<?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 Brandon Aaskov on Medium]]></title>
        <description><![CDATA[Stories by Brandon Aaskov on Medium]]></description>
        <link>https://medium.com/@brandonaaskov?source=rss-cffae3791947------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*LMc_xLqp0O7_F-9aVEZoyA.png</url>
            <title>Stories by Brandon Aaskov on Medium</title>
            <link>https://medium.com/@brandonaaskov?source=rss-cffae3791947------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 05:44:48 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@brandonaaskov/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[Netflix Invented and Killed the Best Rating Scale Ever]]></title>
            <link>https://medium.com/@brandonaaskov/netflix-invented-and-killed-the-best-rating-scale-ever-8dd22248caf9?source=rss-cffae3791947------2</link>
            <guid isPermaLink="false">https://medium.com/p/8dd22248caf9</guid>
            <category><![CDATA[rating-system]]></category>
            <category><![CDATA[netflix]]></category>
            <dc:creator><![CDATA[Brandon Aaskov]]></dc:creator>
            <pubDate>Tue, 27 May 2025 10:02:39 GMT</pubDate>
            <atom:updated>2025-05-27T10:02:39.349Z</atom:updated>
            <content:encoded><![CDATA[<p>Let’s not bury the lede. This was the scale:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*J8F61cKR1Jtf66HGGphv3w.jpeg" /><figcaption>This image came from an article also supporting my theories here, but with big words like “bayesian.” (<a href="https://priorprobability.com/2017/07/29/the-ethics-of-bayesian-voting/">https://priorprobability.com/2017/07/29/the-ethics-of-bayesian-voting/</a>)</figcaption></figure><p>Five stars. No halfsies. Tied directly to emotion. And it was <em>perfect</em>.</p><h3>The Genius of the Emotional Scale</h3><p>You either liked it or you didn’t. That’s the brilliance.</p><p>There’s no true “middle” option in this scale — which sounds like a design flaw, until you realize how well it mirrors how people actually feel. If you’re not sure how you feel about a movie, you’re <em>already leaning</em>. A little good? It’s a 3. A little bad? It’s a 2. You have to pick a lane, and that decision makes your feedback valuable.</p><p>Contrast that with the lazy binary systems we get now — thumbs up or thumbs down. The emotional resolution is completely gone. And when that goes, so does the clarity in recommendations.</p><p>Human emotion isn’t binary. It’s squishy, inconsistent, irrational. Netflix <em>knew</em> that. They built something that respected it. And then they tossed it in the trash.</p><h3>They Really Cared — Until They Didn’t</h3><p>Back in 2006, Netflix announced a $1M prize: if you could improve their recommendation engine (Cinematch) by 10%, the money was yours. They <em>wanted</em> better recommendations. They <em>wanted</em> emotional granularity. The stars mattered.</p><p>That competition — the Netflix Prize — became legendary. Teams from around the world fought over decimal points of accuracy. Mergers happened. Rivalries formed. Code was optimized to hell and back. And eventually, a team hit the 10.05% mark. The prize was awarded. Confetti, probably.</p><p>And then… nothing.</p><p>No second competition. No long-term follow-up. And eventually, no more 5-star ratings.</p><p>What did we get instead?</p><h3>The Dumbing Down of Ratings</h3><p>Netflix ditched stars for thumbs. Just:<br> 👎 I hated it<br> 👍 I liked it<br> 👍👍 I liked it <em>a lot</em></p><p>Cool, so now we’ve got a binary system… plus a weird bonus round. What even is the second thumb? Do I love it? Do I recommend it? Do I want to marry it?</p><p>This isn’t a rating system. This is a <em>marketing filter</em>. It tells Netflix what to push. And guess what gets pushed? Netflix originals. Not the best content. Just the most brand-aligned.</p><p>They took something honest and useful and replaced it with something <em>convenient</em>. Convenient for them. Not for you.</p><h3>Why Half Stars Suck</h3><p>Some folks argue for more precision: “Just add half stars!”</p><p>No.</p><p>Let’s do a little math:</p><ul><li>Full-star system (1–5): 5 choices.</li><li>Add half-stars: now it’s a 9-point scale.</li><li>Add quarter-stars? Now you’re manually tuning decimal levels of emotion like a psychopath.</li></ul><p>Half stars are just indecision with better PR. They inflate the scale, dilute meaning, and let people dodge commitment. If you can’t decide between 3 and 4 stars, congratulations — that’s what <em>3</em> is for. You liked it, but didn’t really like it. Done.</p><h3>Why You Should Steal This Scale</h3><p>If you’re building a product with ratings — a media site, a game, a podcast directory, a sandwich review app, I don’t care — use this model:</p><ul><li>5 points max</li><li>Emotionally anchored labels</li><li>No fractions</li><li>Make the user pick a side</li></ul><p>You’ll get better data. More honest feedback. And, weirdly enough, more user satisfaction. Because it turns out, people <em>like</em> knowing how they feel.</p><h3>Netflix Had It Right</h3><p>They built a rating system that helped people express how they felt. And then they killed it, for reasons that were probably more about “brand alignment” and quarterly projections than product quality.</p><p>But if you’re not chasing the next earnings call? If you actually care about giving your users a way to <em>say something real</em>?</p><p>Then give them a scale that feels like feelings.</p><p>Five stars. Tied to emotion. No halfsies.</p><p>That’s how it’s done.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8dd22248caf9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Silent Cinema: A Novel Business Idea That Should Absolutely Exist]]></title>
            <link>https://medium.com/@brandonaaskov/the-silent-cinema-a-novel-business-idea-that-should-absolutely-exist-3978b7d41b85?source=rss-cffae3791947------2</link>
            <guid isPermaLink="false">https://medium.com/p/3978b7d41b85</guid>
            <dc:creator><![CDATA[Brandon Aaskov]]></dc:creator>
            <pubDate>Sat, 24 May 2025 06:15:20 GMT</pubDate>
            <atom:updated>2025-05-24T06:29:16.586Z</atom:updated>
