The Open Knowledge Canada (OKC) Local Open Data Census is a crowdsourced measure of the current state of access to a select number of datasets in cities and municipalities across Canada. The 2017 census is initially completed by local governments that publish open data, and the submissions are reviewed by members of the Canada’s open data community. Community members may also make the first submission. Anyone is welcome to submit comments or questions about the census.
Open Knowledge International (OKI) began a national level “Open Data Index” globally in 2013 with the goal of providing a benchmark to measure the openness of specific government data around the world (see <http://index.okfn.org/faq/>).
Subsequently, OKI has facilitated the development of Local (city level) Open Data Censuses. Open Knowledge Canada initiated the local version in June 2016 in collaboration with OKI, Open North and Open Data Institute (ODI) Toronto. The Open Data community contributed feedback to determining the top 15 datasets used in the current version. The project is led by OK Canada Ambassador, Jury Konga who may be reached at email@example.com.
Appreciation is extended to Mor Rubinstein and Oscar Montiel of OKI for their assistance and to Jean-Noe Landry and his team at Open North, Bianca Wylie of ODI Toronto and special thanks to Andi Argast, ODI Toronto for her help in getting this launched. We have additional cities that have opened up their data - welcome to the community! This year we are also launching two additional "local" census for regional government and provincial/territorial government.
The census is built on two units — places, which represent local governments, and datasets. In this year’s census we have 20 datasets representing different governmental themes. Each dataset has a description, which can be seen by clicking the link for that dataset. For more information about each dataset, please refer to the methodology page.
Each dataset is first evaluated by the contributor, usually someone at the local government level, by answering the census’ nine questions. For the 2016 census, the first submission is accepted without initial review by census reviewers. After the initial submissions, the data is sent for review to the Open Knowledge Canada census review team. Please contact us if you’re interested in being on the review team.
For the current census, you can contribute in different ways:
Enter a new submission for 2016 when there is no data on a specific dataset in the municipality:
You’ll be asked a series of questions about the dataset you have located to help determine how it is scored. There is a help text prompt next to each question in the “information” column.
Leave a comment. If there is already data in a category where you have more updated information, you can propose an revision by selecting ‘Update’ under the dataset. See Leaving a Comment for more details.
Below, you will find tips for each specific question in the census.
If you choose “yes”, you will be asked to enter the data publisher, a title and a brief description and then you will be guided through the remainder of the questions.
If you choose “no”, your submission will be recorded as such. It is very important to know the data does not exist, so please be sure to indicate so rather than leave the entry blank if you have made an honest effort to look for it but can not locate the dataset.
How do I know if the data exist?
Most governmental data can be found on the local government data portal. Try to search these portals first. If you can’t find the data portal, please contact the local municipality. If you can’t find the specific dataset on their website, email the relevant department and ask them about the dataset and whether it’s online. Still can’t find the data or you didn’t get an answer from the government? Try one last time by using your favourite search engine.
Choose ‘yes’ if the data exists in any digital format, even if it can’t be accessed on the Internet. Data can be digital, but not accessible online. If you choose “no” you’ll notice a couple questions greyed out because if the data isn’t digital, by default it is not online, it is not machine readable, nor is it available in bulk. Move on to the next question.
How do I know if the data is in digital format?
If the data exists, but only on paper, it’s not digital! If you found the data on the Internet, it’s definitely digital, even if it’s just scanned versions of paper documents. Some data might be in digital format on a private government network, but not available publicly on the Internet. If you are aware that the data is digital somewhere (for instance, if a government official tells you so), then mark this one “yes” and add a note about how you acquired that information and any relevant contact details or links.
Choose “yes” if the data is made available to the public in any format without restrictions. If you choose “no” you’ll notice a couple questions will ‘disappear’ because that implies that the data is not available for free, is not online, is not openly licensed, nor available in bulk.
How do I know if the data is publicly available?
If you need a password or some other form of permissions to access the data it’s not publicly available. If the data is only available in paper form without any restrictions on the number of copies you can make, it’s publicly available. If there are limits on photocopying, it’s not considered publicly available. If you need to make a freedom of information (FOI) request to access the data, it is not publicly available. If the data is only available to government officials and not citizens, it is not publicly available.
