UPDATE: (May 17, 2012) For all the people that asked for more editing capabilities, I have updated the editor to provide rudimentary cut/copy/paste capabilities via the normal shortcut keys. More to follow as I get the cycles to refine the capabilities.
I’ve just released my mini-tutorial for the HTML5-based XML editor I am developing for drafting legislation (actually it’s best for tagging existing legislation at this point).
Please keep in mind that this editor is still very much in development – so its not fully functional and bug-free at this point. But I do believe in being open and sharing what I am working on. We will be using this editor at our upcoming International Legislation Unhackathons (
) this coming weekend. The editor is available to experiment with at legalhacks.org site.
There are three reason I think that this approach to building editors is important:
The editor uses an open standard for storing legislative data. This is a huge development. The first couple generations of legislative systems were built upon totally proprietary data formats. That meant that all the data was locked into fully custom tools that were built years ago could only be understood by those systems. Those systems were very closed. That last decade was the development of the XML era of legislative tools. This made it possible to use off-the-shelf editors, repositories, and publishing tools. But the XML schemas that everyone used were still largely proprietary and that meant everyone still had to invest millions of dollars in semi-custom tools to produce a workable system. The cost and risk of this type of development still put the effort out of reach of many smaller legislative bodies.
So now we’re moving to a new era, tools based on a common open standard. This makes it possible for an industry of plug-and-play tools to emerge, reducing the cost and risks for everyone. The editor I am showing uses Akoma Ntoso for its information model. While not yet a standard, it’s on a standards track at the OASIS Standards Consortium and has the best chance of emerging as the standard for legal documents.
The editor is built upon open web standards. Today you have several choices when you build a legislative editor. First, you can build a full custom editor. That’s a bad idea in this day and age when there are so many existing editors to build upon. So that leaves you with the choice of building your editor atop a customizable XML editor or customizing the heck out of a word processor. Most XML editors are built with this type of customization in mind. They intrinsically understand XML and are very customizable. But they’re not the easiest tools to master – for either the developer or the end user. Another approach is to use a word processor and bend and distort it into being an XML editor. This is something well beyond the original intent of the word processor and dealing with the mismatch in mission statements for a word processor and a legislative drafting tool leaves open lots of room for issues in the resulting legislation.
There is another problem as well with this approach. When you choose to customize an off-the-shelf application, you have to buy into the API that the tool vendor supplies. Chances are that API is proprietary and you have no guarantee that they won’t change it on a whim. So you end up with a large investment in software built on an application API that could become almost unrecognizable with the next major release. So while you hope that your investment should be good for 10-12 years, you might be in for a nasty surprise at a most inopportune time well before that.
The editor I have built has taken a different approach. It is building upon W3C standards that are being built around HTML5. These APIs are standards, so they won’t change on a whim – they will be very long lived. If you don’t like a vendor and want to change, doing so is trivial. I’m not just saying this. The proof is in the pudding. This editor works on all four major browsers today! This isn’t just something I am planning to support in the future; it is something I already support. Even while the standards are still being refined, this editor already works with all the major browsers. (Opera is lagging behind in support for some of the application APIs I am using.) Can you do that with an application built on top of Microsoft Office? Do you want to switch to Open Office and have an application you built? You’re going to have to rewrite your application.
Cloud-based computing is the future, Sure, this trend has been obvious for years, but the W3C finally recognizes the web-based application as being more than just a sophisticated website. That recognition is going to change computing forever. Whether your cloud is public or private, the future lies in web-based applications. Add to that the looming demands for more transparent government and open systems with facilitate real public participation and it becomes obvious that the era of the desktop application is over. The editor I am building anticipates this future.
I’ve been giving a lot of thought to where this editor can go. As the standards mature, I learn to tame the APIs, and the browsers finish the work remaining for them, it seems that legislative drafting is only the tip of the iceberg for such an approach to XML-based editing. Other XML models such as DITA and XBRL might well be other areas worth exploring.
What do you think? Let me know what ideas you have in this area.