            <content:encoded><![CDATA[<p>I think we’ve all had that moment in a movie theater when the guy behind us starts whispering. Or worse — <em>commentating</em>. And as much as we want to pretend it’s fine, our brain starts tuning in just enough to ruin the experience.</p><p>But what if the only sound you heard in a movie theater… was the movie?</p><p>I’m not talking about turning the volume up to drown people out. I’m talking about flipping the entire audio delivery model on its head. A quiet theater. No speakers. No booming bass to shake your chair. Just a pair of headphones and a film.</p><p>Turns out, it’s not only technically doable — it’s a really good business idea.</p><h3>The Premise</h3><p>The core concept is simple: eliminate the shared speaker system and deliver high-quality audio <em>directly</em> to each individual seat. You bring your own headphones (or rent/buy a pair at the counter), plug in, and boom — you’re in your own isolated, immersive audio world.</p><p>This solves a <em>ton</em> of problems:</p><ul><li>Noise complaints? Gone</li><li>“My seat’s too close to the left surround speaker”? Doesn’t matter</li><li>Accessibility? Way easier with direct audio + caption options</li><li>Maintenance of giant speaker arrays? You’ve offloaded it to people’s heads</li></ul><p>But let’s talk about how you’d actually make that work.</p><h3>Audio Distribution: The Journey from Projector to Headphones</h3><p>Here’s what the silent theater architecture looks like, if you break it down piece by piece.</p><h4>Step 1: Extract the audio</h4><p>Modern movie projectors or DCP servers output video and multichannel audio over HDMI. We use an HDMI audio extractor that pulls out stereo LPCM audio while still passing video on to the projector. That stereo feed becomes our master audio source.</p><h4>Step 2: Rack-mount it</h4><p>In the projector room, the audio feed runs into a rack-mounted audio interface. That connects via USB to a small rack server, which reads in the stereo signal and handles audio distribution and virtual surround (optional).</p><h4>Step 3: Push it over the network</h4><p>The server slices the audio into small packets and pushes them over Ethernet to each seat. Each packet includes a timestamp for when it should be played. This ensures all seats stay in sync, even if there’s a slight delay or packet gets momentarily held up. We use TCP here because we’re running on a closed, wired network and can afford the reliability.</p><h4>Step 4: Receive + play it</h4><p>Each seat has its own RP2040-based circuit board (specifically, Adafruit’s QT Py RP2040 with their amp add-on). It receives the TCP audio packets, buffers about 100ms of audio to handle jitter, and plays them back via a 3.5mm headphone jack.</p><p>Every seat is powered via Power-over-Ethernet (PoE), which makes cabling clean and simple. We use shielded Cat6 to avoid audio jitter and ground loops.</p><p>Each seat also has:</p><ul><li>A rotary encoder for volume control</li><li>An OLED screen showing your volume and optional audio offset</li><li>A button or two for adjusting that offset or cycling audio profiles</li></ul><p>If something goes wrong with a seat, it fails <em>alone</em> — not as part of a whole row. Each seat has a unique network ID, and logs join/leave events for reliability tracking.</p><p>And yes, this system works with any headphones. Even your old beat-up Sony MDRs from 2005.</p><h3>What About Surround Sound?</h3><p>We’re not doing Dolby Atmos here. That’s licensing hell and total overkill. What we <em>can</em> do is virtual surround — downmixing 5.1 into stereo with HRTF (head-related transfer functions) so it still sounds spatial. Open-source tools like OpenAL Soft make this completely achievable.</p><p>You get that sweet immersive feeling of directionality through plain stereo. That’s basically what an Astro Mixamp does, and you don’t need $300 headphones to appreciate it.</p><h3>Skip the Vending Machine. Go Human.</h3><p>I considered one of those headphone vending machines like you see at airports. But the better play? A real human who understands gear.</p><p>Someone who can ask if you’re just trying to watch the movie or if you’re trying to <em>hear</em> the movie. Someone who can point you to $10 wired earbuds <em>or</em> walk you through why someone dropped $90 on closed-back ANC cans.</p><p>You don’t need a vending machine. You need a guide.</p><h3>But Why Stop at Audio?</h3><p>This system makes it <em>much</em> easier to build in other features, like real-time captions. With a WebSocket feed and some AR glasses (eventually), you could render synced subtitles in the wearer’s vision.</p><p>This can’t be built today for lack of SDK support in AR glasses. The Apple Vision Pro is too expensive for anyone to own. But it’s coming. And the groundwork is already there: we’re really close from being able to build this for real. As a teaser, it would be fed the captions file that is used for viewers with accessibility issues, and piped into the AR glasses so you get subtitles during your movie, but only you get them.</p><h3>Closing Thoughts</h3><p>The movie theater industry is hurting. Streaming is better at convenience. Home theaters are better at comfort. Silent Cinema is a shot at being better at <em>experience</em>.</p><p>And it does that by leaning into the thing theaters haven’t done in decades: rethinking the fundamentals.</p><p>This is an experience-first, tech-forward, comfort-obsessed model that strips away the noise — literally — and puts the focus back on the film.</p><p>If I sound like I’ve already started building this… it’s because I̶ ̶h̶a̶v̶e̶ ChatGPT fluffed up that last line for me without even asking.</p><p><em>Note: this article was completed using ChatGPT, but the entire brainstorm session was very legitimate. </em><a href="https://chatgpt.com/share/683163ad-29b8-800e-9598-3ceb6c332a23"><em>For the sake of transparency, you can see the entire chat here</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3978b7d41b85" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Sveltefire Auth Examples]]></title>
            <link>https://medium.com/@brandonaaskov/sveltefire-auth-examples-d584c7644f24?source=rss-cffae3791947------2</link>
            <guid isPermaLink="false">https://medium.com/p/d584c7644f24</guid>
            <category><![CDATA[authentication]]></category>
            <category><![CDATA[sveltekit]]></category>
            <category><![CDATA[fireship]]></category>
            <category><![CDATA[sveltefire]]></category>
            <dc:creator><![CDATA[Brandon Aaskov]]></dc:creator>
            <pubDate>Sat, 13 Apr 2024 15:13:32 GMT</pubDate>
            <atom:updated>2024-04-15T14:50:50.154Z</atom:updated>
            <content:encoded><![CDATA[<p>I’m working on a project right now and using Sveltekit and Fireship’s Sveltefire libraries. His docs are good and minimal, but only shows the most basic auth examples. If you’re stuck at the point that I was, you’ve probably just missed the incredibly helpful <strong>getFirebaseContext</strong> method.</p><p>Without further ado, here is a very basic example that shows two different ways to use the Sveltefire stores:</p><pre>// +page.svelte<br>&lt;script lang=&quot;ts&quot;&gt;<br>  import type { Auth } from &#39;firebase/auth&#39;<br>  import {<br>    GoogleAuthProvider,<br>    signInWithPopup,<br>    signOut<br>  } from &#39;firebase/auth&#39;<br>  import {<br>    getFirebaseContext,<br>    userStore,<br>    SignedIn,<br>    SignedOut<br>  } from &#39;sveltefire&#39;<br><br>  // these two lines are the key to using sveltefire stores <br>  // from inside the component&#39;s script tag<br>  const { auth } = getFirebaseContext()<br>  const user = userStore(auth as Auth)<br><br>  async function signInWithGoogle (auth: Auth) {<br>    const provider = new GoogleAuthProvider()<br>    provider.addScope(&#39;profile&#39;)<br>    provider.addScope(&#39;email&#39;)<br>    await signInWithPopup(auth, provider)<br>  }<br>&lt;/script&gt;<br><br>&lt;nav&gt;<br>  &lt;p&gt;Data from component (not the store): {$user?.email}&lt;/p&gt;<br>  &lt;!