<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Paula Tulis on Medium]]></title>
        <description><![CDATA[Stories by Paula Tulis on Medium]]></description>
        <link>https://medium.com/@tulis.paula?source=rss-9414f9d99a04------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/2*yCb2ezlcmQH8NGqGp1Uoiw.jpeg</url>
            <title>Stories by Paula Tulis on Medium</title>
            <link>https://medium.com/@tulis.paula?source=rss-9414f9d99a04------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 05:56:39 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@tulis.paula/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Indirect and convoluted and wonderful]]></title>
            <link>https://medium.com/@tulis.paula/indirect-and-convoluted-and-wonderful-50bcab94a893?source=rss-9414f9d99a04------2</link>
            <guid isPermaLink="false">https://medium.com/p/50bcab94a893</guid>
            <category><![CDATA[code]]></category>
            <category><![CDATA[code-newbie]]></category>
            <dc:creator><![CDATA[Paula Tulis]]></dc:creator>
            <pubDate>Tue, 18 Jun 2019 13:49:27 GMT</pubDate>
            <atom:updated>2019-06-18T13:49:27.596Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/428/1*1ZETpznmHKI2gUru95cBNg.gif" /></figure><p>One of my favorite things to do the morning after a long day of coding is to re-watch the lecture from the previous long day of coding. I even like to poke around for additional lectures from different teachers to see if the ways in which they explain the concepts make more or less sense to my brain. The other day, an instructor made a connection between DOM content loading and the starting point of a Rube Goldberg machine.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FJMnnhJ9roRk%3Ffeature%3Doembed&amp;url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DJMnnhJ9roRk&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FJMnnhJ9roRk%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/028a65f15a1a5daedd47fbeafb948540/href">https://medium.com/media/028a65f15a1a5daedd47fbeafb948540/href</a></iframe><p>First, I became nostalgic for my high school physics project when I built a ridiculous homage to Mr. Goldberg. The second thing I thought about was how the process of coding in general can be likened to the quirky contraption.</p><ol><li>The main goal is to create something that works — yes, typically at the end of the day, we want the code to be as easy to read and succinct as possible. That’s the opposite of what you’ll find in a Rube Goldberg machine. However, as a beginner, getting the code to simply accomplish what I want it to accomplish takes the majority of my energy. It’s largely trial and error and seldom linear.</li><li>It should be transparent — When you look at a Rube Goldberg machine in action, the fun part is being able to see every single step in the process. One marble clicks onto a magnet. This action ever so slightly bumps another marble. This newly mobile marble rolls down a slope, falls through a hole, and the chain of events continues. When the final action takes place, whether it’s something exciting like a slice of bread transforming into toast, or something simple, like a final domino falling over, you know exactly what happened to get to that end point. All of the steps are out in the open. You can see the way in which this machine was constructed. Maybe you even get a small peek into the mind of the person who built it. There’s no black box, no magic. But it’s still pretty cool.</li><li>There are a million different ways to do everything — by running a quick search on YouTube for ‘Rube Goldberg machine’, one can conclude there are myriad methods to skin a cat. Spending a few hours on a lab and then comparing work among peers is another way to come to the same conclusion about code. Of course, the rules of physics still apply to everyone hoping to skin their cat with a Rube Goldberg machine just as the constraints of any programming language apply to everyone hoping to skin their cat via variables, methods, and functions.</li><li>Everything should pass tests — If you think about it, every step in a Rube Goldberg machine executes a function and passes a test at the same time. Hooray for built-in unit tests! If a marble doesn’t click onto a magnet with enough force, the next action will not take place, the flow will stop, and the entire machine fails. In programming, it’s perhaps a bit more nuanced. Certain functions or methods may work in one context, but they may not necessarily play nicely given different circumstances. And sometimes these caveats are difficult to notice without testing at every single step. This is why test driven development is stressed over and over again. My takeaway is: build and test your code as if you were building and testing a Rube Goldberg machine. Don’t leave a stone unturned, and you’ll thank yourself when it’s live.</li><li>The frustration and time spent on getting the thing to work is almost not worth it until you experience the satisfaction you get from seeing everything finally come together — when I was a junior in high school, building a working Rube Goldberg machine was a task that haunted my dreams for two weeks. I had a lot of competing demands from other classes, and I wanted to give up at multiple points in the process. Thankfully, my type A personality didn’t allow me to throw in the towel, and when the stupid thing finally went off without a hitch, I was overjoyed at the accomplishment. I didn’t know it at the time, but I was iterating, problem solving, and growing as a programmer.</li></ol><p>On one hand, I’m grateful that I don’t need to ever build an actual Rube Goldberg machine again. On the other hand, I build Rube Goldberg machines nonstop these days, and they are some of the most rewarding I’ve had so far.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=50bcab94a893" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[And then what?]]></title>
