What is Siftr? A tool for collaborative mapping

Siftr is a new(ish) tool by Field Day Lab—best known for their role developing ARIS—in Madison, WI. In this post I’ll go into what siftr is, what it might be for, and how it fits into the space of mapping utilities (a bit).

If you are familiar with ARIS, then siftr is easy to describe. It is a spin-off of the ARIS notebook. If that doesn’t mean anything to you, let’s try this: Siftr is a very easy to use tool for collaborative mapping.

Let’s say you want to mark all the fruit trees in the neighborhood, or all the ADA violations on campus. Siftr makes it super simple for you to set that up and then let either a small group or the public generally use their smartphones to go out and document geographically relevant material. It is also a tool that you can use, down the road, to view and discuss the data as a group on a larger display together.

One area of application is citizen science. There are many citizen science projects that wish to crowdsource data gathering. It has been one of the coolest aspects of the internet that large numbers of people can not only find out about scientific topics, but participate in the creation of new scientific knowledge without needing first to become PhD’s. The example that sticks in my head the best is that of Galaxy Zoo, where scientists created a platform for amateur astronomers to help identify interstellar objects. Michael Nielsen tells their story well in his book Reinventing Discovery. Not only was the crowd able to help identify objects of interest more quickly and accurately than a small number of scientists who were officially running the projects or computers, their curiosity led them to document and discover brand new astronomical phenomena.

Like ARIS then, and a lot of the tools that excite me most, siftr takes an interesting idea and turns it into a potential vernacular. The folks who set up Galaxy Zoo used skills and resources that are not commonly available to create their website. While it is not “rocket science” it’s a whole lot more than a typical teacher or layperson could put together. And other citizen science projects tend to be similar in this regard. They do involve everyday people, but only as unskilled labor, not in the design, planning, or upper levels of execution. The GalaxyZoo’ers who picked out the green pea galaxies could not on their own argue for their originality. But something else entirely can happen when the means of creating this sort of rich, interactive media can be opened up to almost anybody. That’s what siftr does for on-the-scene collaborative mapping.

Also like ARIS, siftr is not limited to this one application area, but instead is a general purpose tool that should be able to find diverse applications because there are so many ways it might help bring together people, ideas, and places.

The competition – Where siftr fits in

Unlike AR game design platforms—the genre ARIS contributes to—in the world of custom mapping, there are other big players in the space. Google Maps and GIS come to mind. While both have their places, neither of them yet allow you to:

• Be a normal person and use the software
• Upload while out in the field

In addition, both Google Maps and GIS concern only about the objective side of mapping. To mark what is there. But so often, trying to understand what is going on in a place or with an issue, working on a map becomes about more than placing a pin. Conversations at particular waypoints matter as much or more than the fact that the point is marked. Siftr is build with these conversations, and the emerging understandings they represent, in mind. Each waypoint created in siftr has a comment thread attached, permitting conversation right where you are.

Let’s take some time to to think about user experience and who it is that you might be involving in the use of these tools. GIS, the standard software that scientists and industry rely on to understand physical place, is designed to deal with and accurately represent data. It is rather a nightmare to learn to use and look at, especially for novices.

GIS is a serious mapping tool, but its depth is accompanied by a complex UI

It is confusing and garish. You wouldn’t wish it upon someone unless they really need the functionality the tool provides. These problems of course become invisible to those professionals who do use it over time, and the way the visualizations feel matters less than the information they represent. They need what the tool does and so they get used to it and take for granted that the difficulties in learning and using the tool are necessary hurdles to joining the profession. While this sort of attitude dominated a good deal of 20th century technologically driven work, in many other sectors we have long sought to put a different sort of ideal into place regarding technology. By developing interfaces that are more intuitive and improving access in other ways, the ability of high technologies to be used by people to make changes in their lives and this world has become much more widespread. Technology becomes a vernacular.

So while GIS is a deep tool with a lot of functionality, learning how to use it is a matter of building up a specific technical skill. It is hard to simultaneously learn about place and how to think about place. The time spent doing this background learning can only be justified as a bearable cost for larger enterprises, not something anyone can pick up for a bit. GIS is not for more casual mapping and so its appropriateness is limited to situations that can support deeper investments.

