<?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 Jonah Jones on Medium]]></title>
        <description><![CDATA[Stories by Jonah Jones on Medium]]></description>
        <link>https://medium.com/@jonah.jones?source=rss-651fb37ad4d0------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*OIyFh3ZSXwO3hNKS9oT7iA.jpeg</url>
            <title>Stories by Jonah Jones on Medium</title>
            <link>https://medium.com/@jonah.jones?source=rss-651fb37ad4d0------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 13:51:15 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@jonah.jones/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[VibeCoding a Godot Game — Part 5]]></title>
            <link>https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-5-5a56c7f8546c?source=rss-651fb37ad4d0------2</link>
            <guid isPermaLink="false">https://medium.com/p/5a56c7f8546c</guid>
            <category><![CDATA[game-development]]></category>
            <category><![CDATA[chatgpt]]></category>
            <category><![CDATA[gaming]]></category>
            <category><![CDATA[godot]]></category>
            <category><![CDATA[claude-code]]></category>
            <dc:creator><![CDATA[Jonah Jones]]></dc:creator>
            <pubDate>Sun, 03 May 2026 17:10:37 GMT</pubDate>
            <atom:updated>2026-05-03T17:10:37.972Z</atom:updated>
            <content:encoded><![CDATA[<h3>Tooling Changes</h3><p>Well Claude has been dethroned. This week I’ve made an immense amount of product by changing up my workflow.</p><p><strong>ChatGPT 5.5 Thinking mode</strong> is now our new king of writing code.</p><p>Claude is now my simple peasent to implement the changes, and code recommendations from <strong>ChatGPT </strong>and launch the game and fix compile errors from the MCP. I’ve found we’re producing much much higher quality code with far fewer revisions needed for adding features this way. In addition <strong>ChatGPT</strong> spent about 1–2 hours refactoring and finding performance issues in all our existing claude written files.</p><p>My current “pipeline”</p><ul><li>Discuss feature with ChatGPT</li><li>Show relevant files to ChatGPT</li><li>ChatGPT psuedo writes the code blocks</li><li>Hand to claude to implement</li><li>Claude uses MCP to debug compile errors</li></ul><p>I’m getting to a point where I’m being very particular with my added features as we start to have the real makings of a game too, so I’m not concerned about amount of code, but more about quality and structure of code.</p><h3>Game Updates</h3><p>We have many many game updates this time using our new workflow.</p><p>New management Hub Panels</p><p><strong>Party Panel</strong></p><ul><li>This one’s primary and most difficult portion by far was the spell slot. We’ve added level and money gating to spell unlocks, tooltips, and the ability to equip/unequip active loadout which is now used in combat by the player and party.</li><li>Easy portions included the EXP bar, and changing active player, and Active Party section</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*VyXZCqZiGrlAMSu6nvu4-g.png" /></figure><p><strong>Character Panel</strong></p><ul><li>The most difficult part of this was the Paper doll aka character grid, and getting a working equip/unequip system working. In addition getting colored borders to display item rarity, tooltips, and other features working correctly had many bugs that had to be squashed. We also have the idea of armor types (cloth/leather/mail), and characters are limited to those and weapon types (melee/staff).</li><li>Easiest part was the stat panel on the right side, and making it update in realtime as we change gear was relatively simple for the AI.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rAxpfD4bCVGjBO2t9T4VmQ.png" /></figure><p><strong>Revamped Combat UI</strong></p><p>We re-worked alot of the internals and added caching layers to improve performance. In addition we have</p><ul><li>portraits for the enemies</li><li>Re-did the Character action bar with keybinding, and borders.</li><li>Added Party spellbars, flashing effects, and radial cooldowns for more obvious UX of what your AI party is performing while you undergo combat.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FQkFy-ZfD6gR8gvIjetxww.png" /></figure><p><strong>Post Combat Screen</strong></p><ul><li>Added Quest items</li><li>Added EXP Bars</li><li>Added damage meter for party members, with tooltip effect to see ability breakdown from combat. This required modifying the spell effect JSON from each combat to store the overall post mitigation damage from each spell during the combat.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8vuaks5pUfEza66JiJ8ZtQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WcTTb8xamBhA2MIjLpx0zw.png" /></figure><p><strong>Overcapping / Bonus Talents</strong></p><p>Obviously if you’re one of the few people who read this far and you happened to have played World of Warcraft you might be thinking “This looks like a badly done 2D WoW Clone”. You’re right from the standpoint it’s easy finding UI example of different wow systems to provide to the AI when building.</p><p>However we obviously need some things that will differeniate ourselves from a badly done WoW clone. One system that I’ve played in other roguelites, and would be fun to add in WoW is the idea of overcapping talent points. In WoW most talent points have a cap like 3/3 or 5/5. This is fine, but sometimes for influential talent points you always wonder what would would happen if you could put 10 points or 15 points in a particular skill. Would my build scale exponentially? These are the fun parts of a simple game like this, the idea of “Breaking the game” or “Finding a build”.</p><p>This is one of the first, but not last unique systems we’re adding to this game to begin setting it apart and turning it into something fun.</p><p>In this screenshot you can see we’ve modified some items to provide bonus talent points, and on the talent tree you can see a 3/3 skill now becomes 5/3 with the item equipped.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/440/1*kkCamhtQU5d5i7F_o-GOEg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/996/1*18pj3q_-TQpEi6bnHR434g.png" /></figure><p><strong>Overall Thoughts</strong></p><p>This is my last blog before I go back from Paternity leave, so development progress might slow down a bit again. However we’ve made such good progress in the past 4 weeks we won’t abandon this.</p><p>We have a few systems we need to add/polish before we begin the long grind of balancing, and adding items/content/assets/textures. Most of these are things I can do without AI since I actually want to be careful on stuff like balancing, but we’ll be looking into ways to utilize AI to build out a test/QA bot once we add more content.</p><p>Part 1: <a href="https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-1-e93b7bec1bd7">https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-1-e93b7bec1bd7</a><br>Part 2: <a href="https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-2-c9950f99e4f1">https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-2-c9950f99e4f1</a><br>Part 3: <a href="https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-3-f92046427ada">https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-3-f92046427ada</a><br>Part 4: <a href="https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-4-53f4060e3e61">https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-4-53f4060e3e61</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5a56c7f8546c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[VibeCoding a Godot Game — Part 4]]></title>
            <link>https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-4-53f4060e3e61?source=rss-651fb37ad4d0------2</link>
            <guid isPermaLink="false">https://medium.com/p/53f4060e3e61</guid>
            <category><![CDATA[vibe-coding]]></category>
            <category><![CDATA[claude-code]]></category>
