The new spate of RIAA lawsuits against MP3 sharers got me thinking again about digital copyright infringement. It's been sort of an article of faith in the copyright world that ever-tightening copyright rules (most recently, the DMCA) are necessary because of ever-advancing copying technology. Back in the days before the printing press, you'd need a horde of medieval scribes to make a copy of a book, so there wasn't much need for copyright. As advances in copying technology have developed, so too has the need for increased copyright protection for authors. Nowadays, it is said, computers can make you a bit-for-bit digital copy of a protected work, which can in turn be perfectly recopied without degradation from one generation to the next, so authors need more legal protection against widespread infringement. That, at least, is what I take to be the accepted wisdom.

Should it matter to this analysis that, in the real world, there is no such thing as a perfect digital copy? It occurred to me that what all those kids who just got sued are trading isn't a copy of the actual Red Book CDDA audio content that RIAA members sell on their CDs. Rather, it's a compressed version of that content that, when played back, gives you a close-enough, but always imperfect and never exact, facsimile of the original audio content.

Ripping a CD Digital Audio file and converting it to MP3 form can turn a large file into a small one, which makes for a faster download over the network. But it does so by throwing away large chunks of the original data. How much gets thrown away is up to the encoder, who can tweak the encoding bitrate up or down, and thereby adjust the precise size-versus-quality tradeoff. But even at the highest quality settings, there is a lot of audio information that was present in the original CDDA data that is no longer there in the MP3. Depending on how good a listener's ears are, they might or might not be able to hear the difference between the original or the compressed version, but the differences are there nonetheless. Because MP3 is a lossy compression scheme, it by definition omits some, or even most, of the copyrighted original CDDA data. (Lossless compression schemes also exist, and are much prized among audiophiles, but they achieve far less compression of the original CDDA data, and probably represent an insignificant amount of P2P traffic.) (Even if the original CDDA data isn't compressed, but merely copied from one CD to another, the copy will in most cases not be a bit-for-bit duplicate of the original, due to differences in CD drives and recording media. But let's stick to the MP3 example for the time being.)

An MP3 file sounds like the copyrighted content, but is not in fact the copyrighted content. Should this matter? Is there a level of copy degradation at which we stop believing that the original author's rights are being infringed? Perhaps the answer is "no"---for instance, if I make a photocopy of Stephen King's new book, why should it matter whether I use a new digital photocopier or the old streaky unit with a worn-out sensor and low toner? But nobody would argue that that's the kind of "perfect" digital copy that we are worried about in the computer age, and ordinary copyright law should be adequate to deal with infringement of that sort. Rather, the argument seems to be that authors need additional legal protection in the computer age precisely because of the availability of bit-for-bit nondegradeable digital copies. If the digital copies aren't perfect, but rather omit some (or most) of the original copyrighted source material, is the justification for the heightened legal protection still present? This sounds like an argumentative, rhetorical question, but it's intended sincerely. Might make a good topic for further research, if it hasn't already been done elsewhere.

One item for further reading---

• Jonathan Zittrain, The Copyright Cage , Legal Affairs, July/Aug. 2003 (explaining widespread distaste for current copyright law among cyberlaw professors).

I recently completed drafting a law review article to be submitted later this month for possible publication. I wrote the first one-third or so using ordinary text editors and LaTeX, but switched over to MS Word after that because many law reviews seem to require (or at least, strongly encourage) the submission of documents in Word format.

Although I find it marginally easier to do my actual drafting in a WYSIWYG editor like Word, I could not help regretting that changing over from LaTeX to Word involved some unwanted costs. First, LaTeX generates more attractive pages than Word, free from awkwardly placed line breaks and so forth. Not such a big deal in itself; people have gotten so accustomed to poor page layout from contemporary word processors that they likely wouldn't even notice the LaTeX difference. But the way LaTeX automates the page layout process made it a trivial matter to switch layouts "on the fly"---I could print my draft in 12-point double-spaced to share with a reviewer, then do a much shorter version in 10-point text with two columns for my own use, just by changing a single word in the preamble. That would be a much more complicated ordeal in Word and the process probably could not be satisfactorily automated.

