Reflections on Agile Development


I learned over the summer that software development teams often use a process called Agile. The short version is that teams work in short cycles to create functional (but not polished) versions of a piece of software. Then they show it off to everyone and discuss bugs and other issues as well as next steps. This way, there is a constant feedback loop and no one team is given too much time to head down the wrong path.

So I decided to run our first project in AP Computer Science Principles using Agile Development. But I figured it was a bad idea to try to learn Agile and also how to write code at the same time, so we used this activity and made a Lego city using Agile development. Here’s our final product:


Nice, right?!

Here’s the rundown:

  • There are three roles in the process – Product Owner, Scrum Master, and Developers. For this activity, I played the roles of  Product Owner and Scrum Master, and the students were split up into 3-4 teams of Developers.
  • First, we needed what’s called the Product Backlog. This is just a list of the features of the city. For this activity, I had the list already done, so we just put each thing on a post it note and put it up on the board.
  • Next, we “estimated” each of the backlog items. This is where we decided about how hard each item would be to complete. We set up “lanes” on the board and took turns placing the post its into the lanes 1, 2, 3, 5, and 8.


The numbers 1, 2, 3, 5, and 8 are arbitrary. As in, they aren’t minutes or days or anything. It just means that if they decided a 1 story Lego building was a 1, then a 2 story building should be a 2 or 3 because it takes at least twice the work to get done.

  • After we had estimated the difficulty of each of the backlog items, we set about splitting up the work among the teams. This is called sprint planning. A sprint is just a short development cycle (we used 7 minutes for one sprint). Here’s our sprint planning board:


A few things to notice. Not all the backlog items were included in the first sprint. We only had 7 minutes, so we chose what we considered to be the most essential things for our Lego city and focused on those first. Also, one student did a little quick calculating there to estimate how many “points” each team would have to complete per sprint in order to finish the project in three sprints.

  • Then we set a timer and they went to work for 7 minutes. It was pretty hectic! After the time was up, a few of them placed the buildings onto the city board so we could review.


  • This is where I needed to let everyone know when I was playing which role. First, as product owner I basically tore apart their city. “The one story buildings should all be the same color!” “The hospital is too small!” “Where’s my bridge?”


Then I switched roles and had a “Scrum” meeting with each team. The purpose of the scrum meeting is for them to let me (the Scrum master) know what their team needed to be successful.


There was a lot of “we need more white pieces if we’re going to make this 1 story all one color” and “can we get a door with hinges?”. It was nice that I had withheld the fact that I had some special Lego pieces set aside (such as doors and windows) until they asked for them. This made the role of scrum master clear to them. I wasn’t answering questions like “how do I build this?”, just removing obstacles in their way of building. Later on, it should be clear that the scrum master isn’t there to help them learn to code (that’s the developer’s job!) but rather to remove obstacles in their way.

  • After the scrum meetings, we went back to the Backlog and planned sprint 2.

sprint 2.jpg

  • Then we repeated the entire process one more time, moving backlog items into the “finished” area as we went.

sprint 3.jpg

After sprint 3, the Product Owner was very satisfied with his city. We reflected on the fact that this happened through a process of quick bursts of working and then feedback. YES!! The point emerges!!

We used this same process for our first project in which they made a game in Scratch. I have a lot to say about that too, but I’ll save that for another post. I also wanna write more about my assessment plan this year because it is awesome! I haven’t “graded” a single thing this year.  I’ve given tons of feedback (formal and informal), and kids are giving feedback to each other. Beautiful! But that also deserves its own post, so I’ll hold off for now. I’ll write it up soon though, promise!🙂

Learning Process: Art, Yoga, and Programming


It’s been a while since I’ve blogged and a lot has happened.  I plan to write a few posts about my plans for next year, especially regarding assessment and PBL. But first I have to write about a project I’ve been working on and some connections I’ve made.


DAY 54

So I’ve been making sketches like the one above using the programming language Processing every day for the last 54 days. I started it as part of #the100dayproject (follow that tag on Instagram, Facebook, or Twitter and you can check out lots of people’s awesome art projects) and I tag everything #owenprocess if you wanna check mine out.

Many connections between art and programming have become super obvious to me.  I think a similar project will be a great way to lead students to make these same connections for themselves. The simplest example I guess is the connection between the tools you are best at using (or like the most) and the art that you produce. As I learn more about what is possible in the Processing language, I make more interesting sketches (to me anyway🙂 ).


DAY 26

But the connections to my yoga practice have been way more striking and came as a complete surprise. Ok so, a few months before I started this art/programming project, I started learning the Ashtanga primary series. I started going to classes with two amazing teachers who showed me the first part of the series and I also started practicing on my own outside of class more often. Doing the same series of poses each time I practice feels right to me. It’s like I am starting from scratch and building something. This is the first connection. Every day at some point, I go to my computer and look at this:


Then I start by writing some of the code I’ll need to draw something on the canvas:


This also feels right to me. It feels like I am starting from scratch and building something one piece at a time. And it is exactly the same comfortable feeling I get when I come to Samasthiti (sama = same, equal / sthiti = to establish, to stand).