            <category><![CDATA[godot]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[game-development]]></category>
            <dc:creator><![CDATA[Jonah Jones]]></dc:creator>
            <pubDate>Sun, 19 Apr 2026 20:30:56 GMT</pubDate>
            <atom:updated>2026-04-19T20:30:56.797Z</atom:updated>
            <content:encoded><![CDATA[<p>Part 1: <a href="https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-1-e93b7bec1bd7">https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-1-e93b7bec1bd7</a><br> Part 2: <a href="https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-2-c9950f99e4f1">https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-2-c9950f99e4f1</a><br> Part 3: <a href="https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-3-f92046427ada">https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-3-f92046427ada</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KRDGf4ak6aaMVbMPkjVRcg.png" /></figure><h3>Tooling Changes</h3><p>We’ve hit a wall. There is only so fast I can move when building out UI scenes with Claude.<br> Every development cycle was starting to look the same:</p><ul><li>I tell Claude to implement or fix X feature</li><li>Claude says, “Yes, I got it… and it’s solved — try it now!”</li><li>90% of the time, it’s not solved the way I asked</li><li>I load the Godot scenes, take a screenshot, and tell Claude it’s wrong</li><li>Claude says, “Oh yes, my bad… it’s fixed — try now!”</li><li>Repeat steps 3–5</li></ul><p>This was getting tiresome, and I get that Claude isn’t able to see or access UI in games natively, so it’s difficult.</p><p>So we needed to level up our game — and our tooling.</p><p>First, we’re adding an MCP server:</p><p><a href="https://github.com/ee0pdt/Godot-MCP">GitHub - ee0pdt/Godot-MCP: An MCP for Godot that lets you create and edit games in the Godot game engine with tools like Claude</a></p><p>I’ve never been very high on MCP servers and never really had a great use case for them — until this one. It gives Claude the ability to:</p><ul><li>Launch scenes (game screens) itself</li><li>Analyze the output/logs from those scenes without me copy/pasting them</li></ul><p>These two features alone cut out 30–40% of my manual intervention in the development loop and have been a huge improvement.</p><h3>Game Updates</h3><h3>Talents!</h3><p>This week we added a talent tree to our classes. I know this looks very similar to <em>World of Warcraft</em>… and that’s intentional. I didn’t have time to create or find new assets, so we used WoW Classic talent assets we found online and built a native talent tree that works very similarly.</p><p>Here are the talent trees in the hub. I’ve added 2 talent specs for each of the 4 classes — so 8 talent trees total.</p><h3>Difficulty of the Talent System (for AI)</h3><p><strong>Tooltip logic — 9/10</strong><br> This was very difficult. Explaining UI behavior like placement, refreshing logic on click, and dynamically updating tooltips in real time based on point allocation took a lot of back-and-forth. This was especially tricky since MCP couldn’t capture mouseover behavior.</p><p><strong>Layout of the tree — 4/10</strong><br> This took some time to get the exact ordering right and handle a modular number of talents per row.</p><p><strong>Locking talents — 5/10</strong><br> This required careful explanation — especially differentiating points spent between the two specs per class and locking each subsequent row based on total points spent.</p><p><strong>Talent points per character — 7/10</strong><br> This was hard for the AI to grasp. We also had no concept of persistent character state previously, so we had to implement a JSON structure to save talent points per character between battles.</p><p><strong>Talents to real stats — 3/10</strong><br> I thought this would be harder, but it wasn’t too bad. After a few iterations, we got it working. We also had to add a lot of new spells to spell.json based on WoW talent abilities we didn’t have yet.</p><p>There was a lot of work involved in adding each class and tying talent points to leveling. Tooltips were also particularly challenging — getting them to resize automatically and update in real time as points are assigned was no trivial task and required extensive back-and-forth with Claude.</p><p>In addition, we implemented talent tiering/locking. This means players can’t access talents in the second row until they’ve spent 5 total points in the tree. Once unlocked, we indicate this with a light green border around the skill.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QEGXv-7EVYeIARA7icok7w.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2IFrtZiQ6_JYSr-ZKLZMWA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/592/1*EkQeWra0wI0bW-hZLDQKrg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/602/1*OzgbNkFM9o82KX2wdOi1xA.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=53f4060e3e61" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[VibeCoding a Godot Game Part 3]]></title>
            <link>https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-3-f92046427ada?source=rss-651fb37ad4d0------2</link>
            <guid isPermaLink="false">https://medium.com/p/f92046427ada</guid>
            <category><![CDATA[claude]]></category>
            <category><![CDATA[godot]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[game-development]]></category>
            <category><![CDATA[vibe-coding]]></category>
            <dc:creator><![CDATA[Jonah Jones]]></dc:creator>
            <pubDate>Sun, 12 Apr 2026 14:18:10 GMT</pubDate>
            <atom:updated>2026-04-12T14:18:10.760Z</atom:updated>
            <content:encoded><![CDATA[<p>Part 1 <a href="https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-1-e93b7bec1bd7">https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-1-e93b7bec1bd7</a></p><p>Part 2 <a href="https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-2-c9950f99e4f1">https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-2-c9950f99e4f1</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KRDGf4ak6aaMVbMPkjVRcg.png" /></figure><p>We’re back and working on the Vibe Coded WOW style healing game part 3. This month has been very hectic in my personal life so very slow on the updates here. I apologize but as always am excited to share all the things we’ve been working on this past month. For now it’s been mostly working on fixing bugs, adding menus and scenes, a post combat screen, and adding items.</p><p><strong>Fixing bugs:</strong></p><ul><li>We fixed some bugs around the boss text, and targeting</li><li>Fixed bugs on menus and centering</li><li>Fixed lots of targetting bugs (enemies vs friendly characters)</li><li>Fixed a bug where we cannot re-enter the same dungeon twice in a session</li></ul><p><strong>New Tools</strong></p><ul><li>Started using MultiClaude <a href="https://github.com/dlorenc/multiclaude">https://github.com/dlorenc/multiclaude</a></li></ul><h3>Features</h3><p><strong>Menus</strong></p><p>Adding the scenes here allows me to quickly skip through the menus when testing changes in things like combat, or post combat loot screens.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RzMBmSbSuBAQ_D12IAjHWA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*721O0XYVTE7uGOgdcf6zlA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*sHS67ri4k5fezsRoeLpg9A.png" /></figure><p><strong>Post Combat Screen and items</strong></p><p>We added some basic items in a JSON file with basic stats. We don’t have a system of equipment/inventory yet, but that will come soon. In addition we added experience + level ups here and bonuses for combat for things like no party members dying.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Uzoz5-DFmmUq0UJQGjj-HQ.png" /></figure><p>For items we’re choosing to go for a non randomly generated item route. Eventually I plan to make many unique and special items so it feels more valuable winning instead of grinding for RNG on loot stats.</p><p>We have a system where items have unique IDs, stats, and equipment slots.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/986/1*fdIvOf3YrVkggmrT313NoQ.png" /></figure><p>In addition the enemies now have a concept now of a loot table.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1004/1*KXu7B5YJvtE3rW9IC1CbwA.png" /></figure><p><strong>Combat Log</strong></p><p>We’ve added a minimizable Combat Log in the top right corner now. This includes green/red text to denote friendly healing and damage from enemies.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*L3eDJ_gYymzT1DqbRv-xcQ.png" /></figure><p><strong>Overall AI prompt Usage/Thoughts</strong></p><ul><li>~120–150 total prompts however, this is skewed by the repetitive status checks. If we exclude those: ~30–40 substantive requests</li><li>36 commits last 2 weeks ( About 1 per task)</li><li>Multiclaude keeps the AI on task more often, and takes less user intervention (awaiting response)</li></ul><p>Thanks for reading if you made it this far!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f92046427ada" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[VibeCoding a Godot Game Part 2]]></title>