            <link>https://medium.com/@tulis.paula/and-then-what-a4af1ef4af7a?source=rss-9414f9d99a04------2</link>
            <guid isPermaLink="false">https://medium.com/p/a4af1ef4af7a</guid>
            <category><![CDATA[markov-chains]]></category>
            <category><![CDATA[bootcamp]]></category>
            <category><![CDATA[coding]]></category>
            <dc:creator><![CDATA[Paula Tulis]]></dc:creator>
            <pubDate>Thu, 23 May 2019 12:26:01 GMT</pubDate>
            <atom:updated>2019-05-23T12:26:01.984Z</atom:updated>
            <content:encoded><![CDATA[<h3>Scratching the surface of Markov Chains</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/220/1*9zohI0eezdUMSSKvQb6BBg.jpeg" /></figure><p>Recently, I was skimming an article that described how a prankster hacked a dating app. The article itself was just okay, but the concept that really piqued my interest was the use of Markov chains to generate content.</p><p>I did some research on this model, named after mathematician Andrey Markov, to see how it works and why these chains are important in programming.</p><p>If I were going to describe Markov chains to someone who has scant statistical or mathematical acumen (me), I would say this: imagine you have a very persnickety friend with a very particular sweet tooth. This friend will choose either a cupcake, donut, or cookie for their dessert, and this friend eats dessert every single day (but only once per day, and after dinner, of course!).</p><p>Here are her quirks:</p><ul><li>If your friend eats a cookie for dessert today, she will choose among cookies, cupcakes, and donuts tomorrow with equal probability.</li><li>If your friend eats a donut today, tomorrow there is a 10% chance she will eat another donut, a 40% chance she will choose a cookie, and a 50% chance she will go with a cupcake.</li><li>If your friend eats a cupcake today, tomorrow there’s a 40% chance she will indulge in a donut, and a 60% chance she will go with a cookie. She will absolutely not choose a cupcake again tomorrow.</li></ul><p>Your friend’s dessert idiosyncrasies illustrate an example of a Markov chain because her choice in dessert tomorrow only has to do with her choice in dessert today. And her dessert choice in two days will only depend on her dessert choice the day immediately before that. In other words, we’re in a world of short-term memory and #noragrets.</p><p>It turns out there are some pretty significant, real world functions that rely on Markov chains. One of these is Google PageRank, which, while not currently the only algorithm used by Google, is the most widely-known.</p><p>Google PageRank takes count of the available links to a page, and based on that number, it judges how important the website is. The underlying thought here is that more important websites are likely to have more links pointed at them from other websites.</p><p>Let’s say there are only five websites on the internet: A, B, C, D, and E. If a user is looking at page A, they will only see one link to page B. If they are on page C, there are three links; one to page A, one to page B, and one to page E.</p><p>As the user navigates through the pages, we can see the probability of the user ending up on each page. If a user begins on page A, they will only be able to go to page B. If they begin on page C, they’ll have a 33.333333% chance of choosing to click through to page A, B, or E. I could keep going here, but <a href="http://blog.kleinproject.org/?p=280">this awesome blog</a> breaks it down with much more expertise. You get the gist.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/714/1*G7D0_DQJ5FaFo3Do0NYXxg.png" /><figcaption>from <a href="http://blog.kleinproject.org/?p=280">http://blog.kleinproject.org/?p=28</a>0</figcaption></figure><p>There are also quite a few funny and completely useless implementations of Markov Chains like <a href="https://www.headlinesmasher.com/best/all">Headline Smasher</a>, <a href="http://joshmillard.com/garkov/">Garkov</a>, and <a href="https://twitter.com/markovtop100?lang=en">@MarkovBot</a>. If you’re feeling ambitious, you can choose from several free step-by-step tutorials and create your own Twitter bot. A quick Google search will put you on the right path.</p><p>I played around with the <a href="https://github.com/zolrath/marky_markov">Marky Markov generator</a> using the <em>Ace Ventura: Pet Detective </em>script as my dictionary. Because the script included all of the elements of a movie script, and not just the dialogue, the quotes generated weren’t particularly interesting or funny. They were just kind of strange.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3xPGsFkZcwzTx0D-55rtZA.png" /><figcaption>Sure, why not?</figcaption></figure><p>I tried again with a dictionary only based off of quotes from <em>Dumb &amp; Dumber</em>. Still, it wasn’t that entertaining.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/690/1*LfxjhZDSJST6m1tJveilww.png" /></figure><p>However, if you take a look at some of the cleverly-built examples on the web, like <a href="https://kingjamesprogramming.tumblr.com/">King James Programming</a>, there could be laughs, or at least subtle amusement, in store.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/638/1*Dk55yXtMn1KrqVZzKpB-OA.png" /><figcaption>from <a href="https://kingjamesprogramming.tumblr.com/">https://kingjamesprogramming.tumblr.com/</a></figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/690/1*TMi8q-bRq9csblahEVCfBw.png" /><figcaption>from <a href="https://kingjamesprogramming.tumblr.com/">https://kingjamesprogramming.tumblr.com/</a></figcaption></figure><p>One day, when I have more time to devote to extra-curricular coding projects for fun, I’ll return to Markov chains. Until then, I’ll continue to appreciate <a href="https://twitter.com/pelafina_lievre">weird Twitter accounts</a> that closely resemble my mom’s texts when she uses the Audio Message functionality on her phone (love you, mom).</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a4af1ef4af7a" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>