-- <br>    components can bind the store using let: so you can <br>    leverage it directly in the markup <br>  --&gt;<br>  &lt;SignedIn let:user let:auth&gt;<br>    &lt;p&gt;Data from store: {user.email}&lt;/p&gt;<br>    &lt;button on:click={() =&gt; signOut(auth)}&gt;<br>      Sign Out<br>    &lt;/button&gt;<br>  &lt;/SignedIn&gt;<br>&lt;/nav&gt;<br><br>&lt;SignedOut let:auth&gt;<br>  &lt;main&gt;<br>    &lt;button on:click={() =&gt; signInWithGoogle(auth)}&gt;<br>      Sign In with Google<br>    &lt;/button&gt;<br>  &lt;/main&gt;<br>&lt;/SignedOut&gt;</pre><p>If you’re trying to use signInWithRedirect with Google auth, but on refresh/redirect the user appears logged out, it’s likely just due to the browser you’re using. Google plays favorites with Chrome but most other browsers automatically block third-party cookies. <a href="https://firebase.google.com/docs/auth/web/redirect-best-practices">Here’s the help doc you’ll need that lists out some options if you really need to use a redirect over a pop-up</a>.</p><p>As noted in that same doc:</p><blockquote>Starting June 18 2024, implementing one of the options will be required for redirect sign-in to work on Google Chrome M115+. This is already required on Firefox 109+ and Safari 16.1+.</blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d584c7644f24" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to Brainstorm What You Can Do With NFTs]]></title>
            <link>https://medium.com/@brandonaaskov/how-to-brainstorm-what-you-can-do-with-nfts-484a464072c9?source=rss-cffae3791947------2</link>
            <guid isPermaLink="false">https://medium.com/p/484a464072c9</guid>
            <category><![CDATA[marketing]]></category>
            <category><![CDATA[strategy]]></category>
            <category><![CDATA[process]]></category>
            <category><![CDATA[nft]]></category>
            <category><![CDATA[web3]]></category>
            <dc:creator><![CDATA[Brandon Aaskov]]></dc:creator>
            <pubDate>Mon, 18 Jul 2022 23:19:35 GMT</pubDate>
            <atom:updated>2022-07-20T13:21:05.220Z</atom:updated>
            <content:encoded><![CDATA[<h3>TL;DR</h3><p>Start thinking about the utility an NFT has as that drives the price/value the most: what does it offer beyond simply owning it? Think like a speculator, not like a fan. What would cause you to want to buy one of these NFTs even as the price went up and up? What does the NFT get you now and in the future that causes the price to keep going up? And more interestingly, what kind of value would the NFT have that would make an owner want to burn, fractionalize or borrow/lend the NFT?</p><h3>Getting In the Right Mindset</h3><p>NFTs as an asset class is still very new to many people, so thinking about <em>how</em> to leverage them can be hard. After all, you don’t know what you don’t know. A common mistake we see made is getting wrapped up in the “how.” I do this all the time myself and it’s hard to break away from. Suffice to say that many marketplaces and tools exist today that simplify the “how.” Instead, I want you to think about this:</p><blockquote><em>“People don’t buy what you do, they buy why you do it.”</em></blockquote><p>Please watch a short section of this video (1:36–5:00) to inspire you and get the creative juices going. I promise it’s a very worthy 3m24s of your time.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2Fqp0HIF3SfI4%3Fstart%3D96%26feature%3Doembed%26start%3D96&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3Dqp0HIF3SfI4&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2Fqp0HIF3SfI4%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/3d44eea9395f34a31f51ac696efe105c/href">https://medium.com/media/3d44eea9395f34a31f51ac696efe105c/href</a></iframe><p>When it comes to thinking about what a collection of NFTs could offer, you need to get your brain in the right mindset first. There are three major categories of NFTs: utility, collectible and art. What we’ve seen to date is really more like a Venn diagram of these categories.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DhHHVgoJDCd6Ej4Li1XuZg.jpeg" /><figcaption>The three major categories of NFTs</figcaption></figure><p>I think it’s safe to say that from here on out, the NFTs you offer will need to have <em>some</em> kind of utility to have any meaningful long term value. The hard part about NFTs is getting them into a customer’s wallet, but the utility an NFT has can evolve <em>forever</em>. Maybe today it’s exclusive access to a channel in your Discord server. And then 6 months down the line, you could have a new campaign that allows those NFT holders access to a high profile experiential event (for example). The utilities your NFTs have can change over time. You can add and remove benefits as you see fit as a business, but it’s worth mentioning that transparency is <em>absolutely critical</em> in this web3 world.</p><p>NFTs-as-art (think <a href="https://www.theverge.com/2021/3/11/22325054/beeple-christies-nft-sale-cost-everydays-69-million">Beeple</a>) is a category very specific to artists and art aficionados. Art NFTs are typically 1 of 1. Most collectible NFTs have unique art too, but it’s more about differentiating the rarity of the NFT than highlighting the art itself. For the rest of your brainstorming, let’s put both of these categories out of our heads and focus on the one that really matters most.</p><h4>Utility</h4><p>When it comes to the audience who <em>wants </em>NFTs, you have to think about two distinct audiences. Again, another Venn diagram suits the bill:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*l1_QKijvuLzOV_gmNm1VOA.jpeg" /><figcaption>The audience of NFT buyers, broken into two broad categories</figcaption></figure><p>You’ll notice that I didn’t add a third category, the “simply curious”, as they don’t provide long term value. Your fans have always been there and this is a new way to tap into that audience and potentially grow it further. Sometimes, fans collect just to collect: that’s enough for them, but if you only focus on this audience, you’ll be missing out on a much larger opportunity.</p><p>Speculators, on the other hand, are a new breed that’s been ushered in rapidly thanks to NFT adoption. Web3 is all about exchange of value so it’s not surprising that even your existing fans are becoming speculators too. If the NFT doesn’t have utility today, you need to be <em>very</em> transparent about what it will have in the future. None of this vague “exclusive benefits and more!” stuff that plagues so many projects. You can be vague about the far-future, but you should explain your plans for at least the next 6 months, ideally a full year. Typically, these roadmap milestones are broken down by quarter.