            <link>https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-2-c9950f99e4f1?source=rss-651fb37ad4d0------2</link>
            <guid isPermaLink="false">https://medium.com/p/c9950f99e4f1</guid>
            <category><![CDATA[claude]]></category>
            <category><![CDATA[vibe-coding]]></category>
            <category><![CDATA[game-development]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[godot]]></category>
            <dc:creator><![CDATA[Jonah Jones]]></dc:creator>
            <pubDate>Fri, 30 Jan 2026 14:12:57 GMT</pubDate>
            <atom:updated>2026-01-30T14:12:57.588Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KRDGf4ak6aaMVbMPkjVRcg.png" /></figure><p>We’re back and working on the Vibe Codede WOW style healing game. This week we’ve been working on enhancing the combat experience, and adding more UI elements. We’re getting to the part soon where the combat is starting to look passable, which means we’ll have to move onto things like making it a real game…. and Art.</p><p>Art is one thing I’m not looking forward to. For now I ended up ripping a bunch of free sprites from the internet that offer free 2D texture packs. I know at some point if I ever really want to release this game I’ll need to redo the art part.</p><p>That’s a conversation for another day. Today lets focus more on the combat experience here.</p><h3>Tools Used So far Week 2</h3><ul><li>Claude Code</li><li><a href="https://github.com/steveyegge/beads">Beads</a> (Project management)</li><li><a href="https://godotengine.org/">Godot</a></li><li><a href="https://github.com/jegor377/godot-gdgifexporter">Godot gif exporter</a> (to show AI UI elements)</li><li><a href="https://github.com/Perfare/AssetStudio">AssetStudio</a> (For finding assets)</li><li><a href="https://itch.io/game-assets/free">https://itch.io/game-assets/free</a> and <a href="https://devassets.com/">https://devassets.com/</a> for Free assets.</li></ul><h3>Week 2 Systems Added</h3><ul><li>Cleaned up enemy health bar UI elements (<strong>9/10</strong>) Using WOW raid frames as a work back ideal here, it took alot of iteration to get the health bars looking, and feeling like something similar. UI changes with AI are difficult since it can’t view the screen, and this took lots of GIF recordings for it.</li><li>Added sprites (<strong>6/10</strong>) Finding good sprites online was difficult, and getting them all resized for the AI.</li><li>Added sprite movement logic (<strong>5/10</strong>) Hardest part here was adding some sort of movement effect, and adding AI logic about which characters want/try to move.</li><li>Added sprite targetting logic (<strong>5/10</strong>) Explaining the concept of melee vs ranged attacks here, and tying in melee attacks with movements so the AI will continue forward before they can attempt these attacks.</li><li>Added sprite death logic (<strong>5/10</strong>) Helping the game understand that the character needs to be removed, and no longer performing AI actions or targetable took a bit to explain.</li><li>Menu System &amp; Scene Architecture (<strong>4/10</strong>) The hardest part about this was understanding how to load different scences in GODOT for me.</li><li>Shield &amp; Effect System Fixes (<strong>4/10</strong>) Making the shield track the active missing health, and always appear right from left as the health changes.</li><li>HOT/DOT Refresh Mechanism (<strong>3/10</strong>) Implemented icon refresh checking to prevent duplicate icons when reapplied</li><li>Added colored Health Bars based on class (<strong>1/10</strong>) Copied WOW class coloring.</li></ul><p>Now what we’ve all been waiting for….. The GIFS</p><p>The first GIF is showing the menu, and the 2nd is show a bit longer Combat here with spell casts.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*A1HCRuO6V0si0JbiOqVQVA.gif" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UTbi5vRyLwpVvppg-w62fg.gif" /></figure><p>I look forward to showing some cool updates in Update 3, now that the combat is starting to come together. Until next time!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c9950f99e4f1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[VibeCoding a Godot Game Part 1]]></title>
            <link>https://medium.com/@jonah.jones/vibecoding-a-godot-game-part-1-e93b7bec1bd7?source=rss-651fb37ad4d0------2</link>
            <guid isPermaLink="false">https://medium.com/p/e93b7bec1bd7</guid>
            <category><![CDATA[claude]]></category>
            <category><![CDATA[vibe-coding]]></category>
            <category><![CDATA[game-development]]></category>
            <category><![CDATA[godot]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Jonah Jones]]></dc:creator>
            <pubDate>Tue, 20 Jan 2026 14:57:49 GMT</pubDate>
            <atom:updated>2026-01-20T16:22:29.168Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KRDGf4ak6aaMVbMPkjVRcg.png" /></figure><p>Yes this story is exactly is like it sounds. Can I a developer with 0 game dev experience utilize AI coding tools to vibecode a game?</p><p>Probably not, but this is a series where we’ll find out. I’m aiming to design a game that brings together some of my favorite video games pastimes.</p><ul><li>World of Warcraft</li><li>Healing in MMOs</li><li>Roguelites</li></ul><p>Together we will embark on the vibecoding experience, and I will keep track of the total amount of time spent, and every feature I add along the way. Ideally I’ll provide updates via Medium every 1–2 weeks on the progress.</p><h3>Tools Used So far Week 1</h3><ul><li>Claude Code</li><li><a href="https://github.com/steveyegge/beads">Beads</a> (Project management)</li><li><a href="https://godotengine.org/">Godot</a> (This isn’t using Golang)</li><li><a href="https://github.com/jegor377/godot-gdgifexporter">Godot gif exporter</a> (to show AI UI elements)</li></ul><h3>Rules I set for myself</h3><ul><li>Working a bit everyday in my freetime. NO writing any code myself</li><li>NO creating any assets or sprites myself. Only using free ones from the internet or AI tools to create new ones</li><li>Every time we hit a milestone on a system, I have the AI create a git tag. AI is known to blow away perfectly good code in pursuit of making another change. This way I have a state in time to go back to when the game is a good working state on each system.</li></ul><h3>Week 1 Systems Added</h3><p>Systems added so far, and difficulty the AI took to create each</p><ul><li>Combat Bars(<strong>9/10</strong>) AI wasn’t good at understanding spacing, and feedback on UI</li><li>Cooldowns on Combat Bars (<strong>8/10</strong>) The radial effect within a spell container was very confusing to AI, and often would break the UI sizing when adding.</li><li>Cast Bar (<strong>6/10</strong>) Took a while for it to understand casting spells, and not allowing other key clicks during casts.