Anyone in possession of a smartphone knows that there’s a ton of applications promising to deliver solutions to any number of problems we encounter on a day-to-day basis. Uber delivers people to destinations without the trouble of driving themselves. DoorDash delivers meals with the press of a few buttons without the time and effort of meal prep problems. Zoom delivers the convenience of meetings with colleagues in the comfort of our homes alleviating the stressors and limitations of living in unnecessarily close proximity to the office. And a plethora of apps have even delivered the possibilities of finding romantic partners without having to mingle uncomfortably in crowded bars and stuffy social events.
These apps arise in response to daily situations that require solutions, and many propose to disrupt some prevailing status quo. So where is the app that will deliver rigorous and engaging lessons that solve the many problems of practice that may surface over the course of any ordinary day in teaching? Though many apps routinely disrupt our students’ attention to learning, no app has “disrupted” our profession at the scale other industries have experienced. For us teachers, it seemed like only a matter of time before an app would do the same in education. Sure, a few apps offer some usefulness, and AI is alternatively promising and terrifying—but as any teacher will tell you, no such app currently exists to solve all the many macro and micro scenarios that arise in the course of even just one lesson.
And yet, it seems so possible, the potential so obvious. If an app like Amazon can deliver 7.7 billion packages in a year, then surely we can figure out an app that can deliver learning to students… right? There is, however, a subtle but significant difference between the problems that other apps have sought to solve and the problems we encounter in teaching. This difference is illustrated particularly well within the framework of the last mile problem.
What is the Last Mile Problem?
The last mile problem is named after the last stage in a product’s delivery chain to its final location. While the name conveys only distance, the “last mile problem” itself is much more concerned with logistics.
Let’s consider the delivery of a tube of toothpaste to your place of residence. The toothpaste originates at a factory where it’s packaged and then delivered to a distribution center. After a distribution center receives your order, a package of toothpaste is loaded onto a delivery vehicle, which is driven by a courier who then navigates the “last mile” to your front door.
The initial stages of this process constitute a fairly large and homogenous distribution stream, similar to the trunk of a tree (the factory) with a few large branches (the distribution centers). Up to this point, the toothpaste company has only had to deal with relatively few variables in the delivery chain, all under control of the toothpaste supplier and the delivery service. Things become much more complicated, though, as we enter the “last mile,” moving to the smaller branches (residential streets) and the leaves (individual residences).
This “last mile” involves more than just the final distance (which may or may not be 5,280 feet). Not only are there exponentially more destinations to reach, each one is a little different, with all manner of unique potential problems to be navigated (closed roads, angry dogs, traffic, mailboxes on a strange part of the building, etc.) The last mile is a dynamic environment which requires a different type of problem solving than the beginning stages of the delivery stream—and this is true for more than just toothpaste delivery.
The last mile problem shows up in any system tasked with distributing something from a few central locations to many distinct final locations; and there is no shortage of entities actively pursuing solutions that they hope will be effective and lucrative. This is true for product delivery, content delivery, public transportation, medical care, and even education.
The Last Mile in Education
Using the subject of mathematics as an example, the current “delivery stream” of content knowledge and understandings in public education can be described as the following: curriculum companies (the factory) deliver their packages to state, regional, and local school districts (distribution centers), who then communicate to teachers (delivery drivers/couriers) what to teach their students (individual customers).
In the interest of fairness, it’s tempting to borrow the strategies from other last mile operations and use them as models for our “delivery system” in education. There is a big difference, however, that we must keep in mind when comparing other last mile problems with ours in teaching.
Most delivery companies are generally working with static objects and static destinations. The objects and destinations do not really change as a result of the delivery process. This is true even when the “objects” being delivered are people. Most public transportation services, for example, are not seeking to change you as a person over the course of your bus ride. The complexity of their last mile problem comes primarily from the dynamic nature of the physical “last mile” itself.
We teachers, on the other hand, are trying to deliver dynamic “objects” (understandings) to dynamic “destinations” (human minds) in addition to the always dynamic “last mile” (a daily lesson). Unlike toothpaste and a destination of a mail box, both the understanding and human mind will change over the course of a successful delivery process. In fact, the goal of responsive teaching is for the human mind to integrate the understanding such that the learning is personalized and the learner is forever affected.
Consequently, the dynamic nature of understandings in interaction with human minds compounds the complexity of the already dynamic “last mile” in education. The task becomes even more complex when we consider that there are generally 30-35 “last miles” happening simultaneously in a classroom. Therefore, as we look for strategies, we must keep in mind that the last mile in education is exponentially more complex than the last mile in other endeavors.
Last Mile Strategies
When we look at the last mile problems in other industries, we often see a two-pronged strategy: increase the organization’s capacity for dealing with uncertainty in dynamic environments while also looking for ways to reduce the amount of uncertainty within the delivery process.
An example of this is the long-promised drone delivery strategy. The delivery companies are attempting to increase their capacity for dealing with uncertain, dynamic environments by increasing the computing power of the autonomous drones. They are also predicting that the comparatively less-crowded sky will be a less dynamic, and therefore less uncertain, environment than residential or city streets. These sound like promising ideas, though we have yet to see them function in the real world.
We can pursue a similar two-pronged strategy within education, but there is a limit to how much uncertainty we can reduce before the responsiveness of the learning environment is also compromised. To attempt to sanitize the teaching environment for the sake of minimizing confounding variables is also an attempt to control for the human elements in learning. We, of course, do not want complete chaos, nor do we want instruction without curricular constraints. But if we try to make the understandings static, we end up producing scripted performances that connect with only a few students—thus undermining the dynamic interaction of understandings and human brains that we know to be essential to the learning process. If we try to conform our students’ minds to accept static content knowledge, we end up with either mindless rote learning, or open student rebellion, neither of which is the goal of a responsive learning environment. Rather, instead of seeking to eliminate uncertainty in the learning environment, we would be better off investing our time, energy, and resources in building our capacities for working within the uncertainty inherently characteristic of dynamic environments.
For many teachers, a big part of the draw to the profession was the chance to work in such a dynamic environment. Most teachers don’t enter the classroom looking forward to reading a script or following the pacing guide of a packaged curriculum. We became teachers with the hopes of building meaningful relationships and creating learning experiences that would bring students into meaningful understandings that change how they see both themselves and possibilities for their own future. One of the most common responses heard from teachers who have engaged in a CRE residency is some version of “This is why I got into teaching. This has renewed my joy and purpose in this job.” The CRE framework gives us the tools and mindset to engage productively and artfully within the dynamic environment of a classroom; and that is often an exhilarating experience that feeds the embers of our professional passions.
We’ve seen unprecedented levels of dynamic uncertainty in the past few years in education, and that trend looks to continue. The algorithmic applications of instructional models won’t suffice in the uniquely human delivery environment of a 21st century classroom. If we are to be fair and responsive, we are wise to resist false promises of reduced complexity and instead develop our human potential to work with the problems of practice in ever changing dynamic environments.