507 lines
22 KiB
HTML
507 lines
22 KiB
HTML
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
|
||
|
||
<html>
|
||
<head><meta charset="UTF-8">
|
||
<title>To vibe or not to vibe</title>
|
||
<meta http-equiv="Content-type" content="text/html;charset=UTF-8" /><script defer src="https://cloud.umami.is/script.js" data-website-id="eb12527f-b713-4afa-905a-8a50f8a7f157"></script>
|
||
<meta content = 'summary' name = 'twitter:card'></meta>
|
||
|
||
<meta content = '16665197' name = 'twitter:site:id'></meta>
|
||
|
||
<meta content = '@martinfowler' name = 'twitter:site'></meta>
|
||
|
||
<meta content = 'To vibe or not to vibe' property = 'og:title'></meta>
|
||
|
||
<meta content = 'https://martinfowler.com/articles/exploring-gen-ai.html/to-vibe-or-not-vibe.html' property = 'og:url'></meta>
|
||
|
||
<meta content = 'Notes from my Thoughtworks colleagues on AI-assisted software delivery' property = 'og:description'></meta>
|
||
|
||
<meta content = 'https://martinfowler.com/articles/exploring-gen-ai/donkey-card.png' property = 'og:image'></meta>
|
||
|
||
<meta content = 'martinfowler.com' property = 'og:site_name'></meta>
|
||
|
||
<meta content = 'article' property = 'og:type'></meta>
|
||
|
||
<meta content = '23 September 2025' property = 'og:article:modified_time'></meta>
|
||
|
||
<meta content = 'width=device-width, initial-scale=1' name = 'viewport'></meta>
|
||
|
||
<link href = 'exploring-gen-ai.css' rel = 'stylesheet' type = 'text/css'></link>
|
||
</head>
|
||
|
||
<body><header id = 'banner' style = 'background-image: url("/banner.png"); background-repeat: no-repeat'>
|
||
|
||
<div class = 'name-logo'><a href = 'https://martinfowler.com'><img src = '/mf-name-white.png'></img></a></div>
|
||
<div class = 'search'>
|
||
<!-- SiteSearch Google -->
|
||
<form method='GET' action="https://www.google.com/search">
|
||
<input type='hidden' name='ie' value='UTF-8'/>
|
||
<input type='hidden' name='oe' value='UTF-8'/>
|
||
<input class = 'field' type='text'
|
||
name='q' size='15' maxlength='255' value=""/>
|
||
<button class = 'button' type='submit'
|
||
name='btnG' value=" " title = "Search"/>
|
||
<input type='hidden' name='domains' value="martinfowler.com"/>
|
||
<input type='hidden' name='sitesearch' value=""/>
|
||
<input
|
||
type='hidden' name='sitesearch' value="martinfowler.com"/>
|
||
</form>
|
||
</div>
|
||
|
||
<div class = 'menu-button navmenu-button'><a class = 'icon icon-bars' href = '#navmenu-bottom'></a></div>
|
||
|
||
<nav class = 'top-menu'>
|
||
<ul>
|
||
<li><a class = '' href = 'https://refactoring.com'>Refactoring</a></li>
|
||
|
||
<li><a class = '' href = '/agile.html'>Agile</a></li>
|
||
|
||
<li><a class = '' href = '/architecture'>Architecture</a></li>
|
||
|
||
<li><a class = '' href = '/aboutMe.html'>About</a></li>
|
||
|
||
<li><a class = 'tw' href = 'https://www.thoughtworks.com/engineering'>Thoughtworks</a></li>
|
||
|
||
<li><a class = 'icon icon-rss' href = '/feed.atom' title = 'feed'></a></li>
|
||
|
||
<li><a class = 'icon icon-twitter' href = 'https://www.twitter.com/martinfowler' title = 'Twitter stream'></a></li>
|
||
|
||