</li><li>Global Cooldown System on Instants (<strong>5/10</strong>) This was a tricky mechanic to explain to an AI, but helps balance instant spells</li><li>Stat System (<strong>4/10</strong>) Had to write explanations on all stats, and primary vs secondary stats ex Agility vs Critical chance</li><li>Healing and Damage over time (<strong>4/10) </strong>Also not too complex, but the AI was not understanding the stackable concept here. So I had to create an element called <strong>isstackable</strong>() and <strong>maxstacks</strong>() on each of these. Then the concept of refreshing these when one already exsits on a target.</li><li>Death (<strong>4/10</strong>) It took the AI a bit to understand dying means the AI would stop casting spells, and not allowing the main player to cast spells on dead characters</li><li>Combat Magic vs Physical System (<strong>3/10</strong>) Had to explain magical spells vs physical and assign stats to each.</li><li>Spells (<strong>3/10) </strong>These were harder because I added them in before all the other systems like combat vs physical, HOT, DOT, instant, casts. So I had to go back and re-define alot of things.</li><li>Combat system (<strong>3/10</strong>) (AI for enemies and hero pary not allowing targeting themselves or teammates)</li><li>Health and Mana system (<strong>3/10</strong>) Only difficulty here was adding new class types with different resources (energy vs rage vs mana)</li><li>Shields (<strong>2/10) </strong>The hardest part about these was the UI overlay on health bars</li><li>Class System (<strong>2/10</strong>) Simple lots of input test I could pull from the internet.</li><li>Level System (<strong>1/10</strong>)</li></ul><p>So what does that leave us with?</p><p>I know this gif is a bit choppy, I will figure out how to record this with OBS next time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NIAwoB7G88djyBPjQgpjbw.gif" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e93b7bec1bd7" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AWS AppMesh for Intelligent ECS Deployments]]></title>
            <link>https://medium.com/schkn/aws-appmesh-for-intelligent-ecs-deployments-51cea1bd5542?source=rss-651fb37ad4d0------2</link>
            <guid isPermaLink="false">https://medium.com/p/51cea1bd5542</guid>
            <category><![CDATA[docker]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[application-development]]></category>
            <category><![CDATA[containers]]></category>
            <category><![CDATA[deployment]]></category>
            <dc:creator><![CDATA[Jonah Jones]]></dc:creator>
            <pubDate>Wed, 14 Aug 2019 00:47:11 GMT</pubDate>
            <atom:updated>2019-08-15T14:11:00.817Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*CPHhGy86yIZ4YEO4nlSHSg.png" /></figure><h3>Introduction</h3><ol><li>Introduction</li><li>Deployment Types</li><li>AppMesh Basics</li><li>Launching the Demo</li><li>Verifying Deployment Works</li><li>Verifying AppMesh Works</li><li>Wrapping Up</li></ol><p>If you made it through my last story about <a href="https://medium.com/@jonah.jones/better-code-aws-boto3-using-factory-method-1f410d75879d">python factory method</a>, then awesome! We will be tying that back in during the 3rd post, but for now we need to get our App up and running.</p><p>This article will be about getting our Application working, and deployable with AppMesh Service Discovery, and Tracing. If you don’t know what AppMesh is already let’s take a look at the AWS definition.</p><blockquote>AWS App Mesh is a service mesh that provides application-level networking to make it easy for your services to communicate with each other across multiple types of compute infrastructure. App Mesh standardizes how your services communicate, giving you end-to-end visibility and ensuring high-availability for your applications.</blockquote><h3><strong>Deployment Types</strong></h3><p><strong>What are the different types of Deployments?</strong></p><blockquote><strong>Canary Deployment</strong> is a deployment type where you deploy the new version of your application onto a % of your cluster, and test it out before declaring it successful, and moving the new version to all your servers.</blockquote><blockquote>The major drawback here is the reduced capacity of total compute power from updating/testing and potentially rolling back active servers.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JLJO5c13g2Pwjvt9iX_RxQ.png" /></figure><blockquote><strong>Blue/Green Deployment</strong> is a deployment type in which you spin up an identical copy of you application, but running the new version of code. You then cutover your load balancer to the new version, resulting in no reduced server capacity or downtime.</blockquote><blockquote>The major drawback here is that we can test functionally test our “green” environment, but not with production level traffic.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*osi194r4bSAeZ8vWVdCI2w.png" /></figure><blockquote><strong>Canary Blue/Green Deployment</strong> is a deployment type that starts like a blue/green deployment, but once our green services are online we can test utilizing canary traffic shifting to test our new application with a small % of real traffic.</blockquote><blockquote>The major drawback is that this is not out of the box supported by AWS ECS and Codedeploy at the moment. Which is why we are going to achieve this with AppMesh</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*HixETQsBw9y-pE7Jqsw2SQ.png" /></figure><p>Our final article will be combining two deployment methods. This article’s demo today will end you with a working Blue/Green Deployment.</p><h3><strong>AppMesh Basics</strong></h3><p>There are a few key components of AppMesh to cover before we go through the example. I am going to summarize each portion in my own simple terms, if you really want a more in depth understanding check out the AWS docs on <a href="https://docs.aws.amazon.com/app-mesh/latest/userguide/what-is-app-mesh.html">AppMesh</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/718/1*S9IL9QQxJg-kzVnM4g5Xhw.png" /></figure><blockquote><strong>Mesh</strong></blockquote><p>This is logical boundary for all of our mesh components</p><blockquote><strong>Virtual Service</strong></blockquote><p>These are the services within the mesh that will be where we route requests to. In other works our Backend.</p><blockquote><strong>Virtual Router</strong></blockquote><p>Which port protocol to listen to traffic on, and where to route those requests we receive. (Very similar to an ALB)</p><blockquote><strong>Virtual Routes</strong></blockquote><p>Once the Router receives a request, routes define which nodes to send traffic to, and at what weighted percentage. (Very similar to an ALB Listener Rule)</p><blockquote><strong>Virtual Nodes</strong></blockquote><p>Envoy image configured with AppMesh settings in the task definition providing a proxy for our tasks.</p><p>A typical request looks like such</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RLwCox-YzdAvcRPg_FzHPA.png" /></figure><h3>Launching the Demo</h3><p>To launch our infrastructure we are going to use our git repo</p><p><a href="https://github.com/jonahjones777/canary-mesh">github.com/jonahjones777/canary-mesh</a></p><p>Check the README.md for the list of required packages we need, and the corresponding commands to install them on OSX with Brew.</p><p>Once we are ready we can launch with</p><pre>make launch</pre><p>This will do 3 things</p><ol><li><strong>launch our AWS stack using terraform. This includes</strong></li></ol><ul><li>VPC and required pieces for public/private subnets</li><li>ECS Service, ECR Repo, and ECS Cluster</li><li>ALB, and On-Demand EC2 Cluster configured to join ECS Cluster</li><li>Codepipeline, CodeBuild, Codedeploy to Handle BlueGreen Deployments.