The same is not true of Google Maps. It was designed to be simple and easy, something anyone anywhere should be able to pick up without distracting too much from whatever use they are putting their maps to. But Maps seems to err too far in the other direction for uses that are not as simple as their interface. While Google Maps is easy to use, the part that is easy to use is very limited in terms of the data it can contain. A marker can have an image, title, description and that’s it. One can also draw figures, directions, and measure distance. There is some visual customization possible of the underlying map. These are enough for many uses. But if you need something a bit more involved, there’s not much you can do to extend these capabilities unless you start getting into special file formats and programming as with GIS.

Maps is also built for a scenario where one author creates a map for an audience to use at a later time. To create a scenario where you are hoping to ask others out in the field to help you gradually build up the information you want in the map, or where a map is a contingent, evolving conversation, it doesn’t really fit so well. Other tools are needed.

Google Maps has a simple UI for simple features

This isn’t a lot of information about Google Maps or  GIS, but it should begin to illustrate why mapping requires many tools. There is a trade-off between depth of data organized and the complexity of UI. And different use scenarios, where in a situation the map is created and how it is used, who gets to author and when, these factors matter a lot for how useful a given tool is. Anyone hoping to work with maps should think about these sorts of trade-offs as they look for useful tools.

Siftr sits somewhere in between these two tools and others like them. It has also been developed in a world where mobile data collection is a given. Now that we have some background on the genre, I’ll describe what siftr is and how it might be used.

How siftr works

Siftr is very simple for both authors, mappers, and viewers. Let’s take a brief look at each activity.

Authoring a siftr

Authors create an account (or use their ARIS account credentials- nice!) at siftr.org. Then, the author creates a

• Name for the siftr
• URL
• Description of what to do with the siftr

then

• Customizes the categories of collection and the colors that represent these categories
• Decides basic permissions regarding viewing and contributing to the map

and

That’s it. If you know what you want, it takes less than a minute to get a siftr up and running. It’s fast and easy and powerful.

Once a siftr is there, and depending on the permissions the author chooses, anyone can both view it in a browser and add data to the siftr.

Settings screen for my siftr, Explore UNM. Not a lot to it. Easy for novices

As I write this, more development of this tool is planned, so please do keep abreast of changes. That said, in its first incarnation, siftr is a responsive web app. It is designed to be used on smartphones (both Android and iOS) as well as on desktop web browsers. There are no apps to download, just the custom URL to type in. Though it is possible to use any of the features from any browser, the basic idea is that users on phones, out in the world, will collect data. Then the group or individuals who care about the map as a whole can get together and bring up the site on a larger screen, maybe projected, to discuss or make other assessments from.

Mapping

Right now, the data types in siftr are rather limited. Basically each waypoint created by mappers can have

• An image
• A title
• Descriptive text
• (Exactly one) category (either from the list provided by the author or created by the mapper)

A siftr on a smartphone. Note the big plus for adding a new waypoint

As mentioned above, each waypoint also is the root for a comment thread. Mappers and viewers can ask each other questions right there in siftr, add additional information, and even share the post (it is after all just a special webpage) via social and anti-social media. Do note though that audio and video are not currently supported (they are in the ARIS Notebook if you need them) and that each waypoint can only have one image.

When a mapper creates a waypoint, the default assumption is to use the mapper’s current GPS coordinates, but the point can be manually moved as well.

Viewing

As a responsive web app, siftr looks a bit different on a bigger screen, giving viewers more ways to look at what has been collected. There are map-based views, image-based views, and ways to filter by category or search for users and text. This is a nice way to review a collection with a group. Note too, that you can also add points right here. You don’t need to be on the phone.

The siftr, UNM in Spring, as seen on a desktop browser. Easy to see and filter your collections, but you can still map if you want too

Using siftr

If you know me then you know that I tend to use these tools in a couple different ways: for my own projects and classes, and with other individuals and groups. In both of these circumstances, but especially important for the latter, siftr takes way less lead time to use than something like ARIS. A teacher does not need the time and resources required to develop interactive media. All they need is an idea for something to go out into the world and look for. Teachers who may be interested in games and mobile media but who do not have the time, expertise, or wherewithal to get involved in a deep way have no problem using siftr with almost no training. For a siftr exercise to be valuable depends only on what becomes of the group, the map, and their conversations. The tool is less heavy, less a focus. It can get out of the way and simply enable.

