(Michael S. Hart is director of Project Gutenberg, and was one of the first private citizens on the Internet back in the ARPAnet days.)
This is a copy of an email I sent to Dr. Hart in October of 1996, after reading his A Brief History of the Internet. It appears to be incomplete, but was posted for public download, so I treat it as a first edition rather than a draft. (You can download it by clicking on the title above.)
I have a few comments to make upon: A Brief History of the Internet The Bright Side: The Dark Side by Michael Hart with Max Fuller (C)1995, Released on March 8th. Firstly, is there a more complete edition yet? If so, some of my comments may be irrelevant, so I'll keep them brief until I hear back from you whether you've finished the first complete version of the book yet. Michael Hart may have been the first person who got on as a private individual, not paid by any of the 23 nodes, or the Internet/ARPANet system, for his work; but who at the time of this publication might have given away 25 billion worth of Etexts in return for his free network access. The passage I quote here is very interesting, even tantalizing: a bit of personal and Internet history from someone who was there almost from the beginning. My chief concern in writing this message is that the title of the book is misleading; most of the book is not about the history of the Internet per se, but about where the Internet is likely going and what we can do about it. This is not a bad thing to write about; far from it. But (1) I would like to learn more about the early history of the Internet, especially from persons like yourself, and (2) if you are going to write primarily about the future of the Internet, the title of the book should reflect this. it in their spare time. For $1,000 per book, I am sure a few people would be turning out a book a week for as long as it took to get all million books into electronic text. It is a bit jarring to have you refer to yourself in the third person in some places, and in the first person in others (as in the passage I quoted others). That's about as far as my stylistic comments are going to go; the rest has to do with your argument. Correct me if I'm wrong, or if this impression is inaccurately formed upon an unfinished version of the book, but your main points seem to be these: (1) That the Internet has been free to most users for the first twenty-five years, and ought to be free to the much larger number of users who start using it now and in the future. (2) That there is a conspiracy, or at least a general inclination, on the part of the "information rich" to keep information out of reach of the "information poor." (3) That one of the main tactics of this game of keep-away is promulgation of etext formats which are machine-dependent and expensive of processor time and disk storage space. (4) That, therefore, we ought to see to it that etexts are prepared in the least-machine dependent format ("plain vanilla" ASCII -- a lovely phrase! and a prettier acronym, PVASCII) first, with fancier, more machine-dependent formats created later, as a lower priority. Let me respond to those points, first briefly to each, then at a bit more length. 1. The Internet has been "free" to employees of government, universities, and large corporations as a job perk, and to students as part of the indivisible package of services they buy (or have bought for them) with their tuition. It cannot be free to everyone in the future without the costs being similarly shifted to others. If one grants your assumption that it ought to be free, one must face the question of who will pay for it. 2. I do not doubt that there is some such attitide as you ascribe in the minds of many "information rich." I hope that you've exaggerated it, but I fear you've understated it. But I doubt that it is an organized conspiracy, and I doubt more that it has much to do with the debate about choice of etext formats. 3. and 4. After reading your etext of the Library of Congress etext conference proceedings, I am inclined to think that the main reason machine-dependent, expensive formats are preferred by some people is simply that they preserve more information from the original book than does a PVASCII etext. And it is usually easier to convert from a high-information format to a lower-information format than vice-versa. I would like to argue briefly against the idea that the want of PVASCII etexts is a major cause of illiteracy, or that an abundance of them would be a major factor in reducing it. When one can buy books -- and very good books, too -- for 25c to 75c at thrift stores, situated for the most part in the places where poor people live, the number of PVASCII etexts available to people with low-end computers is probably not a big factor in the rate of illiteracy. If one needs further evidence, look at Usenet. The average *degree* of literacy seems to be pretty low, though everyone has at least the basic level which allows them to misread what others have written and reply in such a way that they're likely to be misunderstood in turn. :) Yet people on Usenet have had some years of access to hundreds of PVASCII etexts, many of them to hundreds more etexts in HTML and other formats, and most of them probably live in places where the libraries and bookstores are of higher grade than they are in the thrift-store neighborhoods I mentioned earlier. But have they profited by their opportunities? No, for the most part. Again, don't get me wrong: I'd not have spend a thousand-odd hours of my life typing and proofreading etexts if I thought it a worthless pursuit. But I don't think the benefit is likely to come in an increase of literacy, but in an increase of opportunities for those who are of lower-middle class (not poor; the truly poor have no access to computers whatever) and already literate at a more than rudimentary level. OK, now some more detail on each point. 1. There's no such thing as a free lunch. Even if all the content on Internet were in the public domain and had been put there by volunteers, there would be costs involved in giving people access to it: costs in electricity, network maintenance, and depreciation of (and provision for replenishment of) infrastructure. Someone must pay for this: if not the users themselves, then who? The only answers I see are (1) people's employers, (2) charitable foundations, (3) the government. As for employers: lots of people are getting Internet access through their employers, and more people will do so in the future; but it's unreasonable to expect every employer to offer full Internet access to every employee. They already provide net access to those employees whom they judge will benefit from it in increased productivity; to require them to provide it to everyone, or all full-time employees, would have an effect like all other benefit requirements -- it would make employing people (at least full-time) a more expensive proposition, and consequently (full-time) employment would diminish. It is the various benefit requirements that the government imposes, at least to a large degree, that are shifting employment from full-time to part-time positions, and making a college degree more and more a necessity: employers can afford to hire only the most productive people in full-time positions. As for charitable foundations: that would be really cool. If you or someone else were to establish a foundation to provide free or subsidized Internet access to the poor, I would give money to it and probably do some volunteer work for it. But it would be unreasonable