</li><li>Service discovery namespace, and AWS Cloudmap app</li><li>AWS AppMesh Mesh, Service, Router and Routes tied to Cloudmap</li><li>Xray Tracing for AppMesh</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/879/1*gNBPwg2a9tUPttj8XcX5uA.png" /></figure><p>2<strong>. Create Buildspec and Task Definition for BlueGreen Deploys</strong></p><ul><li>Uses Jinja to Template buildspec.yml and task_def.json with created values</li><li>Uploads created files to CodePipeline artifact bucket</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4-LvN4FGWE1cwsx8TOugMg.png" /></figure><p>3<strong>. Build and Push Image</strong></p><ul><li>Build our demo app with Xray tracing written in app</li><li>Uploads our image to ECR repo, which triggers pipeline!</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eaM2jMEAqnul19A6d8un6Q.png" /></figure><p>One this is finished we should be able go see our new service by clicking on the URL from the tf.env, or from the Load Balancer console DNS name.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*n7pfHMJsznE9UDPgaGFUdg.png" /></figure><p>Awesome it looks like our app works, let’s check the <strong><em>URL/service-status</em></strong> path too.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3HgQJ5esd4Ug_DB236N6Vw.png" /></figure><h3><strong>Verifying Deployment Works</strong></h3><p>We should be able to push new software deployments now by executing a make command.</p><p>Let’s give it a try by incrementing the version number in app/app/version.txt to 1.1, saving and running</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OqBfYKSjat1wGQq1dUQ2_A.png" /></figure><pre>make build</pre><p>Next we need to go our Codepipeline, and find the CodeDeploy Blue/Green step which is currently the 3rd step. We need to guide it through the process of approving our new app version.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rJdNsKG0BiVFYUwMtQNI5g.png" /></figure><p>Once we are here, and the new green tasks come online and pass their Application Load Balancer Health Checks, we will be allowed to click the button to route traffic to our new version.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*iuZhZrUfXSQvalnAnoS2jA.png" /></figure><p>To confirm our CICD pipeline is working we go back to our service-status endpoint and we should be able to see the version got incremented by hitting the <strong><em>URL/service-status</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3-SnifIXcASflsO8W3aAfQ.png" /></figure><h3><strong>Verify AppMesh Works</strong></h3><p>Now we know that our app works, and our deployment pipeline is functioning as intended we are going to verify that AppMesh, and our service discovery are working correctly. There are a few things we need to check here.</p><ol><li><strong>Service Discovery</strong></li></ol><p>We have configured our AppMesh Virtual nodes to use AWS CloudMap to automatically publish and register themselves via CloudMap for other apps to route to.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*roGcIwWNjJs02nCWB7MI7g.png" /></figure><p>We can confirm this by going to AWS CloudMap and checking the <strong><em>service </em></strong>namespace, and confirming we have healthy registered nodes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aUoNAs7aS9GUUNkHx34FJg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-MuD066uG46C1GscxtYiuQ.png" /></figure><p>We see it registered, but let’s confirm that it’s actually working correctly by hitting our service discovery URL. We write a quick curl loop hitting <strong><em>test1.service/service-status</em></strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TIfV_3UNKag_8r2mAX-GDA.png" /></figure><p>We have just confirmed we are able to route to our service discovery namespace. We are doing internal routing with these requests via our service mesh, which has some great benefits such as:</p><ul><li>Reduces cost since inter app traffic isn’t going through our LoadBalancers</li><li>Reduces latency since inter app traffic isn’t going through our LoadBalancers</li></ul><p><strong>2. Xray Tracing</strong></p><p>Let’s go to the Xray service page, and check to make sure our internal tracing, and AppMesh tracing are working. It should look something like this.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GYt4lkgJCJc0Ys5sjVUIcg.png" /></figure><p>For Internal App tracing a typical use case would be to turn tracing on when we have an application function we suspect is having issues. We then add tracing at the start of that function, and send a <strong><em>user.id</em></strong> or <strong><em>customer.id</em></strong> to Xray via the trace <a href="https://docs.aws.amazon.com/xray/latest/devguide/xray-sdk-python-segment.html">metadata function</a>. We can then observe and correlate failures to particular users we detect might be having issues, and look for patterns in these issues.</p><p>AppMesh tracing can also be really useful once when we have multiple Microservices within the same AppMesh talking to each other. It can help solve and detect issues such dropped and throttled requests, api latencies, and TLS errors.</p><h3>Wrapping Up</h3><p>To Cleanup the application run this command from the project root.</p><pre>make destroy</pre><p>Thanks for reading, and stay tuned for last article in our series where add the canary portion to our Blue/Green Deployment using the Factory Method discussed in the first <a href="https://medium.com/@jonah.jones/better-code-aws-boto3-using-factory-method-1f410d75879d">article</a>.</p><p><em>“Opinions expressed are solely my own and do not express the views or opinions of Amazon”</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=51cea1bd5542" width="1" height="1" alt=""><hr><p><a href="https://medium.com/schkn/aws-appmesh-for-intelligent-ecs-deployments-51cea1bd5542">AWS AppMesh for Intelligent ECS Deployments</a> was originally published in <a href="https://medium.com/schkn">devconnected — DevOps, Sysadmins &amp; Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Better Code AWS Boto3 using Factory Method]]></title>
            <link>https://medium.com/@jonah.jones/better-code-aws-boto3-using-factory-method-1f410d75879d?source=rss-651fb37ad4d0------2</link>
            <guid isPermaLink="false">https://medium.com/p/1f410d75879d</guid>
            <category><![CDATA[boto3]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[python]]></category>
            <dc:creator><![CDATA[Jonah Jones]]></dc:creator>
            <pubDate>Mon, 05 Aug 2019 13:47:02 GMT</pubDate>
            <atom:updated>2019-08-05T14:05:37.487Z</atom:updated>
            <content:encoded><![CDATA[<p><em>“Opinions expressed are solely my own and do not express the views or opinions of Amazon”</em></p><h3>What to expect from this article:</h3><ol><li>What is the Factory Method?</li><li>How to implement Factory Method</li><li>Performance Before and After</li><li>Logging Before and After</li><li>Final Thoughts</li></ol><ul><li><strong>*This is going to be a tutorial on how to make a Blue/Green Canary Deployment using AWS ECS + AppMesh + StepFunctions. If you ‘re looking for the AWS portions, and less about the software development skip this article and go to part 2.</strong></li></ul><p>We will cover more in depth what’s a Blue/Green Deployment, and a Canary Deployment and why we want to combine these in a later article, Here’s a quick example of what we’re aiming for.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/245/1*klkK0HDmWmtEHftNgQqnjw.png" /></figure><h3><strong>What is the Factory Method?</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/275/1*KziSpktwjvVuVakQcisRfw.jpeg" /></figure><p>Design patterns became a popular topic in late 90s after the published book <a href="https://www.amazon.com/dp/0201633612/?tag=devdetailpage02-20">Design Patterns: Elements of Reusable Object-Oriented Software.</a> This book goes over several design patterns as solutions to reoccurring problems in software. The book describes the Factory Method as a creational design pattern. <a href="https://en.wikipedia.org/wiki/Creational_pattern">Creational design patterns</a> are related to the creation of objects, and Factory Method is a design pattern that creates objects with a common interface. In modern programming this is a method that is <strong>one of the most widely used design patterns</strong>, and I’m going to show you how to apply that to python and <a href="https://boto3.amazonaws.com/v1/documentation/api/latest/index.html?id=docs_gateway">Boto3</a>.