Recognising the volunteers: Jhempenstall is my hero – who is yours?
I’ve been aware for some time of the National Library’s project for digitising old Australian newspapers.¬†But I only recently read the great story of the project told in this article by Rose Holley (pdf) who was appointed in 2007 to manage the¬†program.
From establishing the project at the beginning of 2007 with no idea about inviting the public in to correct errors in the optical character recognition (OCR) done by machines on contract to the NLA, the project is growing into a fabulously successful venture in which unpaid volunteers from the public play a major role in correcting the errors that fancy OCR software can’t get right (though it’s much improved from an aborted attempt to digitise newspapers in 1996).
Here are some highlights from Rose Holley’s write-up.
- In the first month of use over 200,000 lines of text was corrected in 12,000 articles, by the end of 6¬†months 2 millions lines of text had been corrected in 100,000 articles.
- At no point since release of beta has there been a time when text correction is not taking place. It¬†continues 24 hours a day 7 days a week.
- 78% of users were based in Australia but there was also a growing international community with¬†users in the United Kingdom, United States of America, New Zealand and Canada. One of the¬†top ten correctors was based in USA.
- The top ten text correctors were correcting significantly more text than all other users spending up¬†to 45 hours a week on the activity. The top corrector at the end of 6 months had corrected 101,481¬†lines in 2594 articles. The same correctors remained in the top five for the first 6 months.
- No vandalism of text was detected in 6 months so no roll back to previous versions or moderation¬†was required.
Reading about it, I’m struck by the way in which the NLA stumbled upon the idea. If they’d not got a $10 million allocation for what is undoubtedly a very worthwhile program, would the structures of the public service have been flexible enough, would they have encouraged innovation from the ‘bottom up’ sufficiently to have allowed something like this to have gradually emerged from low level experimentation without some imperative to do the project from above? ¬†I mention it because Wikipedia had been around for a good while before the project got going – so one imagines some people had thought of it somewhere. And what’s the best institutional arrangement to spread the skills that the NLA have acquired with this project. ¬†The NLA itself seems to be keen to spread the value of its accomplishments, posting the code it has developed which seems a great start. ¬†But would housing less of the project within the NLA also be a good move? ¬†Might some more generic unit within government (or perhaps outside it) provide a better way of spreading those skills? ¬†I ask those questions quite naively and in full blown enthusiasm for the achievements of the project, not by way of criticism.
But one thing I want to do here is less¬†tentative and more specific. I wanted to pay tribute to the volunteers without which, quite literally, none of this would be possible. The ten biggest contributors to the project – volunteers from outside that is – contribute more than the nearly 1,300 other volunteers who make their own valuable contributions. ¬†(I’m one of them as of a couple of days ago, but so far I’ve only corrected a few lines!) ¬†Go and sign up yourself!
Naturally I want to pay tribute to those people out of gratitude that they’re serving the public interest. I expect you do too. Like Wikipedians, they do it for a variety of motives. For some of them it just bugs them when they can see an error! ¬†But the fact that what they are doing is of public benefit is a substantial motivator for many if not most. And when asked how the project can be made better, as Holley’s paper makes clear, many of them say things like this.
Recognize achievement ‚Äź Make a point to recognize achievements one‚Äźon‚Äźone and also in group settings. We like to think we are being noticed and are making a difference. Show us how we fit into the big picture.
So that’s what I want to do here. ¬†Here is a table of the top eight contributors at the time Rose Holley published her article.
1 Jhempenstall 101,481 lines corrected, 2594 articles
2 Cmdevine 90,823¬†lines corrected¬†1585¬†articles
3 Fwalker13 80,437¬†lines corrected¬†642¬†articles
4 Mrbh 79,248¬†lines corrected¬†1439¬†articles
5 Maurielyn 72,129¬†lines corrected¬†1192¬†articles
6 John F Hall 59,111¬†lines corrected¬†1632¬†articles
7 Jdickson2 28,796¬†lines corrected¬†2407¬†articles
8 JamesGibney 25,106¬†lines corrected¬†479¬†articles
So good on you all you good people. ¬†Good on you Julie Hempenstall from Bendigo whom the NLA tells me is now up to more than a quarter of a million lines. I’ve seen a list of the current top five and Julie’s held her top spot, with all of the rest having been stayers, who were also in the top eight above. ¬†The least this Taskforce can do is to acknowledge your fantastic work. I think that one thing we (the community) should definitely do is to encourage a culture of recognition and public support and approbation for such efforts.
But of course this new world we’re in of open source endeavour is full of such people making their contribution. I wanted to invite readers to nominate other leaders in other projects who have selflessly volunteered large amounts of their time to build the public goods of Web 2.0 in Australia.
If Julie Hempenstall is my hero, who is yours?