I have used siftr in two ways.

1. I asked my Local Games in ABQ students to explore our campus as a preface to exploring the city. To get used to trying a place on with new eyes and discussing their observations with each other.
2. I created a UNM in Spring siftr to document the flowers and other growth on campus. This is mostly a personal, meditative activity for me these last two years. I try to remember to go out, wander around, and look for signs of life at a particular time of year. I share it with others, but there is no structured participation.

And I have worked with a colleague here at UNM who used siftr as part of his collaborative research. He is a historian, mostly of sports, and was taking a few undergraduates to St. Louis in the summer of 2015 to study the 1906 Olympics that were held there. They did research in archives but also in the city. Siftr was a lightweight way, easy to use far from my help, for them to explore and take notes of this place in a way that facilitated future discussion.

The future for siftrs

I think there is a lot of promise for siftr. Because it is so lightweight and yet customizable, because it is so clearly about leveraging our understanding of place, I think there are a lot of causes out there that it can support. Like the early days of ARIS (and maybe these are still the early days) right now we need a bunch of people to get their hands dirty imagining what they can do with this new tool. I hope to hear from many of you in the near future, and help to amplify your voices, when you find a way to do something interesting with it.

Siftr is in very active development as part of a projects/grants at Field Day. They are hoping to make it a more comprehensive data collection tool and integrate it with the functionality of both ARIS (for those who want to build something more involved than a map) and Nomen (the field ID tool—also by Field Day—you may remember from the Seesign project I wrote about last year). Field Day would end up making a better app if they could see what people want to really use it for.

Right now, there’s not a lot out there about siftr, though many projects have been set up in Madison in partnership with Field Day Lab. They wrote a brief post about a photography class that used siftr, and how it brought the class together. Check it out! Another thing to do is to look at work that has been done with the ARIS Notebook, and imagine it being much easier to pull off now that siftr is here. A couple places off the top of my head:

Tim Frandy’s Folklore 101 course at UW-Madison. In his chapter from our book, Mobile Media Learning: Innovation and Inspiration (free PDF from ETC Press and very cheap paper copies!), he describes how he organized his large college class to teach people how to collect and how to think about folklore. He used the ARIS Notebook (and you could use siftr unless you need audio) to seed the world with sample interviews, collect students’ interviews, and review them together to try and understand what good ethnography might be. It is very practical and inspiring work.

Digital Graffiti Gallery. Way back in 2010, before the Notebook, a Local Games in ABQ student’s (Ivan Kenarov) class project here in ABQ that has spun a lot of wheels in a lot of heads since. The idea is simple: capture graffiti, and keep it on a map long after it’s gone. AFAIK, no one has really run with this idea yet, or used the collection to have substantial conversations about the various issues that intersect with graffiti, but this idea is just waiting for the right person to make something great out of it. For me, the student’s game clued me into to how mobile data collection can be a strong motivator for community exploration. To this day, I leave my basement office with the hope of adding to our evolving map (even though intervening software changes mean that map has been reset several times).

One word of caution I will put out there for you though because I have already seen so much of it in my early consulting around the possible uses of this tool: think carefully about the experiences you hope to create, your audiences, and how their use of this tool will do something for them or an idea. A well-meaning but dead-end impulse is to simply create a map in the hopes that it will be of use. For instance mapping campus resources for a specific population. First, this might be a case where the use better suits Google Maps. Second, we too often confuse supplying information with teaching. How would this map be used by your audience, why would they use this instead of the all-encompassing and automatically updating maps on their smartphones already, and how can you see them not just as users, but contributors to knowledge about place.

On a related note, for citizen science projects, we focus too often on the optimistic idea of having an army of volunteers collecting data for the scientists. These are fine, useful and interesting projects, but they don’t always deserve the term citizen science. First, not every bit of data that might be collected will actually help figure anything out. But we have an extreme optimism regarding what we think data can do. Think carefully about what you’re going to do with the data after you get it. Also, citizen science really starts to happen when nonprofessionals can play deep meaningful roles in the work, not just  doing grunt work for those in charge. How can we see a mapping activity as participating in a larger discussion or as a hands-on way to begin learning about something? Coming into a new world and new personality?

Anyway, I hope you like this intro to another wonderful tool developed by Field Day Lab. Once you get started with siftr, head over to our forums and join the conversation.