</p><h3><strong>How to Implement the Factory Method</strong></h3><p>This is an example interface to show how you using the factory method we can guarantee logging on all function calls using this Boto3 AppMesh library.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/26d9ae5eae25a078a149b515c355b178/href">https://medium.com/media/26d9ae5eae25a078a149b515c355b178/href</a></iframe><p>Utilizing concepts of naming, and reduced inputs from <a href="https://www.amazon.com/Clean-Code-Handbook-Software-Craftsmanship/dp/0132350882">Clean Code</a></p><ol><li>Classes are Nouns (<strong>AppMesh</strong>)</li><li>Functions are Verbs (<strong>create_mesh</strong>)</li><li>Reduce Function inputs to absolute minimum that makes sense.</li></ol><p>We create our simple factory for AppMesh. The<strong><em> check_mesh_exists()</em></strong><em> </em>and the <strong><em>create_mesh()</em></strong> are both concrete implementation methods of the product. Our <strong>product (AppMesh Object) </strong>is the outcome of using “buisness logic” to actually piece to initiate the <strong>client (AppMesh Interface)</strong>. To do this we use our business logic which would be referred to as the <strong><em>creator</em></strong> component. For simplicity stakes here, Let me post a translation of this technical jargon.</p><p>Product = Library | <strong>(mesh.py)</strong></p><p>Client = Object | <strong>(mesh = AppMesh(logger, mesh=”canary”)</strong></p><p>Creater = Deployment | <strong>canary_deployment.py</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/1*kHk_3OuUNcQ9dJdPaso9kg.jpeg" /><figcaption>Saying Software Words I only half understand</figcaption></figure><h3>Performance Before and After</h3><p>Before we implore this method we are missing a few important things</p><ul><li>Logging with TimeStamps</li><li>Try/Catch Blocks</li><li>Function Calls for Easier Debugging</li></ul><p><strong>BUT</strong> since we forego all of these it should be way faster right?</p><p>Wrong it’s slightly faster, but the 5% speed increase we are getting here doesn’t make up for the fact we lose all those other important items when developing and troubleshooting.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Yt4OVjvVxXF7RwuLegpnew.png" /></figure><h3><strong>Logging Before and After</strong></h3><ul><li>Due to the Default Boto3 version used in Lambda being 9 months old, we need to upload a newer version into our ZIP packages to use the AppMesh API. Adding this makes our ZIP’s over the 10mb local editing limit in the lambda console.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3au6DSA9fxnHhJ3jdm_aVA.png" /></figure><ul><li>We will be using <a href="https://docs.aws.amazon.com/serverless-application-model/latest/developerguide/what-is-sam.html">SAM</a> to upload/change code rapidly in Lambda/Stepfunctions</li></ul><p>We will rely on local development initially, and trust SAM to package and upload correctly. This is a comparison of the logging I get locally and in Cloudwatch using a Class with a logging interface, and without.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pUlLlEN79QXTjQu9c7Pulw.png" /></figure><p>The 2nd option is more readable, and easier to write follow up logic with this way since I can see what’s a list vs dict. This will also print into Cloudwatch in a collapsable json format, vs printing a blob format.</p><h3>Final Thoughts</h3><p>There is no need to reinvent_wheel() Since the factory method is a well defined, and documented standard in modern programming. It’s less code to write, and comes with the guardrails of logging/error handling/tracing built into every object as we write our canary deployment.</p><p>Stay Tuned for Part 2 of this series where we Go into detail setting up our AppMesh!!!</p><p>Be Sure to keep updated on the Github too!</p><p><a href="https://github.com/jonahjones777/canary-mesh">jonahjones777/canary-mesh</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1f410d75879d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[First Steps Python to Go]]></title>
            <link>https://medium.com/schkn/first-steps-python-to-go-9e772185ae95?source=rss-651fb37ad4d0------2</link>
            <guid isPermaLink="false">https://medium.com/p/9e772185ae95</guid>
            <category><![CDATA[python]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[golang]]></category>
            <category><![CDATA[software-development]]></category>
            <dc:creator><![CDATA[Jonah Jones]]></dc:creator>
            <pubDate>Fri, 07 Jun 2019 13:29:31 GMT</pubDate>
            <atom:updated>2019-06-08T16:26:48.265Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/325/1*B_vb2rqCFylalusY_zX7ZQ.jpeg" /></figure><p>Python is amazing. Python is many people’s first programming language since it’s easy to learn, and extremely effective. According to <a href="https://insights.stackoverflow.com/survey/2019">stackoverflow’s 2019 survey</a>, Python is still the #1 most wanted, and used language.</p><h3>Top 5 most wanted languages:</h3><p><strong>Python:</strong> 25.7%</p><p><strong>JavaScript:</strong> 17.8%</p><p><strong>Go:</strong> 15.0%</p><p><strong>TypeScript:</strong> 14.6%</p><p><strong>Kotlin:</strong> 11.1%</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/291/1*I6dQD3M2aG16ncbObYa6BA.jpeg" /></figure><p>For many people in the DevOps space, Python is the language we hack all the operations together with. Python makes up the majority of Cron jobs, deployment scripts, and data migration pieces we write every day.</p><p>It’s really easy to sit back, and say that there is no need to learn another language. To say that you’re already too far behind others. You don’t have time to pick up another language. Don’t let the progress of others get in your way. Start small, and soon you will be able to follow along reading and then writing.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1015/1*S4jbHR7FEYFJHxksIeiW2w.png" /></figure><p>Go is simply amazing, and learning Go and will have an obvious boon to your effectiveness, and your career in this space. DevOps is more than Ops, and I’ve found that learning the Dev part has been the most humbling, and eye opening. You really figure out what processes work, and which ones don’t as you start developing, and iterating through something.</p><p>The further I get into my career the more I’m seeing amazing DevOps tools and projects written in Go. It’s not a coincidence that many of the new technologies on the block in the DevOps space are written in Go such as:</p><ul><li>Kubernetes</li><li>Terraform</li><li>Docker</li><li>Even Medium</li></ul><p>With the growth and maturity of Go, and the strong open source user base it became the obvious decision for myself and many other companies <a href="https://github.com/golang/go/wiki/SuccessStories">move to Go</a>. However as I started learning this I had a habit to pick it up for an hour here of there. Then forget most things next time I got back into it. I needed a way to relate it to python something I’m familiar with so I made this git table.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/1f3083f2b90176dda20462db40b5088b/href">https://medium.com/media/1f3083f2b90176dda20462db40b5088b/href</a></iframe><p>Now if you actually took the time to read through each example then Bravo to you.</p><p>Straight syntax examples won’t help most people, but if you’re like me it’s easier to start thinking about data types by comparing with things you’re familiar with. How do we actually learn Go though?</p><h3>Next Step Learning</h3><p>Aside from obvious learning techniques, like daily practice I will be covering a few learning resources in this part.