Our government is supposed to be accessible to everybody. Often that's more of an ideal than a reality, but ideals are there for us to strive for, even if we never fully realize them.
The home computer revolution has raised the possibility of access to a new level. Things that used to take a weekend at the library can be done in a few minutes at home--if somebody out there in cyberspace has taken the time to digitize the information. The possibilities are breathtaking. The realities are still sort of under construction.
Mainframe computers have been a part of our government for decades. Traffic violations, military requisitions, election returns, dog licenses--everything is stashed away so a bureaucrat can type your name into a terminal and get the info faster than you can say "panopticon". While some of this is private (we hope), most of the government's business is a matter of public record.
These public records, being public, have been available to us all, but only in certain buildings, at certain times of day, sometimes only in state capitols or Washington DC itself. The lady with the horn-rimmed glasses behind the counter gets to see the good stuff whenever she wants, because she has a computer terminal hooked into that mainframe, but the rest of us have to schlep down to the county courthouse and start shoveling.
Well, if your garden-variety bureaucrat can browse through Uncle Sam's records with the click of a mouse, why can't the home user? If the government can string up dedicated connections across the continent, why can't they provide a modem or two so people can dial in from home over a regular phone line? Or better yet--why can't they put information on the Internet, where you can offer nationwide access and nobody has to make a long-distance call?
While lots of public records remain stashed away in odd corners, there's a project that's been running for a couple of years now to find out how easily and efficiently public records can be distributed on the Internet. Since January 1993, the Internet Multicasting Service has been working with the Securities and Exchange Commission to put filings by publicly held companies onto the Internet where anyone with net access can get at them. Corporations and mutual funds in the United States are required to file quarterly reports and all sorts of other paperwork with the SEC, so that investors in those companies can get some objective idea about where that organization is going and who's involved with it. Many companies now file their papers electronically, using the SEC's EDGAR system. (EDGAR is a cute acronym for the dishwater-dull designation Electronic Data Gathering, Analysis, and Retrieval. Remember, we're talking bureaucrats here.) These filings are public records, and even if you're not going to jump into the stock market anytime soon, there's information in there that could be useful to all sorts of people--employees, small suppliers, environmental gadflies, newspaper writers...anyway, it's our information, and we the public should be able to have a look at it, right?
So let's have a look at what the home Internet user can do with EDGAR. There are many ways to get to EDGAR, and some work better than others. (All of them are listed at the end of this article.) I found that the easiest way to get a document out of EDGAR was through the World Wide Web, either with the Lynx text-based browser or with Netscape. Here's the address you'll be pointing at:
If you're using a unix shell account with Internet access, at the shell prompt type "lynx" followed by that address. If you're using Netscape, click on "open", type the address in the box and hit enter.
You'll be starting at the EDGAR home page. Click on the "Search EDGAR archives" to get to the actual database. If you're using Netscape, you'll have a box to enter your keywords into. If you're using Lynx, you'll need to hit "s" at this point to bring up the place to enter your search.
Searches are by keywords. Enter a keyword like "honeywell", and you'll get all the entries in EDGAR for Honeywell, Inc., quarterly reports, proxy statements, whatever. I tried "valspar" as well, and turned up lots of stuff on Valspar Corporation. The searches use Boolean operators, too. You can enter "dayton AND hudson" to get reports on Dayton Hudson Department Store Corp. (put the AND in capital letters), but if by chance there was a Dayton Alligator Farms, Inc., down in Florida somewhere, you could try "dayton NOT hudson" to eliminate all those department store filings. The searches don't just scan headers; they go into the text of a document. A search under the keyword "3m" pulled up all the reports for Minnesota Mining and Manufacturing. However, a search under the keyword "norwest" pulled up lots of companies that had Norwest Bank mentioned in their paperwork, but otherwise had nothing to do with Norwest itself. Once you've identified the reports you want, choose "file" and "save" in your menu bar if you're using Netscape. If you're using Lynx, you might just want to capture the text with your modem software as it scrolls across the screen.
The EDGAR home page has other useful links--if you want to know just what the SEC is, or what all those forms are and why they have to be filed, you can click around on the home page and get answers to just about all your questions.