</p><p>Let’s finish this off by talking about the two major DeFi concepts that will keep your NFTs relevant and popular for years to come. For the rest of this, put the mentality of your fans aside since that will come naturally for you later. If your NFTs have value to non-fans, you’re off to a good start. Time to start thinking like…</p><h4>Speculators</h4><p>Before switching our brains to the mind of a speculator, let’s get rid of the “non-fungible” part of NFTs. With collectible NFTs we saw that the rarity attributes drove that value. For example a Bored Ape with a “Trippy Captain’s Hat” is more rare than one with a “Cowboy Hat”. Thinking about rarity attributes like this can make this exercise harder than it needs to be, so let’s toss that out for now.</p><p>When it comes to utility, we’re really talking more about a fungible token, meaning that one token is basically the same as another if they both can be swapped but still have the same value. Since the art tied to it would be the only thing making it an non-fungible, and we’re focused on utility, for the rest of this brainstorm let’s try to think about a potential NFT collection as offering the same kind of value across all tokens in the collection.</p><h3>NFT Interactions</h3><p>There are some straightforward ways to interact with NFTs with the tools we have today. Understanding how these mechanics work will greatly help you think up some interesting ideas about an NFT collection you might want to offer.</p><h4>Minting &amp; “Merging”</h4><p>Minting is the act of creating the NFT. Many NFTs are typically created ahead of time, but in some cases the NFT is minted at the time of purchase (using smart contracts). In the latter case, the buyer pays all the transaction fees to create and transfer the NFT into their wallet.</p><p>“Merging” is a little more interesting, and perhaps the wrong term but works for our needs. This is when you combine one or more NFTs together to create something new. The Bored Ape Yacht Club popularized this with their “Mutant Serum” NFTs that allowed BAYC owners to mutate their Bored Ape into a new visual form. When combining NFTs, it’s up to you if owners have to burn those NFTs or not. In most cases it’s helpful to burn the NFTs to simplify the code required to prevent fraud/re-use.</p><h4>Buying &amp; Selling</h4><p>You can’t just “build it and they will come (to buy it).” What’s going to make buyers <em>want</em> these NFTs so much that it’s going to push the price up? This comes down to the utility roadmap for your NFTs and how potential buyers will view it. Though royalty fees on resales can be a great source of revenue, that shouldn’t be your focus right now. If the NFTs don’t have enough utility, the only thing that will affect their <a href="https://www.investopedia.com/terms/p/price-action.asp">price action</a> will be hype, and that’s unsustainable.</p><p>Remember, wearing our speculator hat, we don’t care about the underlying brand/art/mission. We just want to make money. As an example, if Virgin Galactic sold NFTs as tickets to get access to a rocket launch, and one NFT holder would win a trip to space, that could cause some heated speculation on the value of those NFTs since a trip to space is around $250,000 to $500,000.</p><p>The Willy Wonka-style “golden ticket” is just one example. A slight tweak on that idea is that a collection of NFTs could later be broken down into three tiers: bronze, silver and gold, with each tier having differing utility (a la Kickstarter rewards). When the NFTs are first created they all have the <em>potential</em> for future value, and that can drive a lot of speculation.</p><p>Buying and selling is the most straightforward way to interact with an NFT. Ideally your NFTs will have clearly defined benefits before even being sold, as well as a roadmap for what’s to come. By tying in surprise elements to the NFTs that are already in the market, you can help drive the price action.</p><h4>Burning &amp; Redeeming</h4><p>Burning an NFT is the act of “destroying it” (although technically it’s just sending it to a wallet address that doesn’t exist and thus can’t be retrieved). The act of burning a token is helpful for economic reasons: you can balance inflation and deflation when you have the ability to mint and destroy the token supply. For NFTs the burning mechanism can offer interesting opportunities. The simplest to consider is the exchange of a physical good (e.g. apparel), or one-time-access to something (e.g an event). These NFTs are one-time-use but can hold value until they’re burned.</p><p>Burning an NFT can be good for you to control fraud so that there can’t be a double-redemption problem, but that’s not thinking like a speculator is it? As a speculator, the NFT isn’t any good once it’s burned so it has to have value when in that unused, limbo state. To keep the value going up until the moment it’s redeemed, you’ll have to keep adding benefits. If those are real-world redeemable benefits, you’ll want to keep adding those periodically to keep driving up the value. If it’s a combination of real-world and digital benefits, that drives speculative value even further because now it’s a more difficult decision to burn the NFT as the digital benefits would be lost.</p><h4>Fractionalizing</h4><p>Fractionalizing an NFT is simply breaking it up into smaller chunks. The only way this can work and still hold value is if:</p><ol><li>You decide to support those fractions and allow some kind of utility for them</li><li>Another project in the ecosystem offers value for those fractionalized NFTs</li></ol><p>Fractionalized NFTs can be more difficult to think about how to leverage effectively. Using the Virgin Galactic example again, one idea would be to allow owners to fractionalize their NFT <em>up to</em> 10 times, but those fractionalized NFTs would only get you into the pre-party or something of similar reduced worth. Or perhaps you could redeem the NFT at the event for an entry to win a trip to space, or fractionalize it up to 10 times in the form of drink tickets. The sky is the limit in what you can support, but thinking about a <em>limit</em> on the fractionalization will help give some helpful constraints to your ideas.</p><h4>Lending &amp; Borrowing</h4><p>Lending and borrowing NFTs is a relatively new concept that only came about with the rise of highly priced NFTs in 2021. In my opinion, this is the most fun category to try to come up with ideas around.</p><p>As a speculator I could lend an NFT to make money: that side of the equation is simple. But why would someone <em>want to</em> <em>borrow it</em>? Perhaps it gives access to a Discord channel where someone can ask that one burning question they’ve been dying to get answered. Perhaps it’s a season ticket to something sportsball and you want to give someone access for one event. Thinking about the length of time for the lending period can open up interesting concepts as well. For me, the ideas derived from this category flow too quickly. Typically, the best ideas that come out of this will also pair nicely with the buying/selling ideas as they’re so closely related.