I've put together a website, LawTop, that uses Google News to bring you the most recent law-related stories from Canadian mainstream English-language news sources. You can choose whether to see 10, 25 or 50 of the latest stories, and whether you'd like the headlines only or the headlines and a brief excerpt. As I say on the main page, this is really a simple exercise because Google does all the heavy lifting — and the only reason you'd want to use LawTop instead of Google News directly is because LawTop's got a handy and consistent set of terms to search with.
This is a baby toe in the water for my more ambitious venture, Lexmix, an umbrella that can, and over time I hope will, link together various new devices to separate signal from noise for lawyers on the internet. On this score, don't hesitate to let me know if you have ideas about what legal workers would like to see, when it comes to filters and aggregators.
On LawTop I've provided for RSS and email subscription. But the data are most fresh when a particular LawTop page is first loaded or is refreshed, because that runs the search anew.
Please feel free to let me have your views, either by way of comments or by email. I'm not aiming for perfection here — the search algorithm would be immensely long — and neither am I looking to draw items from every town's paper. But there's no doubt that obvious mainstream news sources have been missed and that the search terms can be improved with simple changes that have eluded me.
Good for you Simon! This is a great way to filter things down. I suspect that most Canadian law bloggers could get good value from this site.
I'm wondering if beefing up the number of searches for distinctly Canadian names (firms, courts, etc) would give you a more Canadian flavour? If you're removing duplicates prior to publishing, then it might widen the net & boost cancon?
Also, what about offering an RSS feed for your re-mix? I can see some additional value in allowing others to take your feed and filter further for specific subjects.
Thanks for the ideas, Steve. I'm already working on ways to increase Cancon, as you say. I'm daunted by the thought of thousands and thousands of firm names – which wouldn't do the trick, alas, anyway. I think I'll have some better luck with datelines. A "context machine" would be a fine thing, wouldn't it? and it may come when the semantic web is out of its infancy. It'd be great to filter for "nation: Canada" and "subject: law".
There's an RSS feed now, but I gather that's not what you mean. If I do a remix of various streams of results, I'll have a feed for that, too, of course.
Nice work, Simon. I notice the Creative Commons designation indicates this is not for commercial use. It would be nice to have it available for commercial use so that firms can pull the feed onto their intranets.
I see you are using Feedburner to feed the RSS mix to the page. I was avoiding Feedburner because Google seemed to be changing access to Feedburner around, but I guess they are maintaining this service? A little tutorial on how you did this would have great value, I think.
Cheers,
Connie
Thanks, Connie. I need to think more about the copyright licensing thing: after all, Google and the news sources are doing the work. But… I suppose if Google can take headlines and excerpts, I can, too. I'm quite happy to have my added value used for whatever purpose, with attribution. So, as I say, I'll have to think a bit.
The tut on how I did it will have to wait until I'm a bit further along with the "upgrade" in the works. I'm trying to use Pipes now as well in order to filter for Canadian content. We'll see what comes of it.
Bravo, Simon! Great work — you are definitely onto something here. Lawyers need filtered search more than most — if you can come up with an easily user-customizable filter, you'll have quite the legal news service on your hands.