Choose “yes” if the data is available without any cost. Choose “no” if there is any cost involved in accessing the data. You’ll notice that the openly licensed question is greyed out because being available at no cost is a key provision of the Open Definition and open licensing.
Choose “yes” if the data is available on the Internet and you will be prompted to add the url that links to it.
Choose “no” if the data is not available anywhere on the Internet.
How do I know if the data is available online?
If the data is publicly available (see above) and can be freely accessed on the Internet, it is available online. If the data is available in digital format, but not available on a public website, it is not considered to be available online.
Choose “yes” if the data is in a format that can be easily processed by a computer. Choose “no” if the file format can not be easily processed by a computer.
How do I know if the data is in a machine readable format?
The easiest way to answer this question is to look at the dataset’s file type. As a rule of thumb the following file types are machine readable: .XLS .CSV .JSON .XML The following formats are NOT machine readable: .HTML .PDF .DOC .GIF .JPEG .PPT
Choose “yes” if the entire dataset can be downloaded at once. Choose “no” if you can not access the database in its entirety.
How do I know if the data is available in bulk?
If you aren’t able to download a single file that contains the entirety of the dataset you are looking for, it is not available in bulk. Often times governments will provide access to their data through an online interface. If access is restricted to querying a web form and retrieving a only a subset of results at a time from a very large database, the data is considered to not be available in bulk.
Choose “yes” if the data is licensed in a way that conforms to the [Open Definition] (http://opendefinition.org/).
Choose “no” if the data is protected under a license that does not conform to the Open Definition.
How can I find the licensing information?
Usually, a license or Terms & Conditions can be found at the bottom of the website (in the footer) or under the site’s “About” section. If the site has a search function or a sitemap, those are good places to look as well. If there is no visible license or the license is simply under the country's name (for example — “Copyrighted under the state of Lebanon”) and there are no terms and condition or any other information on the site, the data is not open and you should answer “no”.
How do I know if the data is openly licensed?
In order for data to be openly licensed, it needs to be free to use, reuse, and free to redistribute. The Open Definition website lists the licenses that are certified open. If Creative Commons (CC) licences are used then the data is generally openly licensed. If the non-commercial CC licenses (the ones with “NC” or “ND” in their names) are used, then the data is NOT considered openly licensed. These are partially open, but not fully open according to The Open Definition. Sometimes organizations do not make use of Creative Commons licensing, but the terms and conditions do allow use, re-use and distribution.
Choose “yes” if the data is relevant and complete for the year or time period that it claims to represent.
Choose “no” if the data is outdated or otherwise not representative of the stated or a reasonable time period.
How do I know if the data is timely and up to date?
Check the date-stamp on the data (see below if that’s not obvious). It’s 2016, if the data doesn’t seem relevant for the current year, mark a “no”. It’s important to remember that not all datasets need to be updated with the same frequency. Transportation data can be updated on a daily basis while postal codes might not change for many years. Do your best to determine what is reasonable for a given dataset. Does the data align with how your local government works in a particular area? If budgets are determined yearly, there should be yearly data, if they’re determined every two years, then a two year period for the data would be considered timely and up to date.
How to find when the data was published?
If the dataset was found on a government portal, there will most often be a timestamp attached to it. If the data was found on a government site, sometimes the date will be written next to the dataset link or a release date might be listed in some related content (like a press release or news clip). Sometimes the date stamp is within the dataset itself. For example, a tab in a spreadsheet that is named for the date it represents. You can also download the data and find the creation date of the file. This might not represent the right date but give you some useful clues. Use your best judgement and leave thorough comments about your assessment. Lastly, sometimes there are no timestamps at all. In that case, it might be most fair to mark it not timely or up-to-date.
The census allows only one submission per one dataset. However, you can still help by commenting on a current submission and propose changes. Leaving detailed notes in the comment field goes a long way in supporting work on the Census during the review process.
Here’s some tips for leaving good comments:
You can correct a census submission by clicking “Update” under any of the dataset, entering the revised information, and then submitting your correction. Please include a comment explaining what you have corrected.