- From: Gary Ruben <gdr@cataneo.bitstream.com>
- Date: Thu, 22 Aug 1996 09:02:10 -0400
- To: glen@bitstream.com
- CC: www-font@w3.org
Glen, The recent discussions and proposals in the W3C fonts news list about fonts on the Web have raised some very legitimate concerns about protecting the works of font designers and type foundries in the world of global electronic publishing. While each of the proposals has some potential advantages, and disadvantages, in my opinion none of the proposals - GIFs, PixelFonts, Java App/Fonts - really addresses the issue of how to prevent the misappropriation of fonts in already popular and well-established formats, if they are to be used effectively on the Web. As desirable and flexible, and gee-whiz as a new font format might prove to be, publishers will be reluctant to give up their investments in their current font libraries or to simply wait to do their Web-based publishing until a new font format has been worked out and accepted by the industry. To bring the discussion back into the realm of font security, let me make a proposal that focuses on securing *all* fonts, in any current or future format, that is based on available technology and known good encryption and authentication techniques. 1. I believe, very simply, that fonts should not be embedded on a Web page - that is, the font data and fonts programs in any format that can be handled directly by standard font rasterizers, like TrueType or PS Type 1, should not become part of a Web page in any form. Not re-encoded, not encrypted, as a block of binary data, etc. Not in any form. The mechanisms used in HTML, or CSS1 to specify a font for use on the page should specify either a URL reference to a font on the Web server, or a set of parameters that allows the browser to match local fonts to the desired font (you know, like Face/Family, Weight, Style or PANOSE, or other font mapping systems), or the set of parameters that allows the browser to synthesize a font with particular metrics, using any font synthesis technologies they might build in (like Adobe MM, or Chamelion, etc.) ... THAT IS, there can only be a *reference* to a font, *never* the font itself. 2. Having a full URL for the font might tempt some less than upright net denizens to try to get the font directly from its Web server. To prevent the outright advertisement of the location of a font, the font URL reference should be restricted to *only* relative URLs, not fully qualified URLs. 3. The browser itself must enforce a policy that prevents *any* font from being downloaded from a Web server different than the one originating the Web page the font request is in. 4. When the browser downloads a font from a Web server and "installs" it into the operating system (so it can take advange of the regular OS font machinery), the browser *must* use the existing facilities for "private" installation (both Windows and Mac have parameters for installing fonts as private). This way, the fonts are not seen by or available to other applications running on the same box - only the browser will be able to use the font. For UNIX, because there is no standard font mechanisms (except for X BDF fonts), it is likely that the browser, if it implements font support, will have a built in connection to a font-rasterizer or two, and it can control its use of the rasterizer very tightly. 5. If a browser downloads a font and keeps it in its cache, the font *must* be encrypted (with strong encryption, like RSA or PGP). 6. A Web page author should be able to specify that they want secure _transmission_ for a font. The font reference tag(s) in HTML or CSS1 should have an optional attribute "SECURE". When the browser retrieves a SECURE font, it should use the SSL security mechanisms, The server for a secure font must then, of course, be a secure server. Given these precautions, and the cooperation of browser vendors and manufacaturers, I believe that fonts can be safely used on Web pages without fear of casual appropriation, or even moderately determined (and therefore *deliberate*) attempts to lift a font out of its Web page. ==================================================================== Gary Ruben Bitstream Inc. Senior Software Engineer 215 First Street mailto:gdr@cataneo.bitstream.com Cambridge MA 02142-1270 USA -------------------------------------------------------------------- Bitstream does not necessarily endorse any opinions expressed here. (Come to think of it, you might not either. 8-o) ====================================================================
Received on Thursday, 22 August 1996 09:06:15 UTC