You can get these same documents by ftp, or even by e-mail, if you have limited Internet access. I tried connecting to ftp.town.hall.org and looking around. The help files are right there for you to get and read. Using ftp isn't as straightforward as using the World Wide Web, though. The documents themselves have long, numeric file names that don't mean anything to humans. You have to look inside a text index of all the companies on EDGAR to match names and numbers. This file is easy enough to get. It's in .zip format for DOS/Windows users, .sit format for the Mac, and .Z format for the unix elite, but when you unpack it, the file size balloons to over four megabytes--too big for most Windows users to handle.
Some of the documents themselves are pretty big, too. If you don't have Netscape or Lynx, your best bet is probably to request that these documents be sent to you by e-mail. This way they'll get split into chewable chunks automatically. It wouldn't be a bad idea to check with your system administrator before doing this, to make sure your e-mailbox can handle the sort of volume that's going to be thrown at it.
Admittedly, making use of all this data isn't as easy as getting it. If you're the sort of person whose idea of fun is parsing data into a Lotus spreadsheet, you'll be in hog heaven. It has to help if you've taken some accounting classes at your local community college. Just remember that a financial ratio usually compares the amount of money coming in to the amount of money going out--as you'd expect, it helps if this number is greater than one. The casual browser may not find the makings of a scandal here, but a dogged researcher might trace interlocking ownership among corporations, or see who's doing well and doing struggling.
And again--this is our information. It might not have been: the SEC originally wanted to distribute these electronic filings by selling them to private vendors. Planning for EDGAR started back in 1983, long before anyone thought of using the Internet to get at it. After spending $30 million, much of that in cost overruns, the SEC turned over the rights to wholesale the information in EDGAR to Mead Data Central, the folks who run the pricey Lexis and Nexis on-line information services. Mead, in turn, has been selling the EDGAR info to other information providers for profit.
There are some similarities here to the controversy over West Publishing's case numbering system. The government (our government) makes the law. West does the work of organizing the law. Does that mean that West owns the law? In this case, the SEC does the work of gathering the public records. Mead Data Central does the work of organizing them and making them searchable--and gets compensated for it. Does Mead own the info? Well, they've got the contract to wholesale the info through 1997.
So why is EDGAR being distributed free, while searchable, usable databases of our laws remain proprietary? You can thank the electronic activists who pestered the government into freeing the info--notably Ralph Nader's Taxpayer Assets Project. (Information is an asset too, you know.) And you can thank a new non-profit entity called the Internet Multicasting Service, run by Carl Malamud. In 1993 he got a nice grant from the National Science Foundation to buy the info from Mead and offer it up to the world on the Internet, and his little organization is still running the Internet side of the operations today.
This is also the weak link in the bargain. As anyone who's ever gotten grant money knows, it runs out. Predictions that free EDGAR access would be halted "real soon" have floated around regularly, and the World Wide Web page itself has plenty of disclaimers ("this is a research project", "expect access methods to change without notice"). Yet the Internet Multicasting Service just keeps chugging along. As the NSF money starts running out, New York University, Sun Microsystems, and other companies and groups have been coming in to take over the financial burden. So probably you'll keep hearing the rumors, and you shouldn't be too surprised if you try to get to EDGAR over the net and he isn't there. On the other hand, the rumors said EDGAR wouldn't be on the Internet as of this writing, and those rumors were wrong.
Could EDGAR access be killed in a Capitol Hill smoking lounge? Certainly Mead can't be too happy that thousands of people are getting for free the product they're trying to sell. On the other hand, Mead is still able to sell the full EDGAR feed to high-powered users like investment firms who need the info now, now, now, and are willing to pay very generously for the kind of service only a company like Mead can provide. Funding for the Internet Multicasting Service is starting to come from non-government sources, so it's insulated from Congress' slash-and-burn budget reductions. And didn't Newt Gingrich himself say he wanted to put a computer in every double-wide trailer in America? EDGAR access should be right up his alley.
Meanwhile, by the end of 1996 filings from every publicly-traded company in the United States will be on EDGAR. Even if the Internet Multicasting Service throws in the towel, the Taxpayer Assets Project has a commitment from the SEC to have the EDGAR info put on CD-ROMs and placed in libraries around the country. Either way, it's our info, and we deserve access just as much as the guys in Washington DC or Wall Street.
By e-mail, send mail to:
The body of the letter should be the word HELP, in capital letters, on a line by itself.
By ftp, connect to:
Look in the /edgar directory for the file general.txt; it's got the basic info you need.
By gopher, you can connect to:
And the URL for your World Wide Web browser (Netscape, Mosaic, Lynx, etc.) is: