I want to elaborate on the words from last December that I just posted, in response to some comments about how I had just "discovered" the Web. L. McCarthy just had some good points about differing privacy/political goals here on this list. When we first formed, almost three years ago, one of the first things we did at a physical meeting--and the issue was echoed on the new mailing list--was to conduct a poll of who was using what mail tools, e.g., pine, elm, emacs, Microsoft Mail, MCI Mail (?), Lotus Notes, Eudora, etc. The results--which should be in the archives for sometime around November-December 1992--were, as expected, all over the map. No clear winner. The reason for the poll was obvious: to determine what sort of target markets the various PGP integrators would have. Even then, there were religious wars, with, for example, the emacs crowd arguing that PGP should be integrated into emacs and then the world could just switch to emacs (as they should have long before :-)). Now I wish I could draw pictures here---and that I can't draw pictures here and still communicate with most of you here makes an interesting point about the still-dominant nature of ASCII, which ain't about to change anytime soon for lists like this---so I could better explain my "stable attractors" line of reasoning. But I'll do it instead with more words. Here's an elaboration on my points made last December 15: "I see two "stable attractors" for text/graphics/multimedia/etc. sent over the Net:" What I mean by "stable attractors" are the "islands" or regions in product space that have solidity and success. Leading commercial products are obvious examples, with a cloud of related or ancillary products supporting them. Product versions are like a chain of these islands. As with "attractors" in general (and I assume everyone on this list has read much about attractors, usually in the context of strange attractors), there are not many "occupied" regions in the nether-realm between attractors. That is, between the islands lies open water. "Survival" is difficult in these open waters. My main model for software, borne out by everything I see, is that this "island colonization" model is appropriate. "1. Straight text, ASCII, 80 column format. All systems can handle this, all mailers and newsreaders can handle it, it's what the Usenet is essentially based upon, and it gets the job done. It meets the needs of 95% of us for 95% of our needs." By this I mean just what this list is doing _now_. ASCII is the de facto lingua franca. People with PCs, various operating systems on their PCs (DOS, Win3.1, Win95beta, Linux, OS/2, NeXTStep, Solaris, etc.), Macintoshes, terminals, Amigas, Ataris, Suns, SGIs, NeXTs, and so on, are all mostly able to read what is distributed here. Deviations occur, but mostly unintentially or as "experiments." Occasionally people will still try to send NeXT-formatted mail--I forget what it was called--and various people send their messages as "attachments," even when the text is apparently just straight ASCII. (Hint to the attachment-senders: I periodically go into the "Attachments" folder on my system and empty it of the unread big and little attachments that have accumulated in it...others have said they do the same. So, if your message will fit into the standard text/ASCII "primary format" of the Cypherpunks list, that is how you should send it...and if it _isn't _ straight text, but instead includes attached spreadsheets, JPEG movies of the Waco raid, etc., you might ask yourself just how many people will bother to read or view your message?) Hence my comment that "the written word" is a massively stable, heavily colonized "island" or "attractor." It can be handled in foreign languages, with some difficulty, and on nearly any computer system in the world. It is the language of legal briefs, of economic reports, of crop reports, and of a zillion other forms of communication. Pure text is powerful stuff. But what about pictures, illustrations, diagrams? Magazines and books use them widely, so why can't we? And what about styled text, footnotes, superscripts, hypertext links, etc.? Well, in a different world we might have adopted standards earlier than we did and such things might be more common and acceptable today. (For those who will argue that it is "possible" to exchange e-mail with embedded diagrams, equations, footnotes, etc., "sure." But ask yourself how many times you have ever actually _done_ this, with friends and e-mail correspondents? Some who have done this point out that it usually involved folks within corporations who can standardize on the tools and default settings to make this transparent...then they can send richly-formatted stuff without excessive work. And how many other mailing lists, besides Cypherpunks, have such embedded diagrams and illustrations? I'm not talking about the Web here--I'll get to it in my next point--but about what this list is and what NetNews, for example, is. What will "the masses" likely use to implement a richer communications channel, one that encompass pictures, illustrations, movies, spreadsheets, etc.? What will be the _next_ big island people colonize? (Which is of obvious interest for the deployment of crypto to users.)
