The end of the Word is nigh

Have you written, amended or reviewed a contract today? As you're a Gazette reader, the answer is probably yes. And, because you're reading this on the web, the chances are that you worked on the document electronically. Indeed, if you're above a certain age, you may still be feeling a little smug about embracing keyboards and screens after training on fountain pens: plenty of lawyers of that generation are still around. 

But don't feel too pleased with yourself. Just because a document is held on a computer does not mean it is truly digital. The overwhelming majority of documents exchanged by lawyers today are written in Microsoft Word and presented either in that format - perhaps with the dreaded 'track changes' - or as PDFs. Both programs predate the internet age: Word came out in 1983, in the stone age of personal computing, PDF, originally Adobe's proprietary format, dates from 1992. Both work fine at what they do, but this is essentially to present the words of a document so they can be read by a human being.

In the digital age, this is arguably the least important attribute of a legal document, especially if, once it is agreed, it is filed away and never referred to again. Last week's Smarter Contracts event, organised by the government-backed LawtechUK initiative, heard that this is the fate of 60-73% of all institutional data. Contract documents were likened to a sponge - capable of soaking up data but not of doing anything useful with it. 

In an age where data is supposed to be the 'new oil', this is a manifest waste, causing inefficiency and risk. Of course numerous law tech innovators have spotted this, and have stepped up to the plate with digital systems capable of managing and extracting data from contracts. But these typically require documents in Word of PDF form to be scanned with machine-learning software, which then presents what it detects as the key elements for automated insight and analysis. Such systems make for impressive demonstrations, but conceptually they are the equivalent of translating a document into Chinese and then back into English before it can be understood. Most of the time, they may get it mostly right, but...

An over-riding theme of last week's event, and LawtechUK's well-written and refreshingly jargon-free Smarter Contracts report, was the need to stop pretending that storing information in Word and PDF documents is appropriate for the 21st century. Sir Geoffrey Vos, master of the rolls, did not mince words on the subject, describing them as 'analogue applications'. 

If you don't know what the alternative is, you're already staring at it. The type of mark-up language used to identify different components of a piece of script for a computer program to display as a web page can label all sorts of other information - to take a simple example, the words that comprise the jurisdiction clause of a contract. Data marked up in this way, through so-called schemas, can be unambiguously and automatically extracted for analysis. A UK Legal Schema already exists for this purpose, as an open-source project. The use of mark-up language is also the reason why the National Archives' new court judgments repository is so important. 

Sir Geoffrey Vos, the master of the rolls, is already on record as saying 'the end of the Word is nigh'. (Microsoft, incidentally, doesn't seem to mind: it is basing its offerings to the sector around its Teams collaboration software and the Azure cloud platform rather than 20th century microcomputer code).  At last week's event, he cited the adoption of truly digital documentation as a priority for the sector. 'The UK should, I think, be in the vanguard,' he said.

So let's stop kidding ourselves.