virtual property rights in Second Life.  Most of these revolts have been carried out by ad hoc coalitions of users who were incensed about one particularly acute problem or another.  But there are a wider set of policies which online workers might have potential gripes with, and which they would have a more difficult time organizing against.  An online workers union would be an excellent vehicle for communicating and negotiating online workplace grievances, and could have significant leverage with a web application provider like eBay or Second Life.

Finally, an online workers union could look out for the political interests of online workers.  These interests include net neutrality, intellectual property law like DMCA, and a variety of other regulations.

On balance, I think that there is plenty of good potential for forming an online workers union.  However, I don't think it would be possible to do so within the context of an existing union.  Online and offline workers differ in too many ways; trying to shoe-horn online workers into an existing union would require organizers to stretch too much, and would distract the union from its core mission.

Instead, I think an online workers union would have to take the form of an online membership association.  It wouldn't be organized as a bargaining unit under NLRA, but it could offer a set of protections like those found in traditional unions, and a few more besides.

What I would be most interested in seeing is whether such a union, or association, could still build the ethos of solidarity which traditional unions depend on.  Solidarity - the basic logic of strength in numbers, and unity in the face of powerful enemies - is the fundamental ideology of the labor union, and it's also the reason why so many union members are also progressives.  An online workers union built on this same ethos would be a powerful force for spreading the progressive value system, and bringing more people into our movement.  I am hopeful that it's possible, but I have to acknowledge that online workers, who are essentially in competition with themselves, might not be inclined to accept the concept of solidarity.

Regardless, I'd love to see an enterprising progressive put together such an association, and give it a try.  Online workers need representation, like everyone else.