Pictured: not me

In both of these states, I usually have an idea of where I am going next. In Processing, I’ll create the canvas and then start choosing and adding colors to it. In yoga, I’ll complete 3 cycles each of sun salutations A and B. In both cases it feels like I am building a base, a foundation on which to build more. And so I do. I learn more functions and methods in Processing and then I can make new types of sketches and interactions. But I try not to implement anything if I don’t understand what it is doing. The same goes for my progress in yoga. With the help of my teachers, I add new asanas only when I am ready for them, when I feel like I understand the purpose of the posture (what it is intended to do).

Marichyu0101sana D_0

Image credit:

Another connection I see is abstraction. My computer performs billions of operations per second, so there is no way I’d be able to keep up if I had to pay attention to all of them. So in my interactions with the machine I need to be at a higher level of abstraction to get anything done. For example, I can type in size(600, 600); and the machine knows I want a canvas that is 600 x 600 pixels. I don’t have to tell it every step involved in how to do this (there are a lot); it already knows and it just does it.  In yoga, I am also moving to higher levels of abstraction by memorizing the postures and the sequence. For example, say I want to move into upward facing dog. There are lots of things my brain needs to tell my body to do. Feet pressed into the mat, legs up, shoulders back, chest forward, … But now that I’ve been doing it for a while I just do it. It’s like I’m calling the “upward facing dog” function and it just runs.

There is definitely more I need to write about, but I am at a training for the computer science course I’ll be teaching next year so I have homework to do! Here’s one more before I go, and don’t forget to check out #the100dayproject to see tons of really great stuff being created every day!


DAY 20 #owenprocess


Another Rubric Update


I think I may be on to something with this newest rubric. What it says:

What I did with it (in pictures):

This slideshow requires JavaScript.

How it goes in my gradebook (SocraticBrain):

This slideshow requires JavaScript.

#phonespockets First attempt: quick reflections



So I am jumping on the #phonespockets bandwagon. I just hit record on the voice recording app on my phone, stuck it in my pocket, and then went about teaching. That was yesterday. Today, I listened to the recording (~10 minutes). Here are a few quick noticings:

1.) I was a little self conscious at first, but after a few minutes I forgot I was recording. I set a timer to remind me when 10 minutes was up so I could stop recording. It didn’t interfere with the class in the way that video recording sometimes does.

2.)  Background noise was not as bad as I thought it would be. I constantly worry that the room is too loud, making it hard to concentrate. But I want kids talking to each other (learning is social and all) so I have to resist the urge to quiet everyone down. Listening to the recording, the room doesn’t seem too loud really. I can clearly hear not only my own voice but also the voice of the student I’m talking to at the time. It’ll be interesting to see if this holds for other classes as well, in particular my big class later today.

3.) I only talked to 4  kids the whole 10 minutes. This is something I’ve also been thinking a lot about; how do I make sure to have authentic, useful conversations with students but also be fair about getting to everyone? Talking to 4 kids in 10 minutes means I averaged 2.5 min/kid. So in a class of 27, I would need 68 minutes to get to everyone. Some days that’s doable (I’ve got 90 minute blocks), but I suppose I don’t need to reach every kid every day. But in that case, I need to think more about tracking these convos so that I don’t always end up helping the same students and ignoring others.

4.) I am pretty good about using wait time after asking a question, but I could stand to add a few seconds to my usual wait time. I also need to utilize follow up questions better. For situations like this:

Me: “What are you going to do next?”

S: “I don’t know”

Me: <makes a suggestion>



Updated SBG Rubric – Physics


I am writing this post mostly as an update to this one now that I have been playing with my SBG rubric for a while. To start with, check this out:


This is one of my “Challenge Problems”. It is a chance for students to show me evidence of their mastery of any or all of the learning targets for a unit. Notice the rubric in there? I put a check next to a target if I see solid evidence of that skill (very minor mistakes only). Then I give feedback in the blanks. So far I like the small blanks, they encourage me to keep it short and sweet. If I need to say more to a student than that, I probably need to talk to them in person anyway.

This is the final part of the rubric for the unit. To earn the previous two levels for each learning target:

Level 1: Complete some class activities. Show them to the teacher for feedback (or to level up and move on). When possible, Level 1 can also be earned through a multiple choice scenario. In either case, Level 1 is about showing some basic knowledge of the topic or skill.

Level 2: Ace a multiple choice quiz. These are short and only one topic at a time, although a student could take as many different quizzes as they want in a single sitting. There is a 12 hr wait time for re-takes. All questions are randomly generated, either from a bank of possible questions or the same questions with different numbers. Level 2 is all about applying the skill in a problem solving situation (but still in isolation).

