The company certainly has the backing necessary: investment from billionaire Philip Falcone and its own spectrum. However, it needs partners and is rumored to be in talks with Sprint Nextel over a tower sharing deal. Service could begin as soon as the end of 2011, and Best Buy will resell its service beginning sometime next year.

Certainly more wireless competition is something that a lot of wireless watchers will appreciate. However starting a network from scratch is no easy task, and especially one that would be built from the beginning to support a next-generation technology. It is not going to be cheap at all.

That said, with AT&T-Mobile a possibility, having LightSquared could answer concerns of rising costs due to lack of competition. Look at the company’s deal with budget carrier Cricket. Its desire to operate on a wholesale level — they sell to companies, not consumers — means they could viably serve as a way for those small mom and pop wireless carriers to compete with what soon could be the Big Three.

For the time being, I have settled on Virgin Mobile’s Broadband2Go offering (I’ll have a review of it coming in a week or two after I’ve put it through its paces). It’s cheap, the initial cost of startup is not high, and it’s now Mac compatible. But while at Wal-Mart, I was shocked to see Verizon and AT&T are now offering their own prepaid plans. I must have missed their announcements–and it’s kind of surprising to me that those companies be interested in getting into the game.

While Virgin’s model has always essentially been a prepaid one, AT&T and Verizon’s have not. In fact, they’ve built their broadband offerings on the traditional post-paid model. AT&T offers a $35/month 200MB plan and $60/month 5GB one, and Verizon a 250MB plan for $40 and 5GB for $60.

These are one size fits all plans–so either you use the data or you don’t: the cost will always be the same. For many of us, that just doesn’t work, and in an age where penny pinchers are the norm, a two year expense we just can’t justify ($1,440 over the contract for the highest plans).

Virgin offers several levels of service: a $10 plan for 100MB, expiring in 10 days; and a $20 plan for 300MB, $40 for 1GB, and 5GB for $60, all expiring in 30 days.

My use of wireless broadband itself is very irregular. So a pay-as-you-go plan, where you’re essentially paying for what you use, is very attractive. That’s why I was so excited to see Virgin jumpstart this market, but as I now sit here I’ve begun to think, is this the future of mobile broadband overall?

Take for example the net neutrality debate. Much ado has been made of the decision to leave wireless out of any agreement — with ISPs claiming network management is one of the biggest reasons why. Moving to pay as you go would give these ISPs a way to rightfully charge those who are taxing the network the most while more casual users are not overtaxed for data they don’t use.

Moving back to AT&T and Verizon’s decision to cannibalize themselves, this type of thinking may have led them to think doing prepaid was a good move. Instead of limiting themselves to business customers and the techie crowd, they’ve opened up a whole new revenue stream.

Depending on how well it does, it could trigger a shift in the way mobile broadband is offered. One size fits all plans generally only really work for a small subset of subscribers. Instead of overage fees, carriers could allow customers to “top-up” on an as-needed basis.

With the desire for mobile broadband becoming ever greater as society becomes more connected, I think pay as you go will only become more and more ubiquitous. I don’t necessarily think that’s a bad thing, either.

The most meaningful opinions about the shift will come from those of us who are currently paying AT&T $30 a month for unlimited smartphone data. We’ll all get to vote on the change with our pocketbooks–by either choosing to grandfather ourselves into the unlimited plan or opting to switch to one of the cheaper, limited offerings.

So this silly little poll is for current AT&T 3G smartphone (iPhone or otherwise) owners:

As with the PCWorld tests, PCMag’s have AT&T as the clear overall winner, with download speeds that trounce everything but Sprint’s still-not-widely-available 4G network. AT&T didn’t do as well in the “Consistency” tests, however, which would tend to confirm the theory that AT&T’s 3G network is extremely zippy…when it’s working well. Still, I feel like everyone who declares that it’s a well-known fact that AT&T’s network sucks should be forced to read reports like this one.

To steal a meme from Tip O’Neill, all wireless networks are local–what’s most important is that your carrier work well in the areas you need it most, and performance can very literally block by block. PCMag’s San Francisco Bay Area results show AT&T winning for both download speed and consistency. My iPhone works just fine at home, but remains flaky to the point of being unusable–especially for voice calls, which PCMag didn’t test–in much of San Francisco’s SOMA neighborhood. (The story cleverly uses Google Maps to show the exact locations where tests were conducted, but doesn’t break out the results to that level of detail.)

What the world really needs is a way to get wireless speed and reliability data down to the Zip Codes you hang out in. Anyone know if there’s anyone out there trying to collect this data?

The end of flat-rate data has different implications for the iPad than for smartphones, because Apple’s tablet is (among many other things) a cool portable Internet TV. AT&T’s press release on its new pricing tiers gives example usage for the new $25/2GB plan that includes four hours of streaming video, along with Web surfing, e-mail, and other tasks. Plenty for many people, I’m sure, but anyone who planned to feast on ABC, Netflix, and other video sources over 3G–and who isn’t already grandfathered into the old plan or prepared to buy an iPad before Monday–will be out of luck.

Now Virgin has a new plan: 5GB for $60. That happens to be exactly what I pay Verizon Wireless. If I wasn’t only one year into a two-year agreement, Broadband2Go would be awfully tempting…

LTE isn’t a magic bullet, but it’s promising and I look forward to its arrival. Even though absolutely none of the phones and wireless adapters we own now will be compatible with it…

Except, of course, that it isn’t–you’re on the hook for $1200 over two years in broadband charges, and can’t get out of the contract without paying a penalty. On the other hand, you’re likely to pay sixty bucks a month for wireless broadband even if you don’t buy the Compaq netbook, and Sprint charges after-rebate contract prices of up to $99 for EVDO adapters. Maybe the best way to look at the Compaq is this: It’s an EVDO adapter that happens to have a computer attached to it.

Ultimately, the best way to judge most subsidy deals is to ask yourself what the chances are that you’ll still be happily using the subsidized device and the service in question at the end of the contract. If you can imagine yourself carrying a midrange 2009 netbook and using EVDO through mid-2011–and aren’t worried about being joined at the hip with Sprint for 24 months–it wouldn’t be nuts to consider buying the Compaq. Then again, small computers are going to get a lot cooler over the next couple of years, and 4G data will start to matter; I think the Compaq will start to feel a tad stale by late 2010.

One thing’s for sure: The Sprint offer is more tempting than the equivalent ones from AT&T and Verizon, which bring the price down to $199.99 rather than $0.99.

Full disclosure: I used to insist that I’d never sign another contract again, but I’m currently wed to both AT&T (for an iPhone 3GS) and Verizon (for an EVDO adapter) until 2011.