to expect such organizations to provide access to everyone; they ought probably to focus their limited money and labor on those who need and will benefit most from their help. As for government: I'm not going to spend several pages arguing against such government services on principle, as I could easily do; you're probably working from different assumptions than I, and most of what I said wouldn't have a chance of convincing you. But consider this. If the government, right now, were to establish free access to the Internet as a basic "right" for everyone, those who would benefit most from it, at least in the short run, would be those who have already learned something of the Internet by using it -- that is, the information rich, in your terminology. Even the government would have to limit such "help" to those who need it most desperately, if it were to be cost-effective. Also, there's the question of how it would fund such a program. We could argue all day about where the funding *ought* to come from, if we agreed that it ought to be done. But I can tell you where it probably *would* come from, given political realities. It would be by taking money away from other education programs -- mainly libraries, I imagine -- or by new public debt. As a matter of fact, it would probably be done through existing libraries, by diverting funds from book acquisition to computer terminal acquisition. (See Clifford Stoll, Silicon Snake Oil.) I may not agree with public libraries on principle -- the subscription libraries established by Benjamin Franklin were more just and at least as effective -- but I admit that they're the nearest that government programs ever get to effectiveness and efficiency. Now that I've exhausted the free possibilities, let's look at the low-cost possibilities -- beginning with the one you mentioned and dismissed out-of-hand, Internet Service Providers. most of you probably never heard or saw the word Internet in the media until 1994, with the 25th Anniversay hardly ever mentioned, as the idea was for everyone to think the Internet is the newest thing around, and to get us all to buy tickets for $20-$25 a month. Right now there are 40-50 million people on the Internet-- and if someone could figure out how to make them all pay a $20-$25 fee. . .that would be $100 million a month or over a billion dollars a year. Do you think that the Internet would be provided free if Internet service providers were not charging $20-25 per month for it? Scarcely. If there were fewer Internet Service Providers, there would be less Internet service provided. To most people who have high-end computers capable of using the World Wide Web and so forth, $20-25 per month is not an unreasonable price to pay for such access. But for those with lower-end computers, there exist lower-cost Internet service options. There are BBSs which provide text-only access to the Internet for $20-40 per *year*. The sysops who run these BBSs, like the Internet Service Providers, are entrepreneurs trying to make a profit, just with a different target market... like Henry Ford selling low-cost cars to the poor while others sold expensive cars to the wealthy. All these people -- the local ISPs, the big online service corporations, and the BBS sysops -- are providing useful services for a reasonable price. (The ones that aren't charging a reasonable price will soon be out of business.) We ought to thank them, not vilify them for not giving access away for free. 2. I've said all I plan to about this earlier on. 3. and 4. The people who prefer etexts consisting of high-resolution images of every page of a paper book are scholars who are interested in preserving all possible information about rare books, against the eventuality that the originals may someday be lost. They're also concerned about cost: human labor, needed to produce a PVASCII etext, is more costly than the disk space needed to store high-res images of every page in a book -- for a university library, if not for a volunteer organization like Gutenberg. As for the controversy you relate about whether the Gutenberg etext of "Hamlet" ought to read, "To be or not to be" "To be, or not to be" "To be; or not to be" "To be: or not to be" "To be--or not to be" -- it seems to me that this was more due to simple pedanticism than any conspiracy against the information poor. Don't misunderstand me; I favor PVASCII etexts myself, and that's the format I've used for all the etexts I've produced. But I can definitely see the reason in the university librarians' arguments. A PVASCII text can eventually be produced from a set of page images; the reverse is not true. Suppose that some major disaster were to occur: a meteor strike, a nuclear war, or something of the kind... it might be that the only copies of certain books remaining would be etexts, all the originals being destroyed along with university libraries in big cities. To future scholars, it would make all the difference in the world whether those were page-images or PVASCII. Ordinary people who enjoy good literature of the past would be all right in the latter case; but the scholars would be forlorn forever. And, when the scholars of today are spending money that was donated to a university library, it is more or less reasonable for them to produce etexts that will benefit scholars of the future, as well as ordinary people. Once the page-image or HTML is produced, it will be, at worst, no more difficult to make a PVASCII etext from the high-detail etext than from a paper book. But, again, the reverse is not true. You may have seen grants totalling ONE BILLION DOLLARS to create "Electronic Libraries;" what you haven't seen is a single "Electronic Book" released into the Public Domain, in any form for you to use, from any one of these. Why don't you see huge electronic libraries available for download from the Internet? Why are the most famous universities in the world working on electronic libraries and you can't read the books? This is serious. I shouldn't be surprised if much of the money is being eaten up by the inevitable administrative overhead that occurs in academic organizations. A good deal more may be being "wasted" in creating editions which preserve more information about the original paper books than ordinary readers will ever need. This is probably where your accusations are most accurate, and have the best chance of sticking -- at least, if you identify who and what you're talking about, instead of vaguely speaking of "the most famous universities in the world" and "grants totalling ONE BILLION DOLLARS". What universities? Which grants, granted under what circumstances for what stated purposes? If the grants were explicitly supposed to be for making free- circulation electronic libraries, and the libraries are being hoarded within single academic institutions, this is not only wrong but maybe legally prosecutable. But if they were granted by alumni to build up the prestige and academic resources of their alma mater -- such hoarding may be selfish, but it is not strictly unjust. Still, more publicity might shame these people into giving copies of their libraries away for the sake of the reputaton -- which, after all, is the primary capital of a major university. You are half done writing a good argument here, but it wants more specific facts to be as valid as it could be. I've written a lot more than I meant to; I'll stop here. I look forward to your reply.
To the top of this page.
Back to the Book Reviews page.
Back to my home page.
Email me: Jim Henry.
Get a GoStats hit counter