Level 3: Demonstrate evidence of the skill in coordination with the other skills in the unit. Students know ahead of time what the Challenge Problem will look like because Level 3 is about knowing what good evidence looks like. This is why my rubric in the example above doesn’t have info for the students. Part of this task is figuring out what “Level 3” looks like for each skill. It’s not totally up to them to figure this out though. We have activities built into the unit where we explicitly talk about what Level 3 looks like for each one. Plus they can attempt the challenge problem multiple times, getting feedback each time.

Level 4: Ah, level 4, you continue to prove elusive. I’ve got a place for this level in my gradebook, but so far have not really used it. I’ve talked about using some sort of super-ultra-mega-challenge problems, or a short student directed project (my fav idea) but so far haven’t followed through😦.

Here’s an example of how my gradebook looks now:

Project Rubrics Part 1


I want my engineering class to be project based. So I’m currently reading Setting the Standard for PBL. 


One of the first challenges for me is to make sure that students are learning the stuff I want them to learn. It seems to me that it’s all about documentation. I want them to be able to show me what they’ve learned at different points in the project. In Setting the Standard, the authors describe “checkpoints” in projects. At each of these checkpoints, student work can be critiqued for quality and then revised. The critiques should come from other students, teachers, and (when possible) experts in the field. I think if I design a good rubric, it could be used for these critiques, but also to document key knowledge and understandings based on standards. Here’s the plan:

First, a rubric that includes knowledge/skills/understandings. That way the person filling it out knows what to look for. But I am thinking it should be an all or nothing score (no numbers or other indicators). Either the artifact shows evidence of the skill or it doesn’t. If you’re not sure, then it doesn’t. All sides know ahead of time that this won’t affect the group’s grade on the project, it’s just for documentation.

Next, a feedback section on the rubric. This includes what products should have been created (different for each group, so they probly have to create this part). Here, there is only a description of the perceived quality of the product and some suggestions for improvements. This part is all about getting the group to focus on craftsmanship and quality. Again, both sides know this doesn’t affect the grade.

With these two parts of the rubric completed, each student will have a clear picture of what they’ve learned as well as what revisions to make to their product to improve quality.

So what do they turn in?

  • Each student turns in a project checkpoint. This includes all products their group has created for the project as well as several two part rubrics. The method of submitting will be determined by the nature of the project.
  • Each student turns in the rubrics that they have completed for another group.

So what goes in the gradebook?

Right now I am thinking that none of the info from the rubric needs to go in the gradebook. The students will get credit for two things:

  1. Turning in a project checkpoint. The grade is not based on the quality or quantity of products included, just that it’s turned in. The rubrics need to be included, but the score is not based on what the rubrics say (just that they are included). Documentation is the key!
  2. Turning in the rubric and critique they did of another group’s checkpoint entry.

Because students tend to value what they get grades for, this signals to them that what I value is documentation of the process and critique and revision, not initial quality of their products.

What I still need to think through:

What do I do when a student turns in a project checkpoint that does not show evidence of revision based on previous critiques? Do they lose points or is it enough to have turned something in? Will I even need to worry about this once the right culture is established?

What does the feedback section of the rubric look like? Does it need to be different for each project or group? Or can it be standardized?

Learning by doing, and other things that happened today


So I was really pumped up today in our weekly PD time for two reasons:

  1. We had choices about what to work on and whom to work with.
  2. The task was open ended. No product was required and no methods were prescribed. 

The choices we had were all dispositions that we are looking for in students. The one I chose to work on was “inventive”. As in we want our students to be inventive, how do we make it happen? 

There were three of us that chose this topic and we all seemed to be on the same page about a few things:

  • The problems have to be real. One teacher said “we don’t want their work to go up on the fridge, we want them to make the fridge”.
  • They need an authentic audience. As in not the teacher and rest of the class.
  • The problems have to be important to the individual student. 
  • Grades get in the way. We may not be able to get rid of them, but if we want kids taking risks, we have to get the grades out of the way.
  • Project based learning is the way to get kids inventing. And the less we prescribe, the better. Open ended projects will get us the most creative results. 

What struck me as I started to think about this short list is that these are the exact same reasons why I was pumped up about the work we did today. Chosen by us, no prescribed product, authentic audience, real (and important) problems. 

Learning by doing, I see what you did there. 

Also, this happened today: 



A kid made these after I told the class we needed something better than tape to hold these doors in the unlocked position. A few notes:

  • Not a graded assignment 
  • Kid chose to work on it
  • Audience = the facilities manager and maintenance workers

Our facilities manager liked them a lot. And he brought us another project: 

Apparently these are absurdly expensive if you only need one or two. So we’re gonna make one. Getting the font right and the raised Braille lettering correct will be the hard parts here I think. 

This reminds me, I’ve been meaning to ask, do you have something that’s broken and could maybe be fixed with some inventiveness and 3D printing skills? Bring it on, Lion Design Team can handle it. Will you be our authentic audience? We’ll do a Google Hangout and you can show us if it worked!