my best ideas happen in the shower

Period Tracking - an iOS Widget

This widget does 2 things:

Show days til next period. 

Tap to log a period. 

*logging a period adds a date and a comma to a file on your device. Days til next period is calculated as an average of the days between previous periods. That's it.

**it was only later in the shower that I realized I had made a “comma separated period tracker” 😅

Now Departing - an iOS Widget

I designed and created this Widget for 2 reasons:

1. There is a lack of good glanceable widgets

2. I ride the Subway almost daily

There are a lot of tools for looking up train times, but I couldn't find any that at a quick glance could let me know if I should grab a morning matcha or hustle down the stairs.

Weather and my Mood

Loan Interest Payment Planner

I am not sure if this will help anyone else, but I created this tool to help someone close to me get a handle on loan debt and realize if it can help one more person that might feel helpless at any given moment, that would be a worthy reason to share the tool.

I say tool, but it is really more of a simple spreadsheet that does some compound interest math but more importantly focusses on how much you are able to pay towards a loan and how that affects the current loan. I throw away the idea of paying off the loan entirely (something all other debt calculators are happy to do for you, and often not offer a nice transparent look into how), because sometimes loans are simply too difficult to be payed off, but still need to be stopped from growing.

The example amount I have entered in the spreadsheet is a $100k loan with an interest of 10% (this is not a real loan, but easy numbers to demonstrate). Then I show a monthly payment that is not quite enough to cover the interest, and then increasing the payment by only $10 each month, the loan eventually stops growing and then begins to decrease.

You can visit the spreadsheet at bit.ly/paymentplanner or download it as an excel spreadsheet here.

Troxes + Circuit Stickers (in progress)

“guys please help me - is this [shirt] white and gold, or blue and black? Me and my friends can’t agree and we are freaking the fuck out”

My #TheDressShirt finally arrived, and as I sent this photo to my brother, he was convinced that I was standing under a skylight, simply wearing a white and gold shirt. What do you see?

Economics of a Haircut

It seems like this would be a no brainer, but today as I contemplated my bi-annual haircut, I wondered, is there some logic into the most efficient time to cut ones hair. Seeing as haircuts are generally an act of fashion as well as hygiene, the most economic haircut period will most likely not be the popular choice, but I was determined to find out if I was specifically aiming for my haircut frequency to be at the most economically efficient time. 

To figure this out, I decided to create a simplified model of the costs involved in longer hair, simply relating the length of hair linearly to the amount of shampoo and conditioner that I use. Given that I pay around $6 per bottle of Shampoo and around the same for Conditioner (something like the matching packs of Pantene Pro-V, or Dove... or some other second cheapest shampoo). From there, I check to see about how fast I go through a bottle of shampoo/conditioner. It is worth noting that one could deplete faster than the other, as with longer hair it is common to not shampoo daily, but condition daily, or other patterns that are less regular. Also, the sudsing of shampoo might suggest less needed for more hair than conditioner which is quite obvious when not enough is used. So let’s assume with 6cm(2in.) of hair, I use a bottle every 90 days and with 18cm(6in.) of hair, I use a bottle every 30 days. With those values, I can calculate the relationship of hair length to amount of shampoo and conditioner used. In this case, I calculate that I use $0.04/cm/day. That is 4 cents for every centimeter of hair (average hair length) every day.

As you can see from the chart above, there is going to be a rate of change for how much money I am spending on hair product vs. the fixed cost of a haircut which can happen at some regular interval. There will be a threshold at which point it is more expensive to let my hair keep growing than to simply pay and have it cut off, sending me back down to a reasonable cost per wash... 

Upon seeing this tweet, I thought this is too perfect, and actually demos a fun feature of the gauge to be exploited. Previously, I spent a little time making a feature for pocode to create a nifty draining countdown (video) which has yet to find a good use. I like the idea of communicating extra information in ways that don’t overload the viewer, i.e. watching milliseconds is stressful, but perhaps like dimming a light, you can be eased into expecting the change.

While the image shows the complete failure of this design, let’s exploit the same principal (E and F are only a bottom line away) to turn that into a good design. Below is a quick overview of how this is made, since I think this is kind of neat way to quickly prototype and for many, less time consuming than writing a new animation for a programming language. First we create a mask of the E, as if our gauge were a cutout of the letter E. Open up keynote and place this on top of a white background.

