ILC response to critique

While the ILC refrained from responding to my e-mail, I was surprised to see a public response to my blog post by the ILC Secretariat on the GRAIN blog, farmlandgrab.org (who in contrast did a nicely researched job with their own dataset). Albeit a receptive reaction by the ILC, I feel that they are skirting too much around the issue at hand: Misrepresentation of the data’s quality. Let’s see if the ILC responds to my reaction.

Reaction of the ILC to the blog post:

Posted on 04 May 2012

Thank you for your specific comments on deals in the land matrix. This is exactly the kind of feedback we are hoping to receive with its launch into the public realm. We realise that we have not been clear enough in the website on the specific categories of deals. Our database includes concluded deals, as well as “work in progress” deals in various stages of completeness that may or may not come to fruition. These are included in the overall database as an indication of the level of interest in acquiring land, even where such deals have until now been unsuccessful. We will make sure to verify deals you mention and flag them as “failed” based on conclusive data. We will also modify the public interface to make this a separate and more distinct category. We also appreciate your pointing out several errors. Although we have put much effort into sorting and checking the data that has been collected by a virtual team over the past two years, the lack of transparency and frequent changes in the status of deals mean that errors are inevitable. Our decision to make the existing data public is not a claim that it is all verified, but a strategy to use crowdsourcing to contribute to the continual updating and improvement of data. Your comments are therefore extremely useful, and will be taken into account in the ongoing revision of the data.

Michael Taylor, International Land Coalition Secretariat

Reaction of Rural Modernity to the ILC response:

Dear Michael,

I appreciate your humble reaction, though remain concerned about how the dataset is presented by the ILC. For example, it claims that “This public component of the database includes only those deals with a reliability ranking of >1, hence all figures on the website only refer to this subset.” I won’t bother you with another expose, but a large proportion of the data entries do not meet the basic criteria of the corresponding reliability codes (as your heavy reliance on the deeply flawed GLP illustrates). So, in some way you are claiming the data is reliable, while it obviously is not – which I feel constitutes misrepresentation. I would urge to make very specific in that case that the data does not meet any reliability standards or remove those entries that do not meet the criteria of code 1 and above. Moreover, as you claim that this is merely to open up the data to crowdsourcing, I am curious to know why a technical report, based on this flawed data, has been published prior to having opened up the data to public scrutiny? As you have no doubt noticed, the media have already started widely quoting these aggregated figures. I feel, as many colleagues would tend to agree, that there is a huge moral hazard in presenting so publicly such figures without having carried out the necessary fact-checking. Additionally, I have made a large number comments on the accuracy of specific deals within the Land Matrix. Up to now, none of these have been published. Effective crowdsourcing should enable uncensored public input. Lastly, it should not have to be proven that a project doesn’t exist – it should be proven that it does. That’s like saying fairies exist, until proven otherwise.

 

Rural Modernity