</p><h4>The DAO Difference</h4><p>A DAO is a “decentralized, autonomous organization.” Basically it’s a fully decentralized company where all the owners of the tokens operate the business together. A shared company with a shared bank account and a somewhat shared vision (established through governance votes). This is the extreme end of the spectrum regarding project concepts. To operate as a DAO, a lot of things need to be considered (beyond legal requirements). It’s probably too much for this article, but if you’re dying to hear more, here’s just one example: <a href="https://linksdao.io/">LinksDAO</a> sold their NFTs as memberships for a golf course that doesn’t exist yet. The owners of the NFTs are the owners of the DAO and get to decide where to buy a golf course with all the money generated from that NFT sale. Those members get exclusive access to the course, and as an organization they can collectively vote and decide on the path for the future.</p><h3>Wrapping Up</h3><p>If you find yourself swimming in a sea of ideas, good job! Try to get it down to 3–5 ideas. If you’re struggling with that, take the Hot-Or-Not approach and compare two ideas at a time. Which of the two is better? Put aside the lesser idea, and compare again. If you find yourself struggling still, please reach out (<a href="https://engineering.deptagency.com/">DEPT</a>, <a href="https://www.linkedin.com/in/brandonaaskov/">me</a>) and we can help you get there: that’s what we do 🙂</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=484a464072c9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to Use AirPlay with Javascript]]></title>
            <link>https://medium.com/@brandonaaskov/how-to-use-airplay-with-javascript-5dadce53263a?source=rss-cffae3791947------2</link>
            <guid isPermaLink="false">https://medium.com/p/5dadce53263a</guid>
            <category><![CDATA[airplay]]></category>
            <category><![CDATA[javascript]]></category>
            <category><![CDATA[tutorial]]></category>
            <dc:creator><![CDATA[Brandon Aaskov]]></dc:creator>
            <pubDate>Sun, 10 May 2020 18:28:48 GMT</pubDate>
            <atom:updated>2020-05-10T18:28:48.574Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*v2jOTwndgUpef811-e4hbg.jpeg" /><figcaption>I promise you it works, and it’s not hard</figcaption></figure><p>I had to do more searching on the web than I wanted to, so hopefully this will save future travelers some time.</p><p>It’s actually very simple, and instead of wasting words, here’s the snippet:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/6d277427b5b9603eba5a2d184f8c907f/href">https://medium.com/media/6d277427b5b9603eba5a2d184f8c907f/href</a></iframe><p>If you would like to see a working demo, <a href="https://codepen.io/brandonaaskov/pen/eYprLMv?editors=1000">load up this Codepen</a> in either Safari or on your mobile device.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5dadce53263a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to Test Nuxt Stores with Jest]]></title>
            <link>https://medium.com/@brandonaaskov/how-to-test-nuxt-stores-with-jest-9a5d55d54b28?source=rss-cffae3791947------2</link>
            <guid isPermaLink="false">https://medium.com/p/9a5d55d54b28</guid>
            <category><![CDATA[nuxtjs]]></category>
            <category><![CDATA[vuex]]></category>
            <category><![CDATA[vuejs]]></category>
            <category><![CDATA[testing]]></category>
            <category><![CDATA[jest]]></category>
            <dc:creator><![CDATA[Brandon Aaskov]]></dc:creator>
            <pubDate>Sat, 18 Jan 2020 23:31:36 GMT</pubDate>
            <atom:updated>2020-01-18T23:43:51.402Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YUwr1vuE_l0x9S8fPRtp_Q.png" /><figcaption>This is not Photoshopped</figcaption></figure><h4><strong>The Quick Answer</strong></h4><p><em>If you just want to clone a repo or tinker in a sandbox, I get it.</em></p><p><em>Github: </em><a href="https://github.com/brandonaaskov/how-to-test-nuxt-stores"><em>https://github.com/brandonaaskov/how-to-test-nuxt-stores</em></a></p><p><em>CodeSandbox: </em><a href="https://codesandbox.io/s/github/brandonaaskov/how-to-test-nuxt-stores"><em>https://codesandbox.io/s/github/brandonaaskov/how-to-test-nuxt-stores</em></a></p><p>When you first install Nuxt, you have the option to install a test framework. When you do, it gives you a sample test… of a component. That’s great and all, but in most enterprise web apps you’re going to have the majority of your logic in your stores.</p><p>But… where are those examples? It’s almost as if this had never been considered by the nuxt team or community. This plagued me for nearly a year, until pimlie on the Nuxt Discord server helped me understand more about how to access and control what nuxt is building under the hood. Huge props to pimlie.</p><p>In short, this is now a solved problem. Let’s get into it.</p><h3><strong>The Problem</strong></h3><p>The great thing about nuxt is that it takes care of <em>a lot </em>of tooling for you. This is largely thanks to conventions, like how the pages directory becomes the routing, or how your stores directory becomes a bunch of <a href="https://vuex.vuejs.org/guide/modules.html">Vuex modules</a>.</p><p>In nuxt, you can get access to the store most commonly from <a href="https://nuxtjs.org/api/context/">the context object</a>, or by using the helpers that come with Vuex (e.g. mapGetters, mapMethods, mapActions). If your /stores/ folder has more than just an index.js file, then nuxt is building your stores into Vuex modules behind the scenes. It also surfaces that store on the context object for you.</p><p>So the problem becomes, how do we rebuild/structure our stores in a way that we can test them in the <em>same way </em>we use it in our pages and components? Do we write all that store building logic ourselves? How can we reuse what nuxt is already doing for us when it builds?</p><h3>The Solution</h3><p>There are two key parts here:</p><ol><li>using a setup file that runs once before Jest starts running any test suites</li><li>dynamically importing the built store into the test files</li></ol><p>We’re going to add a globalSetup reference in our jest.config.js file. I’ve named mine jest.setup.js and it lives right alongside the Jest config file.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/1db698f257b9976ee058742672d677b0/href">https://medium.com/media/1db698f257b9976ee058742672d677b0/href</a></iframe><p>The purpose of this setup file is to run once before any tests start running. This gives us a chance to run a nuxt build, which is the critical piece to this whole thing. We tell nuxt to <em>only</em> build the store by overriding all the config elements of what to build. Lastly and very important, we expose the path to that built store’s location as an env var. We’ll later be able to grab that in our tests:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/64614359689acbc325c50a8b981e2022/href">https://medium.com/media/64614359689acbc325c50a8b981e2022/href</a></iframe><p>And finally, the test itself. The piece to focus on is the setup: the beforeAll and beforeEach blocks:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/301cca0a3b0f5d2b54ad2888e967507c/href">https://medium.com/media/301cca0a3b0f5d2b54ad2888e967507c/href</a></iframe><p>In the beforeAll block, we leveraged that env var to dynamically import our built store. In the beforeEach block, we create a fresh store each time to keep our tests clean and independent.