As is often the case, you have supplied the answer to a problem that I've posed in print.

In my previous column, I argued that although affordable high-speed Internet links are currently something of a luxury, the Internet's growing importance could mean that in 10 or 20 (or perhaps 50) years, communities with nothing better than 56K dial-up would find themselves severely disadvantaged, potential "e-ghost towns." I pointed out that some technological infrastructure changes, such as highway and railroad placement, damaged some communities at the expense of others (and often created real ghost towns). I also observed that in the case of other infrastructure changes, such as telephones and electricity, the government intervened to ensure universal access.

I suggested that although today high-speed Internet access is clearly more of a "nice-to-have" than a "need-to-have" feature, as time goes on, a community without affordable high-speed Internet access would soon find itself shrinking. Therefore, we should be thinking ahead to policies that would prevent such population losses.

The vast majority of you agreed with my analysis and entertained my potential solution of a government policy that would ensure access for everyone. And many of you related that you live in a large urban area and still can't get cable or DSL because of cable and DSL providers' inefficiencies. Many tales included phrases such as, "If I lived the next block over, I could get it." (One of you pointed out that DSL's cable distance range isn't 18,000' but twice that, 36,000'. Thanks for the correction; I didn't know that!) Some of you disagreed politely, and I had a few interesting exchanges of opinion over the issue.

Not everyone understood that I wasn't suggesting some kind of e-welfare system: the whole idea of policy intervention in infrastructure (e.g., the government's action to ensure telephone and electric coverage) isn't to be nice guys and share the wealth. The idea is this: Economic growth requires people to do the work—the "labor" part of the labor/capital equation—and they need a place to live. We can either spend a relatively small amount to keep already developed areas livable, or we can spend a much larger amount to create communities located close to high-speed bandwidth. I didn't convince all of the dissenters in subsequent email messages, but I appreciate the exchange of ideas. (And, of course, the usual three or four people tried to change my mind by insulting me—but that's what the Delete key's for.)

Some of you, however, told me about a group of technologies that could make high-speed Internet access affordable and possible, even in the most remote boonies: a better kind of satellite Internet link. I've known for years that Hughes offers a small satellite dish for a PC that supports high-speed Internet access, but I didn't (and don't) think it's the answer to the two-way telecommuting infrastructure needs that I've described. The Hughes system uses the satellite only for downlinks—you must still dial up to an ISP to uplink data and information requests. Three firms—Gilat Satellite Networks, WildBlue, and Teledesic LLC—are preparing to offer satellite-based Internet connections that don't require dialing out to an ISP. According to their Web sites, none is up and running yet, but all are worth watching.

Gilat plans a Fall 2000 debut and uses geosynchronous satellites to transmit and receive data. Uplink speeds are "about twice dial-up speed," and downloads are said to be "about 10 times" dial-up speed. WildBlue also plans to use geosynchronous satellites, but will not be ready until late 2001. (Apparently, Gilat has access to satellite bandwidth that's already in the sky, and WildBlue has yet to launch its birds.) WildBlue claims that its speeds will be faster, with 3Mbps downlinks and 400Kbps uplinks.

Because both Gilat and WildBlue use geosynchronous satellites, link speeds will be limited by the speed of light—the 44,000-mile trip to the satellite and back takes 270 milliseconds. And, although a quarter of a second doesn't sound like much, it's extremely irritating for interactive technologies such as voice and might be troublesome for Internet transactions. Teledesic won't be online until at least 2004 according to its Web site, but the company's different approach might eventually make its system the most attractive of the bunch. Teledesic plans to offer up to 64Mbps downlink and 2Mbps uplink speeds and imperceptible transmission delays—by building a system of 288 low-Earth orbit satellites. And, like Gilat and WildBlue, the Teledesic connection will be a persistent 24 x 7 connection—always up and ready.

If you detect a tentative tone to my commentary, you're right. After all, Motorola's Iridium project turned out to be an expensive failure. I'd hate to develop a business plan that depended on a future satellite-based system starting and then staying in business. (And, as an amateur astronomer, I'm not thrilled with the prospect of 288 bright new distractions in the sky. Terrestrial light pollution is bad enough. We don't need more celestial light pollution such as the "Iridium flares" that we've endured for the past few years.) But keep your fingers crossed, rural surfers, and keep your eyes on Gilat, WildBlue, and Teledesic. Soon, we might all be able to "dish up" hot Internet access speeds.