From Dec. 15:
"2. The Web, for graphics, images, etc. This will be the next main stable attractor, deployed on many platforms. (I'm assuming the debate here about Netscape standards does not imply much of a fragmentation, that Mosaic, Netscape, MacWeb, etc., will all basically be able to display Web pages in much the same way.)" Enough people are starting to "surf the Web" (whatever you think of that expression) that this is becoming the _de facto_ next attractor, or island. _Millions_ of users will have whatever tools and "helper apps" in their versions of Mosaic, Netscape, MacWeb, etc. such that this will be the platform/environment of choice. As the browsers add e-mail (receiving, as most or all can send mail), and as applets/helpers proliferate, then these platforms/environments will allow new forms of e-mail to finally become _widespread_ (note that I did not say "possible," but instead said "widespread"). Many folks I have expressed this view to have said "But the Web is not a two-way medium like e-mail." That is, most people spend most of their time on the Web "reading" (viewing) the stuff others have put on the Web. Three points: 1. This is changing already, as "feedback" is included on pages. This feedback is beginning to look like local newsgroups, and will become more so (IMO). (Speculation: The current Usenet "feed" is of course huge. It may get replaced, via evolution/revolution, by a shift to a Web-oriented system of local newsgroups. What I mean by this is that instead of reading, say, "alt.cypherpunks," one points one's browser at "http://www.cypherpunks.org/" and uses one's various Web tools to browse, sort, search, read, and respond to comments of others. BTW, this could be done today, and might be a better alternative than creating "alt.cypherpunks." The current approach of shipping the entire Usenet feed to all the sites that carry it is likely to eventually break down.) (Even more speculation: Currently I point my Netscape at which news server I wish, from a choice of several. The idea of "subscription-based" News sites is an interesting one. I might pay extra money for a site that is very current and carries all News groups, while parents might pick a site that is sufficiently sanitized for them, a site they let their children access. Much more to say here, but I see several Cypherpunks themes.) 2. The easy-to-use integration of helper apps into Web browsers will confer the same capabilities on mailers that are now associated with these Web browsers. (Again, don't tell me what _your_ mailer can now do, look to what the millions of people are using...they'll gain a lot when their mailers, whether part of Netscape or MacWeb or not, can automatically handle things their browsers can now handle.) 3. The main development seems to me to be in Web tools these days. Being a user of "tin" for several years, and seeing minimal development of it the past two years, I've seen tin get almost no new features. Ditto for "elm," my mailer (when logged-on to Unix systems). (Before you comment, I can't speak for trn, nn, rn, etc., or mailers such as pine. But friends of mine have told me the same stagnation is happening with other mailers and newsreaders. Many of the developers of tin, elm, archie, gopher, etc., have moved on to bigger and better things.)) So, given that it has long been recognized as a valid Cypherpunks goal to see what people are using for mail and newsreading, I think an analysis of what's likely to be popular amongst the "masses" is valid. (I don't disdain the "masses," at least not in this context. The needs of a lawyer wanting to communicate securely with his client are not the needs of a C hacker wanting to configure his Linux box to auto-sign his emacs messages.) My views on the Web have *not* changed dramatically since last fall and winter when I was talking in these terms, though all I had then was a text-oriented browser (lynx), and it was not very exciting (as Ray Cromwell also notes). (And I recall this discussion on the importance of the Web going on several times earlier, including a prediction/hope by the Extropians list organizers, including Harry Shapiro (Hawk) that the Web could be the solution for distributing graphics-content mailing lists...this was around 1993 sometime.) I'm definitely not "dissing" Unix, though I personally never had much use for it. The world is made up of all kinds of people. Some are hackers, some are expert cryptographers, some are lawyers, and so on. The needs of a lawyer for computer tools and writing aids are quite a bit different from those of someone who wants to put together a Linux box for C hacking. If I sound snide in my comments about Linux, I don't mean to be. What I mean is that very, very few users, even fairly sophisticated users, are going to be doing their work on Linux boxes. (If I'm wrong about this, and Linux becomes a serious deployment system--as opposed to a Sun-killer, which is what it looks like now--then I'll acknowledge that I was wrong.) As cheap boxes to deploy remailers and Web sites on, Linux sounds like a win. I'm unconvinced that it has a future for _general_ users, though. (And by general users I don't mean computer-phobic newbies, I mean the folks buying Windows in the tens of millions and Macs in the millions per year. Wider use of crypto means these users, not just the current PGP users.) Nothing has changed my view that the Web is clearly the next big attractor, the next big island. Integrating crypto into it is likely the next big win, which is how this latest thread started. (And by "integrating crypto into it" I don't necessarily mean getting the source code from Netscape or Spry or whomever and adding it...the integration can be done in multiple ways, I think, and as several folks here are already thinking about.) In any case, the future will unfold as it unfolds. Maybe I'm right, maybe not. Maybe only partially right. Debate is healthy, and at least this debate is closer to being on-topic than discussions of red mercury (and the even rarer columbium-niobum alloys the Japanese have developed) and Cypherpunks logos. --Tim May .......................................................................... Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay@sensemedia.net | anonymous networks, digital pseudonyms, zero 408-728-0152 | knowledge, reputations, information markets, Corralitos, CA | black markets, collapse of governments. Higher Power: 2^756839 | Public Key: PGP and MailSafe available. "National borders are just speed bumps on the information superhighway."