7 Tips To Get Your Open Data Project Started
There is so much open data available - whether it’s on a national level, state level, or local level. With so much data available, it can be daunting to figure out what to do with it. How can you actually accomplish something that is both important to you and important to your community? It’s a tough question.
At this past week’s Open Data Delaware Meetup we kicked off our 2017 Open Data Challenge (look out for way more info coming soon!) by helping to give people a place to start when looking at starting their own open data project (full slide deck here). These aren’t hard and fast steps, but they’re a start. They certainly don’t need to be followed in order. Take a look below and then reach out to us to let us know - what project are you going to start?
1. Look through the available datasets
Sometimes the best way to get started with an open data project is to simply explore a little bit. Delaware just launched it’s own Open Data Portal, so that’s a great place to start. There are all kinds of great things to look through there - like State Employee Credit Card Transactions or Baby Names.
2. Find datasets that you want to work with
Eventually, something is going to pop out at you that seems really interesting. Looking through the datasets that are available, it would be interesting to see what someone could do with the Professional and Occupational Licensing dataset and the Disciplinary Actions for Professional and Occupational Licenses dataset. Perhaps something else would inspire you, though.
3. What could you build with what is there? Would it be useful?
This is where you should start taking a deeper look at what is actually in the datasets available. This can really help you to figure out what you could build. Let’s talk about the Professional and Occupational Licensing dataset first. This document gives a ton of great information:
- Last Name
- First Name
- Combined Name
- License Number
- License Type
- Issue Date
- Expiration Date
- Whether there had been a disciplinary action or not
- License status
Now let’s talk about the Disciplinary Actions for Professional and Occupational Licenses dataset. This also has lots of great information:
- License Number
- Disciplinary action taken
- Discipline start date
- Discipline end date
There’s a lot of potential for this data. One quick idea would be to build a portal that lets you look up disciplinary action records of licensed professionals (this already exists - nice job Delaware!). That would definitely be useful, so it’s something that should definitely be pursued.
4. What other data would you need? Don’t see it - suggest it!
Sometimes (read: all the time) you’ll run into moments where you need to pair the data that you want to work with and another dataset. In our example, it might be great to pair licensure discipline information with the location where the discipline took place (mapping is a really great way to help visualize data). Looking through Delaware’s Open Data Portal, this information doesn’t seem to be available. In that case, it’s possible to suggest a dataset. That doesn’t mean that the data you’re looking for will become available right away, but it helps to make it possible in the future. For your project, you might need to re-evaluate how important that dataset actually is to accomplishing your goal. It’s okay to not have every feature that you want in your first iteration of the project. You can always add on more features later.
5. Find people interested in working with the same data
Projects need team members. You might be able to do this alone, but generally more people working on a project means that you can start to churn out better, more creative ideas. Generally speaking, we recommend working with teams of 4 to 5 people. It keeps your group small and manageable. You can find these people anywhere, though, being biased, we recommend an open data Meetup - like ours. It might also be useful to go into communities that are actively dealing with the issue that your project concerns. These groups will typically have a lot of passion about the topic and will have great insight into how the data could best be used. For example, if you’re interested in working on an education project, then go find a group of parents, teachers, or students who might want to work with you.
6. What skills do you need to make the project real?
Every project has different needs. Take a few minutes to map out what skillsets would be needed to make the project a reality. Does the team that you’ve started to assemble have the right tools? Perhaps you need to add another person to help round out the group. Often, you’ll want to have someone who has some experience with front end web design. You’ll also probably want someone who is relatively familiar with the topic that you’re tackling.
7. Who could lead the project?
The last part is probably the most important step. Any project that will get completed needs to have a leader. Because you’re the person that thought of the project in the first place, that probably means you. However, it doesn’t have to be. Whoever plans to take on the role of the leader in your group needs to be prepared to wrangle cats. Oftentimes, open data projects are side projects. They’re volunteer work that people want to contribute to, but that means life can sometimes get in the way. The project leader’s role is to keep the team together with clear goals. You organize times to meet and tasks for people to take on. You’re the person with the overall vision for where the project should go. Projects without this key person are much less likely to meet their goals.