<li class = 'icon'><a href = 'https://toot.thoughtworks.com/@mfowler' title = 'Mastodon stream'><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="currentColor"><path d="M21.2595 13.9898C20.9852 15.4006 18.8033 16.9446 16.2974 17.2439C14.9907 17.3998 13.7041 17.5431 12.3321 17.4802C10.0885 17.3774 8.31809 16.9446 8.31809 16.9446C8.31809 17.163 8.33156 17.371 8.3585 17.5655C8.65019 19.7797 10.5541 19.9124 12.3576 19.9742C14.1779 20.0365 15.7987 19.5254 15.7987 19.5254L15.8735 21.1711C15.8735 21.1711 14.6003 21.8548 12.3321 21.9805C11.0814 22.0493 9.52849 21.9491 7.71973 21.4703C3.79684 20.432 3.12219 16.2504 3.01896 12.0074C2.98749 10.7477 3.00689 9.55981 3.00689 8.56632C3.00689 4.22771 5.84955 2.95599 5.84955 2.95599C7.2829 2.29772 9.74238 2.0209 12.2993 2H12.3621C14.919 2.0209 17.3801 2.29772 18.8133 2.95599C18.8133 2.95599 21.6559 4.22771 21.6559 8.56632C21.6559 8.56632 21.6916 11.7674 21.2595 13.9898ZM18.3029 8.9029C18.3029 7.82924 18.0295 6.97604 17.4805 6.34482C16.9142 5.71359 16.1726 5.39001 15.2522 5.39001C14.187 5.39001 13.3805 5.79937 12.8473 6.61819L12.3288 7.48723L11.8104 6.61819C11.2771 5.79937 10.4706 5.39001 9.40554 5.39001C8.485 5.39001 7.74344 5.71359 7.17719 6.34482C6.62807 6.97604 6.3547 7.82924 6.3547 8.9029V14.1562H8.43597V9.05731C8.43597 7.98246 8.88822 7.4369 9.79281 7.4369C10.793 7.4369 11.2944 8.08408 11.2944 9.36376V12.1547H13.3634V9.36376C13.3634 8.08408 13.8646 7.4369 14.8648 7.4369C15.7694 7.4369 16.2216 7.98246 16.2216 9.05731V14.1562H18.3029V8.9029Z"></path></svg>
|
||
</a></li>
|
||
|
||
<li class = 'icon'><a href = 'https://www.linkedin.com/in/martin-fowler-com/' title = 'LinkedIn'><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="currentColor"><path d="M4.00098 3H20.001C20.5533 3 21.001 3.44772 21.001 4V20C21.001 20.5523 20.5533 21 20.001 21H4.00098C3.44869 21 3.00098 20.5523 3.00098 20V4C3.00098 3.44772 3.44869 3 4.00098 3ZM5.00098 5V19H19.001V5H5.00098ZM7.50098 9C6.67255 9 6.00098 8.32843 6.00098 7.5C6.00098 6.67157 6.67255 6 7.50098 6C8.3294 6 9.00098 6.67157 9.00098 7.5C9.00098 8.32843 8.3294 9 7.50098 9ZM6.50098 10H8.50098V17.5H6.50098V10ZM12.001 10.4295C12.5854 9.86534 13.2665 9.5 14.001 9.5C16.072 9.5 17.501 11.1789 17.501 13.25V17.5H15.501V13.25C15.501 12.2835 14.7175 11.5 13.751 11.5C12.7845 11.5 12.001 12.2835 12.001 13.25V17.5H10.001V10H12.001V10.4295Z"></path></svg>
|
||
</a></li>
|
||
|
||
<li class = 'icon'><a href = 'https://bsky.app/profile/martinfowler.com' title = 'BlueSky'><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24" fill="currentColor"><path d="M12 11.3884C11.0942 9.62673 8.62833 6.34423 6.335 4.7259C4.13833 3.17506 3.30083 3.4434 2.75167 3.69256C2.11583 3.9784 2 4.95506 2 5.52839C2 6.10339 2.315 10.2367 2.52 10.9276C3.19917 13.2076 5.61417 13.9776 7.83917 13.7309C4.57917 14.2142 1.68333 15.4017 5.48083 19.6292C9.65833 23.9542 11.2058 18.7017 12 16.0392C12.7942 18.7017 13.7083 23.7651 18.4442 19.6292C22 16.0392 19.4208 14.2142 16.1608 13.7309C18.3858 13.9784 20.8008 13.2076 21.48 10.9276C21.685 10.2376 22 6.10256 22 5.52923C22 4.95423 21.8842 3.97839 21.2483 3.6909C20.6992 3.44256 19.8617 3.17423 17.665 4.72423C15.3717 6.34506 12.9058 9.62756 12 11.3884Z"></path></svg></a></li>