The other big issue for me was information loss. I used pdfLaTeX to generate PDF files with embedded hyperlinks to the court decisions I was citing (at least where they were available online, but all Supreme Court decisions and most recent circuit court opinions are). I lost those hyperlinks in the Word conversion. I could have added them back in, but Word doesn't handle hyperlinks as well as LaTeX. In particular, Word doesn't distinguish between on-screen presentation of a document and printed presentation. If I added a hyperlink to a Word document, Word would underline it and format it in blue to make visually apparent that it was a hyperlink, but that formatting isn't desirable when printing. If I wanted to print the document, I could edit the style sheet so that hyperlinks showed up in ordinary, black, non-underlined type. But then there would be no visual cue on-screen to let me know that any given passage was actually a clickable link. Having to constantly edit the style sheet to switch back and forth according to whether I wanted to see the document on-screen or in a hard copy is pointless effort.

The experience got me thinking again about whether it would be possible to adopt some higher-level markup language to my needs. What I'd like is to be able to draft in some form of markup language, either with authoring software or in plain text. Easy conversion to (X)HTML (for online display, with embedded hyperlinks), PDF (for printing, probably via LaTeX), or Word (for submission for publication, probably via RTF or some other known intermediary format since Word is a closed file spec) is a must. Is there anything like that in use in the legal community?

Markup schemas based on SGML or XML are the obvious candidates. There are well-developed suites of tools to convert from SGML and XML to any of the other electronic formats discussed above, and because of their essentially unlimited extensibility, SGML or XML would make superior archival formats. The Oasis Cover Pages site includes a listing of known academic applications of SGML and XML, although I don't see anything specific to legal academic publishing on their list. But I haven't browsed the complete listing, and maybe it would be worth taking a little time to do so.

Legal XML is an umbrella project that seems to be organizing efforts to develop XML schemas for representing a variety of types of legal documents, so it may be something to consider. But its concern seems to be mostly about evidentiary records like contracts, transcripts, and on government records like court pleadings and legislation. There's currently nothing there that is aimed at scholarly writings (or, for that matter, court decisions as distinguished from the underlying evidentiary record on which they rest).

The other idea I've had so far is to use DocBook or perhaps something based on it. Pluses: it has been in existence for more than 10 years and has been widely used, especially in computer science. I like the idea of adopting a standard from a discipline that has been worrying about stuff like this longer than lawyers have; there is no need to reinvent the wheel. It comes in both SGML and XML variants, which have minor differences in syntax, but nothing major. (Those differences could be avoided in any case by using a dedicated authoring program instead of doing my own markup in a text editor---I need to do more research into what SGML/XML authoring software is out there. Conglomerate looks like an attractive option, and includes a DocBook mode, although it's still in alpha). There are tools to convert DocBook-marked-up documents to the other formats I'm most interested in. Minuses: DocBook is a much more complex tool than is needed for this particular job. It includes tons of predefined tags, the great bulk of which would never be needed in legal publishing. Its origins in the CS community are clearly reflected in the tags defined in the full DTD---there are a horde of tags just for displaying computer programs, GUIs, mathematical equations (via MathML), and the like. So the great bulk of DocBook would be superfluous at best---and at worst, confusing and counterproductive---for use in legal scholarship. The maintainers seem to be generally aware of this, however; there is a Simplified DocBook variant that omits most of the excess tags. May be worth looking into in more detail.

Further inquiry needed: are there other similar systems out there that are more general in focus than DocBook, something that would allow relatively easy markup of documents intended for legal scholarship, possibly via an SGML/XML authoring tool?

Bibliography -- a few items I found while taking advantage of the free weekly Westlaw training access:

• Maria Perez Crist, The E-Brief: Legal Writing for an Online World, 33 N.M. L. Rev. 49 (2003).
• William A. Fenwick & Robert D. Brownstone, Electronic Filing: What is it? What are its implications?, 19 Santa Clara Computer & High Tech L.J. 181 (2002).
• Rogelio Lasso, From the Paper Chase to the Digital Chase: Technology and the Challenge of Teaching 21st Century Law Students, 43 Santa Clara L. Rev. 1 (2002).
• Edward L. Rubin, Computer Languages as Networks and Power Structures: Governing the Development of XML, 53 SMU L. Rev. 1447 (2000).
• Winchell "Todd" Vincent, III, Legal XML and Standards for the Legal Industry, 53 SMU L. Rev. 1395 (2000).
• Hugh Calkins, Legal XML---What Does it Mean to You?, 17 Maine B.J. 210 (2002).