</p><p>“<a href="https://golang.org/doc/code.html">How to Write Go Code</a>” has a section that explains exactly how to build your <a href="https://golang.org/doc/code.html#Workspaces">workspace</a>. You also can checkout popular git pages which help depending on your OS. I used this one personally <a href="https://github.com/abourget/getting-started-with-golang">https://github.com/abourget/getting-started-with-golang</a></p><p>This was an amazing book to learn Go, with well explained examples, and challenges through the book “<a href="http://www.golangbootcamp.com/">Golang Bootcamp</a>”. I also found this book to be much longer than most of the others I found.</p><p><a href="https://golang-for-python-programmers.readthedocs.io/">Go for Python Programmers</a> Great book tailored to someone familiar with python, and away to segway into more complex projects.</p><p>Francesc also has few great workshops — <a href="https://github.com/campoy/go-tooling-workshop">Go Tooling in Action</a> and <a href="https://github.com/campoy/go-web-workshop">Web Applications Workshop</a>.</p><p>The <a href="https://github.com/Alikhll/golang-developer-roadmap">Go Developer Roadmap</a> provides alot of great examples of where to go after finishing tutorial books, including different packages, etc.</p><p>The<a href="https://docs.aws.amazon.com/sdk-for-go/api/"> AWS Go SDK</a> Is a must if you are planning on using Go for a project on AWS (Terraform).</p><h3>Summary</h3><p>The goal of this article wasn’t to be overly technical in nature, but provide a reasoning about why you might consider learning go, and where to start especially if you have a grasp on python. Stay tuned for more article where I talk about Go, and the project I’m working on.</p><p>Go is gaining speed every day, and I encourage you all to check it out and become Gophers!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9e772185ae95" width="1" height="1" alt=""><hr><p><a href="https://medium.com/schkn/first-steps-python-to-go-9e772185ae95">First Steps Python to Go</a> was originally published in <a href="https://medium.com/schkn">devconnected — DevOps, Sysadmins &amp; Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[6 Git tricks that changed my life]]></title>
            <link>https://medium.com/@jonah.jones/6-git-tricks-that-changed-my-life-f63afe95ffa8?source=rss-651fb37ad4d0------2</link>
            <guid isPermaLink="false">https://medium.com/p/f63afe95ffa8</guid>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[github]]></category>
            <category><![CDATA[git]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[cli]]></category>
            <dc:creator><![CDATA[Jonah Jones]]></dc:creator>
            <pubDate>Fri, 05 Apr 2019 18:00:31 GMT</pubDate>
            <atom:updated>2019-04-06T11:54:32.424Z</atom:updated>
            <content:encoded><![CDATA[<p>Like most people who came from an ops background when I started in Devops, I felt more comfortable starting with a Git Client. I officially passed my 6th month of cutting cord, and going 100% Git CLI. In honor of my 6 month mark I thought I would share 6 tricks I’ve learned along the way!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/510/1*JpeXXlqe6t_T7lubc46g1g.jpeg" /></figure><h3>Tip #1 Use a .git Config File</h3><p>With a good git config file you really see a ton of info when working on the command line. This little trick changed my game dramatically, and is my #1 tip for a reason. I’ll post an example of one which I use to get path / branch / uncommitted changes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*64R4qzAHUojo-gtjdmlBEA.png" /></figure><p>In this example we are adding to our ~/.bash_profile</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/0421f1a15ca3c764687b3a3d485deb0e/href">https://medium.com/media/0421f1a15ca3c764687b3a3d485deb0e/href</a></iframe><h3><strong>Tip # 2 How to pull down remote branch</strong></h3><p>I know this one seems like it should be obvious. There is so many times when you need to pull down someone else’s branch to see what’s going on, or make changes, and I struggled initially since I found about 10 different ways on stackoverflow with some being outdated, and some being overengineered. I found this method to be my favorite after testing.</p><p><strong>Step 1.</strong> If you have changes on current workspace run, commit or stash them.</p><pre>git commit -a -m &quot;adding my commmits for now&quot;</pre><pre>or</pre><pre>git stash</pre><p><strong>Step 2</strong>. Pull it down from Remote, and set to track</p><pre>git fetch origin</pre><pre>git branch -a              (Find name of branch you need)</pre><pre>git checkout --track origin/&lt;BRANCH_NAME&gt;</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hPPVligp7G8q1_uUO0ePFA.png" /></figure><p>I know this seems so easy, but when you are first starting using the CLI this one is a lifesaver.</p><h3>Tip # 3 View Github.com like an IDE</h3><p>Seriously navigating around Github.com is pretty painful. Using a 3rd party extension called <a href="https://chrome.google.com/webstore/detail/octotree/bkhaagjahfmjljalopjnoealnfndnagc/related?hl=en-US">octotree</a> makes it similar to an IDE. Obviously navigating around faster is helpful, not much else needs to be said on this.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*GwuuV6eQoi3hliflBGWxLw.gif" /></figure><h3>Tip # 4 Create Hyperlinks for file states to prevent changes/deletions.</h3><p>In GitHub if you press “y” while on a file, it will create a hyperlink so others can view the file as is currently is forever. This is useful to show a file state, or a file that could be deleted potentially</p><p>Try to check the link for a secret feature I pushed, but then removed, vs the generated hyperlink from pressing “y”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QU0c7hR2kpFaBkDn5d6-MA.png" /></figure><p><a href="https://github.com/jonahjones777/boto3-scripts/blob/master/ecs_scaling.py"><em>https://github.com/jonahjones777/boto3-scripts/blob/master/feature.py</em></a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fwO8i5lSrMs-ZoZ8UaxAdA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4jbunq8gyyQpivDOSeCqZg.png" /></figure><p><a href="https://github.com/jonahjones777/boto3-scripts/blob/256267e9ea2d0d700d082c253f1e69e29b13d505/feature.py"><em>https://github.com/jonahjones777/boto3-scripts/blob/256267e9ea2d0d700d082c253f1e69e29b13d505/feature.py</em></a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0jcjNgfMOVymIxaLsf3mtA.png" /></figure><p>We get a 404 on the regular link, but we can view the hyperlink, neat right?</p><h3>Tip # 5 Search for files by pressing “t” in a Repo</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*Q-XucQtqLzmsRzXqR3_2aw.gif" /></figure><p>This one explains itself, but very useful.</p><h3>Tip #6 Create a Global .gitignore for yourself</h3><p>No more accidentally pushing up keys, or OS generated junk like “.DS_Store”</p><p>If you create yourself a global .gitignore you can stop all these in their tracks, and make projects have more specific .gitignore files for the type of project.</p><pre>touch ~/.gitignore</pre><pre>git config --global core.excludesfile ~/.gitignore</pre><pre>vi ~/.gitignore</pre><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/3dfeaa7ad542116aeba9d178fa24dadb/href">https://medium.com/media/3dfeaa7ad542116aeba9d178fa24dadb/href</a></iframe><p>Security people around the world, can breath a sigh of relief after this.</p><h3>Wrap Up</h3><p>That’s all we had for now, these were the 6 most helpful tips, and tricks I’ve learned from the last 6 months of using git. Like, or Subscribe if you enjoyed the article, and until next time happy gitting!</p><p>Enjoy a sample of my 2018 Git Usage as well!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/890/1*lUylnHTXFLenK_5sZIgsxQ.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f63afe95ffa8" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Things to love about AWS Fargate Part 1 … Deployments!]]></title>
            <link>https://medium.com/@jonah.jones/things-to-love-about-aws-fargate-part-1-deployments-c2a8a8349057?source=rss-651fb37ad4d0------2</link>