|
||
</ul>
|
||
</nav>
|
||
</header>
|
||
<nav id = 'top-navmenu'>
|
||
<nav class = 'navmenu'>
|
||
<div class = 'nav-head'> <div class = 'search'>
|
||
<!-- SiteSearch Google -->
|
||
<form method='GET' action="https://www.google.com/search">
|
||
<input type='hidden' name='ie' value='UTF-8'/>
|
||
<input type='hidden' name='oe' value='UTF-8'/>
|
||
<input class = 'field' type='text'
|
||
name='q' size='15' maxlength='255' value=""/>
|
||
<button class = 'button' type='submit'
|
||
name='btnG' value=" " title = "Search"/>
|
||
<input type='hidden' name='domains' value="martinfowler.com"/>
|
||
<input type='hidden' name='sitesearch' value=""/>
|
||
<input
|
||
type='hidden' name='sitesearch' value="martinfowler.com"/>
|
||
</form>
|
||
</div>
|
||
|
||
<div class = 'closediv'>
|
||
<span class = 'close' title = 'close'></span>
|
||
</div>
|
||
</div>
|
||
|
||
<div class = 'nav-body'>
|
||
<div class = 'topics'>
|
||
<h2>Topics</h2>
|
||
|
||
<p><a href = '/architecture'>Architecture</a></p>
|
||
|
||
<p><a href = 'https://refactoring.com'>Refactoring</a></p>
|
||
|
||
<p><a href = '/agile.html'>Agile</a></p>
|
||
|
||
<p><a href = '/delivery.html'>Delivery</a></p>
|
||
|
||
<p><a href = '/microservices'>Microservices</a></p>
|
||
|
||
<p><a href = '/data'>Data</a></p>
|
||
|
||
<p><a href = '/testing'>Testing</a></p>
|
||
|
||
<p><a href = '/dsl.html'>DSL</a></p>
|
||
</div>
|
||
|
||
<div class = 'about'>
|
||
<h2>about me</h2>
|
||
|
||
<p><a href = '/aboutMe.html'>About</a></p>
|
||
|
||
<p><a href = '/books'>Books</a></p>
|
||
|
||
<p><a href = '/faq.html'>FAQ</a></p>
|
||
</div>
|
||
|
||
<div class = 'content'>
|
||
<h2>content</h2>
|
||
|
||
<p><a href = '/videos.html'>Videos</a></p>
|
||
|
||
<p><a href = '/tags'>Content Index</a></p>
|
||
|
||
<p><a href = '/fragments'>Fragments</a></p>
|
||
|
||
<p><a href = '/boardgames'>Board Games</a></p>
|
||
|
||
<p><a href = '/photos'>Photography</a></p>
|
||
</div>
|
||
|
||
<div class = 'tw'>
|
||
<h2>Thoughtworks</h2>
|
||
|
||
<p><a href = 'https://thoughtworks.com'>Home</a></p>
|
||
|
||
<p><a href = 'https://thoughtworks.com/insights'>Insights</a></p>
|
||
|
||
<p><a href = 'https://thoughtworks.com/careers'>Careers</a></p>
|
||
|
||
<p><a href = 'https://thoughtworks.com/radar'>Radar</a></p>
|
||
|
||
<p><a href = 'https://www.thoughtworks.com/engineering'>Engineering</a></p>
|
||
</div>
|
||
|
||
<div class = 'feeds'>
|
||
<h2>follow</h2>
|
||
|
||
<p><a href = '/feed.atom'>RSS</a></p>
|
||
|
||
<p><a href = 'https://toot.thoughtworks.com/@mfowler'>Mastodon</a></p>
|
||
|
||
<p><a href = 'https://www.linkedin.com/in/martin-fowler-com/'>LinkedIn</a></p>
|
||
|
||
<p><a href = 'https://bsky.app/profile/martinfowler.com'>Bluesky</a></p>
|
||
|
||
<p><a href = 'https://www.twitter.com/martinfowler'>X</a></p>
|
||
|
||
<p><a href = 'https://boardgamegeek.com/blog/13064/martins-7th-decade'>BGG</a></p>
|
||
</div>
|
||
</div>
|
||
</nav>
|
||
</nav>
|
||
|
||
<main class = 'memo '>
|
||
<div class = 'frontMatter'>
|
||
<div class = 'main'>
|
||
<h1>To vibe or not to vibe</h1>
|
||
|
||
<div class = 'front-grid'>
|
||
<div class = 'author-list'>
|
||