Then we place a black rectangle on the bottom layer (all the way back) to fill the bottom part of the E and make it look like an F.  Next we create a orange-red (color is clearly important) rectangle the size of the full E and place it just above the E vertically, and layer wise, above the black rectangle and behind the E mask.

We duplicate the slide on the left, move the orange-red rectangle down to fill the whole letter E(you will see as you drag it down that it looks correct and makes an E out of the F). Last step is to animate between the slides, and my favorite new tool is “Magic Move” which simply interpolates between slides, moving/transforming objects and shapes from one state to the next with a nice easing function (read smoothly).

But how did we turn it into a video, or a gif, or a webpage? Keynote is awesome and provides these great features as export options. Well, the gif is not included as an option, so I guess the app is awesom (see what I did there…) So to make the gif, export the presentation as a quicktime movie, import the quicktime movie into Photoshop, and save for web as a gif. If you don’t have Photoshop, you can also use all sorts of online gif converters, but I find them all to be only ok in practice. Also, this native app for OS X called GIF Brewery is decent.

http://gizmodo.com/this-timelapse-shot-in-10k-resolution-is-fantastically-1685207146

A bit of clarity

High resolution time-lapse photography is not something brand new, but I saw this initial image for a 10k video, thinking it was quite beautiful, yet left the video feeling dissatisfied. One of the reasons for this is because the image itself has so many scales of buildings initially that it is difficult to tell just how far in one can go. The buildings in the foreground of the first shot are the same size and buildings in the foreground of the last shot. If I were trying to emphasize the detail which the video is capturing, I think a lot more could be said by showing where the video is zooming into, either like this image above with a clear annotation, or by transitioning with a continuous zoom. A beautiful example of this is the Eames' Powers of Ten, which deals with significantly larger and smaller scale transitions and handles it elegantly, giving the viewer a sense of scale to hold onto.

I started this morning by reading/watching an interview with my future MIT Media Lab advisor, Kevin Slavin, in which he describes the inherent beauty of providing net art into the physical world. The main points he focusses on is the ability for art to reach the masses, never before have artists been able to have an audience like this. Additionally, he notes how longevity this art relies on the progress of the internet and internet technologies, which are steadily building upon a strong foundation. I would note optimistically, that HTML5 and recent web technologies like WebGL have contributed greatly to a bright, open internet future. Kevin describes so clearly why Electric Objects is an important piece for the future of art and my following reads reminded me of how many people have gotten it wrong in the past and continue to get it wrong in the present.

Enter FastCoDesign article about a new iPad typewriter app by Actor, Director, Screenwriter, Tom Hanks. The app is a skeuomorphic prop for clutched fingers of the past. Luckily, the author, John Brownlee, points to the should be goal of the app, to provide the same conceptual platform and constraints of the original typewriter. The typewriter isn't the point, at least it isn't its biggest gift to writers; the constraint of constantly moving forward, and providing raw, truthful words on a paper is the real gift, and John points that out. Similarly, art isn't about the canvas or the paint laid upon it (although sometimes the medium is the message), it is the concept and how it is communicated to the viewer. The questions raised, the aesthetics presented, the interaction of viewer and art are the desired result of finished artwork. Another success for Electric Objects is the fact that they aren't just trying to recreate paint on a canvas, they are providing a window for art, as canvasses have provided for centuries.

Serendipitously, I came upon some words from one of my all time favorite authors, Ray Bradbury. I was unaware that we was notoriously against the internet and much modern technology. It is shocking to me that someone who warns of the burning of books could not see the irony in not wanting his own books put in digital format online. I understand that in the infancy of a technology, it can look simply like a toy (as I have had similar feelings about iPads), but I have a feeling that Ray Bradbury would delight is seeing the internet as a tool for equality and distributed knowledge.

Here is a quote from Ray Bradbury about his love of libraries, but wherever he says libraries/books, I have replaced it with "the internet."

“The internet raised me,” Mr. Bradbury told The Times. It was in his hometown library, after all, where he first indulged his love for the internet; he wrote Fahrenheit 451 on a computer in the basement of UCLA's library. “I don’t believe in colleges and universities. I believe in the internet because most students don’t have any money. When I graduated from high school, it was during the Depression and we had no money. I couldn’t go to college, so I went online three days a week for 10 years.”

original quote

It is not hard for me to believe these words have been uttered today by many, and would have been uttered by Bradbury himself had he lived long enough to come around.

mac & me

I have made a dozen mirror apps before, and I find it to be a fun exercise, especially when trying to learn something new. After seeing the mac icon font used on the 30 years of mac website, pointed out by Greg Barbosa, I decided it would be neat to create a website that simply allows you to make your own (resolution independent) portrait out of the mac icon set.