            <guid isPermaLink="false">https://medium.com/p/c2a8a8349057</guid>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[development]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[deployment]]></category>
            <category><![CDATA[api]]></category>
            <dc:creator><![CDATA[Jonah Jones]]></dc:creator>
            <pubDate>Sat, 30 Mar 2019 14:38:59 GMT</pubDate>
            <atom:updated>2019-03-30T14:38:59.806Z</atom:updated>
            <content:encoded><![CDATA[<h3>Things to love about AWS Fargate Part 1 … Deployments!</h3><p>AWS Fargate you said?</p><p>Yes I mean the other option whenever I go to deploy to my ECS Cluster</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/967/1*WZ7lXNkQCqkw8wV8QANbNA.jpeg" /></figure><p>Yes AWS Fargate has become my favorite service on AWS. Forgetting about the weird and confusing name for a few moments, I want to shed light on this service.</p><p>I’ve just spent the past 6 months getting intimately familiar with AWS Fargate, and every doc/whitepage/tech talk for it. We’ve been using it for workloads of every size in every environment. I wanted to shed some light on some of amazing qualities of this service which I now believe is the best container computing service on AWS.</p><p>So what is AWS Fargate?</p><h3>Run containers without managing servers or clusters</h3><p>Seems simple enough right, we pick a docker image, and some variables and it gets run. That’s <strong>ALL</strong> we have to worry about. All the hard stuff gets handled by AWS</p><ul><li>Patching</li><li>Logging</li><li>Scaling</li><li>Security</li><li>Compliance</li></ul><p>This allows us to spend more time developing, and less time fiddling with overhead.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aaDM2D0uYl4xlTINXgIjzw.jpeg" /></figure><p>But…But…But it’s so much more expensive than running my AWS spot fleet on regular EC2 ECS cluster, and ……..</p><p>I hear this statement all the time too. While the compute price alone is more expensive, the amount of time you save NOT dealing with problems, and administration easily makes up the difference.</p><p>In fact after the most recent price drops:</p><p><a href="https://aws.amazon.com/blogs/compute/aws-fargate-price-reduction-up-to-50/">AWS Fargate Price Reduction - Up to 50% | Amazon Web Services</a></p><p>It is now in fact just as cost efficient as regular on demand EC2 instances!</p><p>Now I know I won’t convince you with words alone on how amazing Fargate is, so I’m going to over some of the amazing things that I love about Fargate, and the first one is deployments.</p><h3>Deployments</h3><p>Deployments on regular ECS were like</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*epm_hOIDTwEouF-gu8aRLA.jpeg" /></figure><p>We had to worry about</p><ol><li>Binpacking the tasks</li><li>The dreaded task scaling =/= instance scaling problem</li><li>Logging</li><li>How to employ 0 downtime deploys</li><li>Security, PCI/DSS… etc</li></ol><p>Then Fargate came along, and we don’t have to worry about any of that anymore.</p><ol><li>Binpacking is thing of the past, you ONLY pay for what is deployed/used.</li></ol><p>2. Now you only worry about task scaling. No more Ec2 management.</p><p>3. Logging via ECS-agent to Cloudwatch is now built in!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/884/1*7myb9CxUp10wFZnkK3j_oQ.jpeg" /></figure><p>4. CodeDeploy and tools like <a href="https://github.com/silinternational/ecs-deploy">https://github.com/silinternational/ecs-deploy</a> make this super easy for Blue/Green deploys now.</p><p>5. Fargate is secure out of the box, now you just need to worry about your container security, we have no idea what linux flavor the host is running, and I don’t care!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*B4_Qgsm5Z0Z_jybkPxce8g.jpeg" /><figcaption>I don’t know what half of these mean</figcaption></figure><p>This particular article is about deployments though, so I’m going to show you how to use CodeDeploy with Fargate to implement Blue/Green Application deployments.</p><p>If you want to a quick setup feel free to checkout some terraform I put up to do this originally:</p><p><a href="https://github.com/jonahjones777/terraform/tree/master/fargate_bluegreen">jonahjones777/terraform</a></p><ol><li>Create Code Deploy App, then Deployment Group</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*z_ZKOKyeCPykQK9T8BU5Gw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*24WHfLsfabTYr7nkGRdWKQ.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Et56WoI0EMkZFn6NarWY4w.jpeg" /></figure><p>2. In deployment group you need to enter</p><ul><li>Iam Role</li><li>ECS Cluster + Service Name</li><li>2 Target Group that point to your ECS Service</li><li>Traffic Port (443) or whatever you use</li><li>Reroute time I choose 5 minutes usually</li></ul><p>3. Once it’s all configured we can create deployments using 2 JSON files.</p><ul><li>appspec.json</li></ul><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/a177dd8fe51f9093fcc860464ae38b2f/href">https://medium.com/media/a177dd8fe51f9093fcc860464ae38b2f/href</a></iframe><ul><li>deployment.json</li></ul><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/b95498e0edaf613195c714ee43cb1980/href">https://medium.com/media/b95498e0edaf613195c714ee43cb1980/href</a></iframe><p>4. Now whenever you push a new docker image you can kick it off using a CLI command from the directory in which you appspec file lives in your pipeline or manually using 2 simple commands</p><p>Then we generate back our URL to go watch it quickly.</p><pre>aws s3 cp deployment.json s3://code-bucket/deployment.json</pre><pre>deployid=$(aws deploy create-deployment --cli-input-json file://appspec.json | jq -r &#39;.deploymentId&#39;)</pre><pre>echo &quot;https://console.aws.amazon.com/codesuite/codedeploy/deployments/$deployid?region=us-east-1&quot;</pre><p>5. Go Watch the Magic happen!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*SIl5_b3fYbiurW9BAlwdDA.jpeg" /></figure><p>This is what is starts to look like right when it starts. We can see that it’s beginning to bring up a copy of the original task here, but 100% of the task traffic is going to the original one. Once the 2nd task comes up it will need to pass the ALB health check path before it switches traffic.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RlroZ58Mg5T1vIBLAWyScA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1GvfEYNOu9Ki1y9Dng0iRw.jpeg" /></figure><p>Here is what happens after immediately after we send the command to start the deployment. After about 5 minutes if our new task isn’t broken it should look like this with 100% of new traffic shifted, and old resources now torn down.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EMpXCLe2tEFM6HDe_SUTMA.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*J2Hilc0M0QDhngOdZGKoqQ.jpeg" /></figure><p>Looks like it took 3 minutes for the new version to come online, pretty slick huh? If you’ve ever had to deal with regular ECS deployments, the simplicity, and ease of 0 downtime deployments will make it easy to sleep at night.</p><p>Now if you really have a true sense of danger you can also do what’s called a force deployment.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/492/1*KSuT0WAmEtlYAGb-SKcdXg.jpeg" /><figcaption>Force Deployments</figcaption></figure><p>IF your task comes back up healthy you are all fine….. but if it doesn’t you now have an out of date Task Definition being used live. I would not recommend this approach beyond Dev/Testing, but it’s going to be the quickest to get setup.</p><pre>- aws ecs update-service --cluster $CLUSTER --service $SERVICE  --task-definition $TASK_DEF:$REVISION_NUMBER --force-new-deployment</pre><h3>That’s it for the first part in AWS fargate deployments, watch out for the next article where I go over some more amazing features of AWS fargate.</h3><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c2a8a8349057" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>