</p><h3>Wrapping Up</h3><p>I’ve linked to a Github repo and CodeSandbox at the top of the article, and I encourage you to play with it. The sample gists above were pulled from there. If you’re having trouble getting this to work in your codebase, try first bringing your stores and test files into this repo, so you can see it in a known working environment. That way you can isolate any errors that might be due to your test files before bringing them back to your codebase to get Jest configured correctly. Your mileage may vary, every codebase is different.</p><p>Good luck!</p><p>If you have any improvements you want to make, please submit a PR. I’d love that. And if you do, you should know that <a href="https://blog.rocketinsights.com/rocket-insights-makes-built-in-bostons-best-places-to-work-list-for-2020-2/">I work at Rocket Insights, we’re awesome, and we’re hiring</a> 🤓</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9a5d55d54b28" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Using Zoom with Cisco Gear]]></title>
            <link>https://medium.com/@brandonaaskov/using-zoom-with-cisco-gear-cdf1b7a15a93?source=rss-cffae3791947------2</link>
            <guid isPermaLink="false">https://medium.com/p/cdf1b7a15a93</guid>
            <category><![CDATA[zoom]]></category>
            <category><![CDATA[video-teleconferencing]]></category>
            <category><![CDATA[cisco]]></category>
            <category><![CDATA[webex]]></category>
            <dc:creator><![CDATA[Brandon Aaskov]]></dc:creator>
            <pubDate>Fri, 17 Jan 2020 22:32:11 GMT</pubDate>
            <atom:updated>2020-01-17T22:32:11.761Z</atom:updated>
            <content:encoded><![CDATA[<p>Our parent company, <a href="http://deptagency.com/">Dept</a> standardizes on Webex and they set us up with <a href="https://www.cisco.com/c/en/us/products/collaboration-endpoints/webex-room-kit-mini/index.html">some nice gear</a> so we could have some industry-grade teleconferencing. The problem? <em>Using</em> Webex is not fun, and it’s such a bad experience we all started creating and paying for our own personal Zoom accounts. To be clear, I’m not even referring to the hardware. Everything Webex feels like trying to teach my grandmother how to rotate a PDF: you’ll eventually get there but what an unnecessarily painful process.</p><p>After chatting with Zoom, we discovered we could re-use our pricey Cisco gear with our Zoom accounts as well as still use it for Webex meetings. This is clearly the best of both worlds, but there’s a lot of unclear documentation on how to achieve this. After playing with it for a couple of hours, I discovered it’s far simpler than it seems on the surface.</p><h3>Cloud Room Connector vs. Virtual Room Connector</h3><p>First, I should point out that this feature is an additional cost. I believe it’s $49/month at the time of this writing. For a while, Zoom only supported using a <a href="https://support.zoom.us/hc/en-us/articles/202134758-Getting-Started-With-Virtual-Room-Connector">Virtual Room Connector</a>. Setting up the VRC is a complicated process, and requires you run a virtual machine image on your internal network to support the connection between Zoom’s services and Cisco’s. Suffice to say, you can easily go in circles with this.</p><p>The <a href="https://support.zoom.us/hc/en-us/articles/201363273-Getting-Started-With-H-323-SIP-Room-Connector">Cloud Room Connector</a> on the other hand is <em>way</em> more straightforward. Once that’s been added to your plan, you’re done with all things “room connector.”</p><h3>Pro Setup Tips (Optional)</h3><p>This step isn’t required, but it’s going to make your life a lot easier. In your Zoom app under that little “New Meeting” dropdown arrow, make sure to select “Use my Personal Meeting ID (PMI)” and then mouse over the number and head to settings so we can change it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ui_yPzHCBBEmIjR8PiBKgA.png" /><figcaption>Using a PMI keeps the number consistent from meeting to meeting.</figcaption></figure><p>That PMI number can be anything so long as it’s _at least_ 10 digits long. What else is 10 digits? Your phone number! Just change it to your phone number so you can remember it forever without an extra thought. If that number is already taken… well that sucks, I’m sorry.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6C9HFK9ocae4V1z7MNrY6w.png" /><figcaption>I also turn off password requirements to make it easier.</figcaption></figure><h3>Joining With Cisco Hardware</h3><p>The last step is to actually use the camera with Zoom, and it’s pretty close to magic. I’ll use the number from the example above: 617–555–1212.</p><p>All you have to do is dial your-meeting-number@zoomcrc.com (e.g. 6175551212@zoomcrc.com). Depending on how the meeting was configured and if it’s started already, you’ll see either a splash screen on your TV or you’ll just auto-join the meeting. That’s it! There’s no actual change to the hardware itself, just that lovely little connector doing all the work.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=cdf1b7a15a93" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Using Multiple USB Mics on a Mac]]></title>
            <link>https://medium.com/@brandonaaskov/using-multiple-usb-mics-on-a-mac-27f7847a0560?source=rss-cffae3791947------2</link>
            <guid isPermaLink="false">https://medium.com/p/27f7847a0560</guid>
            <category><![CDATA[podcast]]></category>
            <category><![CDATA[audio-engineering]]></category>
            <category><![CDATA[microphone]]></category>
            <category><![CDATA[mac]]></category>
            <category><![CDATA[equipment]]></category>
            <dc:creator><![CDATA[Brandon Aaskov]]></dc:creator>
            <pubDate>Tue, 15 Jan 2019 18:45:58 GMT</pubDate>
            <atom:updated>2020-01-18T21:59:35.495Z</atom:updated>
