Tomorrow’s Texts

From time to time we let our columnists here at Slaw take a break, if they’ve been good… and if Your Editor has something he’d like to say and wants the column inches for himself. Herewith, then, the first of the very occasional “ed hoc” columns.

You can’t manage a WordPress blog for four years and not think a fair bit about electronic publication. But it’s surprising how easy it is to get caught up in day-to-day matters and to let the functionality that working on the web makes available slip beneath your awareness. And what marvels these are! Lists drop down when you hover a cursor over a heading; text leaps instantly from place to place with a mouse click; any of a billion images (literally) might grace the document you’re reading; text too small for failing eyesight? make it any size you wish with a couple of mouse clicks; dynamic charts reveal graphically the import of the latest data; copies of what you’re reading can be transferred to your mobile phone or to a network of friends anywhere in the world…

I forget, too, that these wonders, commonplace as they have become, aren’t universally available — that there are still spheres where a click gains you nothing, or, at least, a whole lot less than it might. I’m not talking about print, marvellous in its own way, or even about the boundary between print and digital — something discussed with a lot of interest and skill recently here on Slaw, thanks to a great post by John Gregory. I’m referring to the practice of law and the online material that we use in that practice every day — plain old judgments, journal articles, textbooks, etc.

Perhaps it’s because it is so truly wonderful these texts are available online at all that we don’t stop to think often enough about how they might be better structured, about what we’d truly like our legal materials to do online — how the legal article of the future should be structured, in effect.

You read Slaw, and so you’ll know that this observation of mine was prompted by an announcement from Elsevier that they’re experimenting with the “article of the future” for their science publications. It’s a bit unnerving to see such a good problem posed by one of the world’s two fat publishing companies, even if their vision is a bit humdrum. It’s not been our experience that these publishers’ legal branches have been so willing to open discussions.

But I’d suggest that the unadventurous nature of the big legal publishers is for once a Good Thing. The relative quiet on the commercial front presents an opportunity for the legal profession to set standards for “future legal texts” while the issue is still fluid. We can develop a list of document desiderata, some of which might quickly be met by a few publishers and organizations, and the rest of which could guide technical development generally.

So what do we want in online legal documents? How wildly should we imagine this textual future?

For me, the answer to the latter question is dead easy: we should be as bold as our imaginations will let us be. And we should work to stoke the fires of our imaginations as much as we can, because our ordinariness of vision is if not a professional deformation then at least a common handicap.

After all, we know that technical changes will inexorably come, whether we ask for them or not: which of us begged for a replacement for Word Perfect 4.1? which of us demanded Twitter? Though there’s no sensible prospect of controlling this sort of change, even in the law, it is a good idea to try to guide it somewhat in order to get a little of what makes sense to us.

Yes, but what do we want in online legal documents?

As I said at the outset, odd as it may seem in an online publisher, I’ve only just begun to think about this issue in broad terms. I have some small ideas. For instance, the paragraph should be the basic unit of text within a document (much as it is now) and each paragraph should have an “anchor” so that you can link directly to that paragraph. All documents should be legible on small screens, i.e. mobile phones. There should be a Canada-wide system for according every online legal document a unique URI or uniform resource identifier so that we don’t have to depend on fragile URLs to identify it. In documents over a certain length, tables of contents should be available anywhere in the document as popups with a set keyboard shortcut.

See how limited in vision these are? Small, indeed. So how do you see things? What in your wildest (legal) dreams would the online legal document of the future look like?


  1. We have lawmakers (both legislators and courts) making norm statements which are then fleshed out by either supplementary norm statements or explanation. Think statutes, regulations and first reading statements. We have courts providing a logic of justification by reference back to earlier decisions.

    Yet in each case, the formal structure of the law text is set by accidents of legal history. The outline of what a case should look like was set when the nominate law reporters were systematized by the ICLR in the 1860s.

    The structure of legislation is slightly older.

    Yet there is nothing inevitable about how we present the law through generic norms or by individualized decision-making. Think what court decisions look like in Italy or France and how other countries manage the balance between legislation, regulation, policy statement and travaux préparatoires.

    As domains of knowledge legal texts are extraordinarily cross-referential (even if the excess of a Harvard Law Review footnote is baroque)- ideally suited for hypertext.

    But it seems extraordinary that fifteen years after the explosion of HTML into our daily consciousness that the makers of legal texts use such traditional formats and structures.

    Layered texts are the future of the law. We may have to go back to considering Jay David Bolter’s prescient book from 1991, Writing Space: The Computer, Hypertext, and the History of Writing.

  2. The most important characteristic of the future legal texts that I hope for and dream about is that they should be marked up in such a way as to facilitate syndication. Some people say that “RSS” means “really simple syndication”, and that’s the syndication I’m talking about. The text should be marked up with XML in such a way that it can be parsed and extracted from according to taste with appropriate software. It mean that people can write blog readers, and that people can do things like this if they’re so inclined. The possibilities are endless. Of course, RSS provides a good illustration of the challenges too. There are many competing standards.

    For law, one initiative is OASIS Legal XML. In Europe, there is LEXML.

    I’m not sure who, in Canada, is doing serious work on this. Maybe this discussion will draw out some comments. One thing that I do know is that, looking only at the surface of current web publishing, we have a long, long way still to go. Retrieve a case on Lexis and Westlaw and you see frames. View the source of individual frames, and it’s pretty ugly. There are, of course, many reasons to believe that both Westlaw and Lexis have invested heavily in good markup for their internals. It would be interesting to know if they have plans ever to provide consumer products that could be syndicated in conjunction with SQL or Xquery.

    I believe the good people at CanLII would be working on this if they had the funds. The impression I have gotten from talking to people, however, is that the necessary upgrades to their markup are currently well beyond their means.

    Moving forward, it would help if the courts and legislatures were creating appropriately marked up XML at the source. This would involve thinking about the innards of their wordprocessing software, i.e. things like ODF and OOXML. It would be great to hear from anybody involved with a Canadian court or legislature who is working on this.