In the last Flokk post we arrived at the following summary:
Flokk would be a platform to host protocols and portals that would achieve the following two goals:
Remove the redundant parts of building an application
Drive adoption and engagement across the various protocols and portals
Let's now look at each of these goals in the context of the tools that already exist.
Removing Redundancy
Building a citizen science application requires a few key ingredients: a UI (and all of its styling and design), a backend (with cache, database, API, etc), authentication, security, a data model, somewhere to host, etc. Now if you are familiar with the citizen science applications and software bootstrapping libraries that already exist you'll know that you can cobble together many of these components. For example both iNaturalist and SciStarter have authentication that you can piggy back off of. iNaturalist and GBIF both provide APIs that let you use their data model and backend. Flask-RESTful makes building APIs incredibly easy and Flask has several other plugins for building out other components quickly. If you insert data in well known platforms like iNaturalist and add the right meta data, that data will flow into other portals like Globi. The point is that while all the bits and pieces may be a bit scattered, most of the tools to put together a citizen science application already exist and just need to be cobbled together. Which brings us to the following conclusion:
Catalyzing new application development is not about building new tools but about making it easy to discover and use the tools that already exist.
In other words our goal in Flokk should not waste its time replicating all of the great work that's already been done but make the tools discoverable and, especially for someone new to these kinds of things, accessible.
Driving Discovery and Engagement
In a similar vein to the last section, the central driving observation here is that so many amazing applications (protocols and portals) exist. Flokk should take advantage of this ecosystem, not exist in ignorance of it. Indeed there are already so many great apps built around this ecosystem already. For example iNaturalist has a series of associated applications like Seek, Globi, or Caterpillar Counts that enrich their data, protocols, and experience. Yet, unless you know what to go looking for you'd never find these apps! This underscores the need for a place where all of these various applications and tools can be linked together and made discoverable. If you're working on a project concerning insects within iNaturalist you shouldn't have to stumble across Caterpillar Counts. If you're making multi species observations Globi should be recommended to you. What we need is a marketplace that helps you navigate the vast array of citizen science projects and apps, provides recommendations, and links together apps that synergize well.
Driving discovery and engagement will come from building a marketplace for citizen science tools and applications that links together relevant apps, provides recommendations, and helps you navigate the vast space of projects.
Wonderfully SciStarter is working to do exactly this! The already have and are developing a recommendation engine and have a vast catalog of citizen science applications and projects that they are enriching with metadata meant for enabling navigation. Rather than having to solve this problem ourselves we can simply go support them!
Marketplace Engagement
One challenge that popped into my brain while I was looking at SciStarter was that of keeping people engaged in the marketplace itself. As a very personal example I actually used SciStarter several years ago to find some citizen science projects. Subsequently, because my attention was focused on the projects SciStarter helped me find and not on SciStarter itself, I actually forgot SciStarter existed. I only rediscovered it recently! Keeping people engaged with a marketplace that they use to find things outside of the marketplace that will draw their attention for a long time is a tough problem and one that'll need to be solved in order to allow the marketplace to really drive continuous and thorough adoption and engagement
A central challenge for our marketplace is how to keep people engaged with the marketplace when the marketplace is effectively driving their attention elsewhere.
Pulling it All Together
Okay, let's go ahead and summarize our findings:
We have a series of refined tenets
There are tons of components for quickly building out citizen science apps that already exist. They just happen to be scattered about, so Flokk needs to make them easily discoverable and accessible.
Discovery and engagement will come from building a marketplace that links relevant apps together, provides recommendations, and makes the space of citizen science projects more navigable.
Such a marketplace needs to find a way to keep folks engaged, especially given the purpose of the marketplace is effectively to direct attention elsewhere.
There are already folks working on these kinds of problems that we can work with! (How exciting)
SciStarter is driving after that marketplace.
iNaturalist, GBIF, Flask, etc. have built and are building the components necessary to make building new apps quick and easy.
What's Next
With all that's been said it should be clear that Flokk, above everything else, is about making things discoverable. Whether you're a scientist looking for components to reuse in your citizen science app or you're a citizen looking for new projects and cools ways of viewing your data, Flokk should make finding things super easy. Given this is all about creating that marketplace, and in the spirit of jumping into the currents that already exist, getting involved in SciStarter development seems like an obvious next step.
That being said, you can't guide people on a trail you've never taken, or teach people a skill you've never used. So actually building an app with the aforementioned components is an obvious next step as well. Through the experience of building such an app we'll be able to identify what works well, stumbling blocks, missing components, and how to best guide folks through the process of using these components.
Finally, it is highly unlikely that every component ever needed has already been built. It is also highly likely that the folks in the best position to build these components will be the organizations already hooked into the field like iNaturalist, GBIF, and the like. Therefore becoming involved within these communities is another obvious step forward.
Comments