            <content:encoded><![CDATA[<h3>Connecting Multiple USB Microphones on a Mac (probably for Podcasting)</h3><p>This was surprisingly hard to find, but the answer was straightforward in the end. I’m using Adobe Audition for this example, but the same solution works for GarageBand or any other audio editing app (at least, it should).</p><h3>The Problem</h3><p>If you’ve ever used an audio interface before then you’re used to being able to select multiple inputs from the interface, which is necessary for multitrack recording. That’s because the audio interface is providing the system clock to be the digital sample rate driver. Connecting multiple microphones to a USB hub will only allow you select <em>one </em>of those microphones to be the master clock, which causes a problem. So no matter how many USB mics you connect, you can only use one in your multitrack software. Bummer.</p><h3>The Solution</h3><p>Apple already has a built-in solution for this. In your Applications/Utilities folder is an app called Audio Midi Setup, and it’s got a cool little feature in it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/350/1*M_Wkaf0dMfU2aq8kY4mrmw.jpeg" /><figcaption>I’ve never needed this software ever, up until this multiple USB mic thing</figcaption></figure><p>In the bottom left of the Audio Midi Setup app is a little plus icon: click it and select “Create Aggregate Device”.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/470/1*IbxmJQ5BJC4QosFEK_U5dw.png" /><figcaption>Click that one. You’re going to aggregate your USB mics into one device.</figcaption></figure><p>In my case, I had three USB Blue Yeti microphones I wanted to work with. Here, you can see I’ve selected both channels from each microphone.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/596/1*3AFydMwr42quy93NS8iEtw.png" /><figcaption>This is where the magic happens. Now all my mics are grouped into one device.</figcaption></figure><p>In Adobe Audition (or equivalent), select your default input to be the “Aggregate Device” now showing up as one of your inputs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/590/1*4_H0qMSE5d6oz58A3P-HCg.png" /></figure><p>Now you can select the stereo (or mono) inputs for each of your tracks. You’re done!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*q0BpvjxPS_CdrUd1zElJsw.jpeg" /><figcaption>That’s it! It’s so simple (if you know where to start).</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=27f7847a0560" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Learning Anything via Middle-Out]]></title>
            <link>https://medium.com/@brandonaaskov/learning-anything-via-middle-out-99cb8955d25b?source=rss-cffae3791947------2</link>
            <guid isPermaLink="false">https://medium.com/p/99cb8955d25b</guid>
            <category><![CDATA[front-end-development]]></category>
            <category><![CDATA[silicon-valley]]></category>
            <category><![CDATA[learning]]></category>
            <category><![CDATA[learning-to-code]]></category>
            <dc:creator><![CDATA[Brandon Aaskov]]></dc:creator>
            <pubDate>Tue, 11 Dec 2018 20:12:29 GMT</pubDate>
            <atom:updated>2018-12-11T20:12:29.402Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/700/1*FYyR8YvV94Q_WeW-t6VRrA.jpeg" /><figcaption>This is not what this article is about :)</figcaption></figure><p>This is not about <a href="https://www.youtube.com/watch?v=l49MHwooaVQ">Pied Piper’s famous (and fictional) “middle-out compression.”</a> Sorry to let you down. Instead, I’m making a point: that we learned “how to learn” the wrong way, and it’s not our fault.</p><p>When we (as humans) were doing our formative learning, we learned from the very baseline up. “Here are letters, these are numbers.” It was only later that we started to combine them to form more complex ideas, like language and math. And that’s all well and good for toddlers, but that’s not a great way to learn complex stuff <em>fast</em>.</p><p>From the time we start school, we’re all forced to start with chapter one and never skip ahead. You have to know addition before you can multiply, multiplication before exponents, etc.</p><p>I’ve found that’s a terrible way to learn almost any hobby or subject (except math, I guess). We’ve been taught to think that learning in this way will net us a solid understanding of the foundations.</p><p>We shouldn’t be learning that baseline first. If I want to learn how to change the spark plugs in my car, I <em>don’t </em>need an understanding of internal combustion first. If I want to try my hand at sewing a pair of cotton pajamas, I don’t first need to learn about the many types of fabric. And yet that’s how so many courses, books, and instructional videos start the process. It’s as if they’re teaching us an entire career instead of showing us how to sample one. Put another way, if I want to learn how to build a house I need to start with something smaller. But I don’t need to learn about all the different types of wood and screws to get moving.</p><p>For the sake of simple analogies, let’s assume software and house development are similar in complexity. You can start with the simplest shape (the <strong>sandbox</strong>), and work up to a medium sized project (a <strong>shed</strong>). One day you’ll have earned the confidence to build a <strong>house</strong>-sized project and beyond.</p><p>When it comes to learning a new subject, you should dive into the <em>middle </em>of that learning curve and work your way <em>out. </em><a href="https://codeburst.io/the-2018-web-developer-roadmap-826b1b806e8d">Codeburst wrote a great article on the 2018 web developer roadmap</a>. I’m going to borrow their image for learning front-end development to make my point.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*60PpLKHsPMDgZ9qZh1VegQ.jpeg" /><figcaption>Codeburst 2018 Front-End Developer Roadmap</figcaption></figure><p>This roadmap view makes it seem like you have to dive deeper into each subject to properly learn it. Even worse, it might leave you with an impression that you need to know most of this to be able to function as a front-end engineer. In reality, you should learn what you need to get by. Let’s use that same example again, and let’s say you wanted to learn React. Your journey would look more like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/690/1*H3TMOf6VwGN17svl3EaWtg.png" /><figcaption>Learning React from the middle, out</figcaption></figure><p>The downside to this method is that the learning curve is more complex. Because you’ll be learning <em>as you go, </em>each time you dive into some new piece, your progress will slow down. That can frustrate some. But learning in this way allows you to focus on only learning what you need <em>right now</em>. It also (typically) means that your sandbox/shed is going to look a little funky in some areas. It’s the nature of “the first project” because at that point, you don’t know what you don’t know. When you learn by doing, you get a real-world experience that teaching can’t match. It’s those insights that create better, future projects: more stability, lower cost, better use of time, etc. This is also a good way to become very quickly familiar with the tools and materials you’ll need to use to get it done.</p><p>So dive into the middle of the next project you need to learn. Don’t build something <em>just to learn. </em>Build something you actually need/want and don’t expect perfection (this is key). By the time you finish your first project, you’ll have insights that take much longer learning the traditional way. You won’t know it all, but you’ll know “enough to be dangerous”, as they say.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=99cb8955d25b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Ignore Your Burndown Charts]]></title>
