On Journalism & Crowdsourcing: the good, the bad, the ugly

Last week the Vancouver Sun (my local paper) launched a laudable experiment. They took all of the campaign finance data from the last round of municipal elections in the Lower Mainland (the Greater Vancouver area in Canada) and posted a significant amount of it on their website. This is exactly the type of thing I’ve been hoping that newspapers would do more of in Canada (much like British newspapers – especially The Guardian – have done). I do think there are some instructive lessons, so here is a brief list of what I think is good, bad and ugly about the experiment.

The Good:

That it is being done at all. For newspapers in Canada to do anything other than simply repackage text that was (or wasn’t) going to end up in the newsprint sadly still counts as innovation here. Seriously, someone should be applauding the Vancouver sun team. I am. I hope you will to. Moreover, enabling people to do some rudimentary searches is interesting – mostly as people will want to see who the biggest donors are. Of course, no surprise to learn that in many cases the biggest donors in municipal elections (developers) give to all the major parties or players… just to cover their bets. Also interesting is that they’ve invited readers to add “If you find something interesting in the database that you want to share with other readers, go to The Sun’s Money & Influence blog at vancouversun.com/influence and post a comment” and is looking for people to sniff out news stories.

While it is great that the Vancouver Sun has compiled this data, it will be interesting to see who, if anyone uses their data. A major barrier here is the social contract between the paper and those it is looking to engage. The paper won’t actually let you access the data – only run basic searches. This is because they don’t want readers running off and doing something interesting with the data on another website. But this constraint also means you can’t visualize it, (for example put it into a spread sheet and graph) or try to analyze it in some interesting ways. Increasingly our world isn’t one where we tell the story in words, we tell is visually with graphs, charts and visuals… that is the real opportunity here.

I know a few people who would love to do something interesting with the data (like John Jensen or Luke Closs), if they could access it. I also understand that the Vancouver Sun wants the discussion to take place on their page. But if you want people to use the data and do something interesting with it, you have to let them access it: that means downloading it or offering up an API (This is what The Guardian, a newspaper that is serious about letting people use their data, does.). What the Sun could have done was distribute it with an attribution license, so that anybody who used the API had to at least link back to The Sun. But I don’t know a single person out there who with or without a license wouldn’t have linked back to the Sun, thanked them, and driven a bunch a traffic to them. Moreover, if The Sun had a more open approach, it could have likely even enlisted people to to data entry on campaign donations in other districts around the province. Instead, many of the pages for this story sit blank. There are few comment but some like these two that are not relevant and the occasional gem like this one). There is also one from John Jensen, open data hackathon regular who has been trying to visualize this data for months but been unable to since typing up all the data has been time consuming.

At the end of the day, if you want readers to create content for you, to sniff out stories and sift through data, you have to trust them, and that means giving them real access. I can imagine that feels scary. But I think it would work out.

The Ugly:

The really ugly part about this story is that the Vancouver Sun needed to do all this data entry in the first place. Since campaigns are legally required to track donations most track them using… MicroSoft Excel. Then, because the province requires that candidates disclose donations the city in which the candidate is running insists that they submit the list of donations in print. Then that form gets scanned and saved as a PDF. If, of course, the province’s campaign finance law’s were changed so as to require you to submit your donations in an electronic format, then all of the data entry the Sun had to do would disappear and suddenly anyone could search and analyze campaign donations. In short, even though this system is suppose to create transparency, we’ve architected it to be opaque. The information is all disclosed, we’ve just ensured that it is very difficult and expensive to sort through. I’m sadly, not confident that the BC Election Task Force is going to change that although I did submit this as a recommendation.

Some Ideas:

1) I’d encourage the Vancouver Sun to make available the database they’ve cobbled together. I think if they did, I know I would be willing to help bring together volunteers to add donation data from more municipalities and to help create some nice visualizations of the data. I also think it would spark a larger discussion both on their website, and elsewhere across the internet (and possibly even other mediums) around the province. This could become a major issue. I even suspect that there would be a number of people at the next open data hackathon who would take this issue up.

2) Less appealing is to scrape the data set off the Vancouver Sun’s website and then do something interesting with it. I would, of course, encourage whoever did that to attribute the good work of the Vancouver Sun, link back to them and even encourage readers to go and participate in their discussion forum.

3 thoughts on “On Journalism & Crowdsourcing: the good, the bad, the ugly

  1. shepsil

    I would agree that what the Vancouver Sun published of the donation info was good for transparency and honest politics, but you have to ask yourself if they did it because of an underlying need to do something positive and altruistic for our democracy or if they did it simply to garnish more readers? After all, it is 3 years before the next election, what they publish now will have little effect on the next election. Anyone going back to the year before this past provincial election by reading CanWest and BlackPress publications will be shocked by the clear tone of the articles written to either favour their choice or not written at all that could have favoured the opposition.Until the Vancouver Sun comes up with clearly unbiased articles and policy, I will continue to ignore most of what they publish. The glimmer of hope is that the Toronto Star may take over their reigns.

    Reply
  2. John Jensen

    Hi David,Thanks for the links. I should point out that I wrote to the Sun to ask for a copy of the database to do further analysis, but was told that was not possible. Perhaps when they have won the lion's share of the website visits, newspaper sales, and advertising revenue that this set of stories will generate they'll reconsider.

    Reply
  3. John Jensen

    Hi David,Thanks for the links. I should point out that I wrote to the Sun to ask for a copy of the database to do further analysis, but was told that was not possible. Perhaps when they have won the lion's share of the website visits, newspaper sales, and advertising revenue that this set of stories will generate they'll reconsider.

    Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s