Am I the only one who finds it redundant to have both the first two digits and the second part of the org identifiers say the same thing?

aa-bbbb-cccccccccccc is a sample structure of the org ids. Why would AA refer to where the registering agent is located and then the bbbb be the reference to said registration agent rather than have aa refer to where the organization is and the bbbb-be the registration agency. Who cares where the registration agency is. Isn't it more important to know where the organization is? This would also be consistent with which actually spells out the aa-bbbb for you before providing a link the the registration site where the cccccccccc can be found.

I get that we have some with historical XI-IATI and XM-DAC codes and I'm not saying we should change them. But I do think going forward, having XI (which means nothing really) be replaced with the country of the organization IATI is giving a code to makes more sense. The same with XM-DAC for codes which are listed there. Why would we use the country code of the government then DAC to indicate that is the place where the code comes from and then the actual DAC code.

Just a thought I'm putting out there as I've spent more time than I care to admit looking for codes and always end up with more grey hair after the fact.

Herman van Loon
Herman van Loon

Hi Michelle,
I agree that the identifiers could maybe be better structured. I rather hesitate to start this discussion again because if this leads to changes in the standard, lets say IATI 3.0, the consequences for hundreds of publishers would be huge because all changed IATI identifiers, wherever the are published should be changed. I do not think the benefit is great enough to justify such a major change. It will cost enormous amounts of time and energy from everyone and the gains seem not great.

Michelle Levesque
Michelle Levesque

Herman van Loon I agree to a certain point. Happy to keep this on the radar screen for a 3.0 but why couldn't the geographical part (the aa- in my example) be improved on a go-forward basis. No need to go backwards. But if IATI is issuing a number going forward they could simply stop with the XI and start with the 2-digit ISO code for country of the organization? The identifiers are all over the place as it is for the same organization so at some point we all need to bite the bullet and clean things up. The less of a mess me make going forward the less there will be to clean up when the time comes.

I also wonder if the major lifting in cleaning is programming by each publisher or rather having a single source of what the numbers are beyond simply the IATI publisher list. Perhaps a project for the Data Use Task Force funds?

Yohanna  Loucheur
Yohanna Loucheur

I agree with you Michelle on the identifiers that are issued by IATI. These identifiers are a problem anyway: the IATI Secretariat is not a registration agency and sooner or later, the existence of these identifiers will come back and bite the organisations using them. Including a geographic reference to them would at least give them some value from a user perspective, a more immediate sense of where the organisation is located.
Regarding the XM-DAC identifiers, nobody is forced to use them. However, as far as I know, very few governments have a well-governed list of identifiers for their public organisations. We did it for Canada and it's way more complicated than one would imagine... Whereas the DAC identifiers already exist, they're already used for DAC reporting, so it made sense for the donor agencies to use them. As long as aid data continues to be published in a silo (rather than becoming part of a broader open data ecosystem), there's no imperative to move away from these identifiers.
TL;DR I'm afraid you'll continue to end up with more grey hair.

Michelle Levesque
Michelle Levesque

Yohanna Loucheur To be clear, I have no issue with the DAC reference. It is the use of XM before it that makes no sense. Why not use LU-DAC for the Luxembourg government rather than XM-DAC. While the DAC code itself does number the countries (22 for Luxembourg), the numbering scheme is not a commonly known identifier. The 2-digit ISO codes are.

As for the lack of interoperability of the code between standards or even cleaning up the mess we have within IATI today, I still firmly believe that there needs to be a tool that summarizes all of the codes (not just publishers) used for participating organizations so 1) we can see if more than one code is being used for the same participating org and 2) if a publisher needs a code they have a place to get what is already being used rather than make one up. Even if the code isn't 100% in accordance with a standard at least we have consistency and not 5 codes for one donor as we did (and perhaps still do) for GIZ.

And it is very much appreciated that Canada did number their government ministries and make that code list publicly available for use. We recently pointed one of our donor contacts to it as they were unaware that they had their own number. It helped as we are often able to get the org reference IDs into our donor agreements. Ours is always stated so there is no ambiguity but that in itself was a process that took time to work though legal and templates.

Please log in or sign up to comment.