            <link>https://medium.com/@brandonaaskov/ignore-your-burndown-charts-2e252c16e4f?source=rss-cffae3791947------2</link>
            <guid isPermaLink="false">https://medium.com/p/2e252c16e4f</guid>
            <category><![CDATA[agile]]></category>
            <category><![CDATA[process]]></category>
            <category><![CDATA[sprint]]></category>
            <category><![CDATA[scrum]]></category>
            <category><![CDATA[saas]]></category>
            <dc:creator><![CDATA[Brandon Aaskov]]></dc:creator>
            <pubDate>Fri, 27 Apr 2018 03:49:07 GMT</pubDate>
            <atom:updated>2018-04-29T17:34:53.064Z</atom:updated>
            <content:encoded><![CDATA[<p>Burndown charts are a popular aspect of the Sprint methodology and why wouldn’t it be: <a href="https://gph.is/28R1o2g">it’s a chart</a>. <a href="https://cl.ly/rCO1">Business people</a> love charts! Ironically, the goal of this simple chart is to get all the stories done and reach 0 in time, and that flies in the face of everything it means to be agile and productive.</p><h3>A Quick Recap</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*t7AubCNLIsSB-lN6LNERkA.png" /><figcaption>Sample burndown chart from Atlassian (makers of JIRA)</figcaption></figure><p>This is a burndown chart. The gray line is the “ideal” trending line for stories to be “<a href="https://gph.is/1U1c6Fn">burnt down</a>” (i.e. make their way to complete/done). The red line is reality. It’s well-known that the ideal line is never really achieved. In this example, the team started with 40 points of work and ended with about 8 points remaining.</p><h3><strong>The Argument</strong></h3><p>The basic problem with a burndown chart is that its goal is to get to 0, which implies that everyone working on stories will finish their work at some point. But it doesn’t account for the <em>next</em> work item that needs to get done. Typically this means pulling in a story from the backlog. While that sounds simple, not all teams work that way since that kind of thing can make for one ugly burndown chart.</p><p>When stories get pulled into a sprint from the backlog, it causes spikes in the trend line. It also risks that this new, pulled-in story won’t be completed by the end of the sprint and thus the trend line will <em>never </em>reach zero. In some businesses, never reaching zero in a burndown chart is a bad sign. Even the chart itself implies this is a bad thing. This creates a strange atmosphere of “the perfect sprint” where everyone completes their work in a timely manner and all work is done just in time for the sprint to end. But the reality is that people will finish their work at different intervals, and just because a story is still open, doesn’t mean anyone can just work on it. Theoretically, if we lived strictly by the chart’s rules, we’d all stop working when we had nothing left to pick up, <a href="https://cl.ly/rCSp">just waiting patiently for the sprint to end</a>. No extra stories pulled in that can’t be finished in time: none of that funny business here!</p><h3>Velocity Chart</h3><p>Everyone loves a good chart with a trending line, but instead of <a href="https://cl.ly/rDUa">focusing on a line to no-more-worksville</a>, let’s <a href="https://cl.ly/rBde">focus on a line that highlights productivity</a>. Keeping track of a team’s velocity should be a nice, steady line. But spikes/anomalies in this graph actually mean something significant. Just resist the urge to over-estimate when grooming stories just to see that line go up. The goal is to keep it steady.</p><h3>Engineers, Not Robots</h3><p>The shape of a team is one of the biggest reasons that a burndown rate will never be ideal. Let’s say we have a team that includes 3 engineers:</p><ol><li><a href="https://cl.ly/rC3v">Dade: there since the beginning (legacy knowledge)</a></li><li>Andrea: came aboard two months ago (still learning, but confident)</li><li>Seth: brand new to the team</li></ol><p>In this common scenario, the person writing the stories will most likely <em>not </em>be the same person working on the story. Sometimes certain stories are best suited for certain engineers, which means the average team velocity is kind of hard to gauge. Dade’s velocity would be above average, Andrea’s velocity is probably close to average, and Seth’s average is close to 0. Dade’s going to have a busy sprint. There’s also a good chance that stories are light on details since <a href="https://cl.ly/rCgl">a lot of the pieces in play are in the head of the person that wrote the story</a> (dammit Dade!). Working on stories and <em>not </em>pulling them into a sprint will negatively affect this average velocity calculation. You should care <em>a lot </em>more about your velocity being accurate than an impossible-to-achieve line in a burndown chart.</p><h3>Sprint Purgatory</h3><p>Right near when the sprint ends and many of the last stories are making their way to completion, there’s a gap of working time for many engineers on the team. Once a story leaves the engineer’s workflow and goes to QA/staging/prod, their involvement with the story typically drops dramatically. If this happens at the end of the sprint, and the team has been encouraged to <em>not </em>pull in a story unless it can be completed in time, it’s uhh…. <a href="https://cl.ly/rBsh">engineer party time</a>?</p><p>There should always be allowance to pull stories into the sprint, because work being done should be visible and tracked. If the team is focusing on <em>only </em>pulling in stories that can be completed in the sprint, that means the backlog’s order is irrelevant. You’re then picking stories based on the convenience of their size, and <em>then </em>considering their priority. This also means that you’ll have an unexpected influx of work landing at the beginning of the next sprint that no one knew about.</p><p>Personally, I take this opportunity to tackle <a href="https://gph.is/1pHkkC7">tech debt that the business doesn’t deem important</a>, or update some docs to make everyone’s lives easier in the future (<a href="https://cl.ly/rCgi">bad docs make sad engineers</a>).</p><h3>Let’s Wrap This Up</h3><p>Process is critical in just every organization, but blind acceptance of its rules can be dangerous. Kanban is flexible by nature, and focuses on a constant workflow, so that might be a better process fit. I will admit that the root of <em>many </em>sprint process related issues is how detailed the stories are. They need to be written with the mindset of “I’m giving this to <a href="https://cl.ly/rDSt">a new team member</a>”, but I digress: that’s a different blog post entirely.</p><p>Let’s face it, you could just show a fake burndown chart to whoever wants to see it, and <a href="https://cl.ly/rDHS">no one is going to be able to tell the difference</a>. 🤓</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2e252c16e4f" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>