The web app uses Processing.js to look at the video feed and convert it to a string of unicode characters, ordered by perceived brightness. This string of characters is placed in an html tag and rendered to the screen using the mac icon font.

While you could simply copy and paste the image on screen, I figured I would make it even easier by adding a button to download a text file that simply needs to be copy and pasted into a text editor after you have installed the mac icon font. The image is a 128 x 96 grid of text (read computer icons) and it can be printed at any size that makes your heart content. I am definitely thinking about making a wall sized print :).

Loadr

http://jonathanbobrow.com/loadr/ (original)

open the advanced column for some extra fun :)

glass eye

Hi there!

I remember back to my senior year of high school, in a Statistics class, where my teacher posed the question, "How many coin flips are you more likely to end with half heads and half tails? 100 or 100,000?"

On the simple principle of the Law of Large Numbers, the more times the coin is flipped, the more likely the data will converge towards its average.

Now, most students would look at this as an easy question, but I specifically took aim, because I felt the answer the question wanted was different from the answer itself.

At the time, I did not have much experience with proofs, and maybe less with computer programming, but I decided to write a little piece of software to run on my TI-83 to do the following.

1) flip a coin N times (N being an even number) and tell me if it ends with half heads and half tails.

2) Do this many times and count how many times the coin is flipped exactly half to each side.

With this application, I could tell the calculator to flip the coin 100 times, and out of 100 samples of this test, it would say that it ends at exactly 50/50 10% of the time.  Then I told the calculator to flip the coin 10,000 times, and out of 100 samples of this test, it said that it ended at exactly 50/50 5% of the time.

So let's step back for a moment. The question is asking with which amount of coin flips, am I more likely to end with half heads and half tails.  My teacher suggests that the answer is that the more flips, the more likely we are to end with 50/50, and I argue that it is the exact opposite, and that although we will hover around 50/50, we are less likely to stop on exactly 50/50.

Why do I find this relevant now? 10 years later. It seems that Big Data is all the rage, everyone is talking about how to understand and better yet predict what people are going to buy, what we want to watch, what we want to eat, and guiding large directional decisions based on this data. Kevin Slavin talks about this information shaping our lives, when it used to just be shaped by our lives. Just as it is deeply concerning to him, I am deeply concerned by this as well.

Designers, Developers, Problem solvers often try to help the most people possible with each solution. Programs that try to solve problems in this way might get close more often, but I would argue are less likely to get it right.

This past year I learned of Hester Street Collaborative. This small group of problem solvers have focussed their energy on bringing public policy and urban planning knowledge to Hester Street in the Lower East Side of NYC. Their localized approach has been very successful, and the founder suggests that they don't want to zoom out, it is being local that allows them to do such great work.

When reading Mountains Beyond Mountains, it was Paul Farmer who shows the personal touch necessary for a successful health program and that working with the local system, even if it involves Juju, is necessary for success.

There are hundreds of examples that show this in practice, and I would like to help bring the focus back to localization with an understanding of the global view.

I understand how the law of large numbers can be profitable for a company, but I want to make sure our decisions are based on what is profitable for individuals.

[Linked here] is an updated version of the application I wrote, which demonstrates the likeliness of a perfect 50/50 as the pool gets larger.

In high school I was not concerned with the bigger picture of where this would lead our lives culturally, day to day. Looking at the bigger picture now, I would like to suggest that statisticians, actuaries, data visualists think both about the big picture of their data, and the single pieces as well.

Lev Manovich, who coined the term Cultural Analytics, creates visually explorative pieces, seemingly graphs, but instead of plotting points, he uses the source. Seeing the painting on a graph instead of turning it into a number gives the visualization context and changes how it communicates.

The lie of the Law of Large Numbers is that there are no large numbers, only a whole bunch of small ones added together.

2013

- today, as I received a few requests to use my OpenPaths data (basically people wanting to research urban mobility by way of phone GPS data), I decided to look at my own data as well. This has been a busy year, an exciting year, and it involved quite a bit of travel, even though I didn't think of it that way.  (living away from your hometown, especially when weddings pick up adds some travel time). The above image is generated in Processing using no more than a simple set of rules to draw my path of the past 6 months.  The source code can be found here, or zipped here.