<div class = 'author'>
|
||
<div class = 'photo'><a href = 'https://birgitta.info'><img alt = 'Photo of Birgitta Böckeler' src = 'bb.jpg' width = '80'></img></a></div>
|
||
|
||
<address class = 'name'><a href = 'https://birgitta.info' rel = 'author'>Birgitta Böckeler</a></address>
|
||
|
||
<div class = 'bio'>
|
||
<p>Birgitta is a Distinguished Engineer and AI-assisted delivery
|
||
expert at Thoughtworks. She has over 20 years of experience as a software
|
||
developer, architect and technical leader.</p>
|
||
</div>
|
||
</div>
|
||
</div>
|
||
</div>
|
||
</div>
|
||
|
||
<div class = 'right'>
|
||
<div class = 'catalog-notice'>
|
||
<p><a href = '../exploring-gen-ai.html'><img src = 'donkey-card.png'></img></a></p>
|
||
|
||
<p>This article is part of <a href = '../exploring-gen-ai.html'>“Exploring Gen
|
||
AI”</a>. A series capturing Thoughtworks technologists' explorations of using gen ai technology for
|
||
software development.</p>
|
||
</div>
|
||
|
||
<p class = 'date'>23 September 2025</p>
|
||
</div>
|
||
</div>
|
||
|
||
<div class = 'paperBody'><p>The discourse about to what level AI-generated code should be reviewed often feels very binary. Is vibe coding (i.e. letting AI generate code without looking at the code) good or bad? The answer is of course neither, because “it depends”.</p>
|
||
|
||
<p>So what does it depend on?</p>
|
||
|
||
<p>When I’m using AI for coding, I find myself constantly making little risk assessments about whether to trust the AI, how much to trust it, and how much work I need to put into the verification of the results. And the more experience I get with using AI, the more honed and intuitive these assessments become.</p>
|
||
|
||
<p>Risk assessment is typically a combination of three factors:</p>
|
||
|
||
<ol>
|
||
<li>Probability</li>
|
||
<li>Impact</li>
|
||
<li>Detectability</li>
|
||
</ol>
|
||
|
||
<p>Reflecting on these 3 dimensions helps me decide if I should reach for AI or not, if I should review the code or not, and at what level of detail I do that review. This also helps me think about mitigations I can put in place when I want to take advantage of AI’s speed, but reduce the risk of it doing the wrong thing.</p>
|
||
|
||
<h2 id="1-probability-how-likely-is-ai-to-get-things-wrong">1. Probability: How likely is AI to get things wrong?</h2>
|
||
|
||
<p>The following are some of the factors that help you determine the probability dimension.</p>
|
||
|
||
<h3 id="know-your-tool">Know your tool</h3>
|
||
|
||
<p>The AI coding assistant is a function of the model used, the prompt orchestration happening in the tool, and the level of integration the assistant has with the codebase and the development environment. As developers, we don’t have all the information about what is going on under the hood, especially when we’re using a proprietary tool. So the assessment of the tool quality is a combination of knowing about its proclaimed features and our own previous experience with it.</p>
|
||
|
||
<h3 id="is-the-use-case-ai-friendly">Is the use case AI-friendly?</h3>
|
||
|
||
<p>Is the tech stack prevalent in the training data? What is the complexity of the solution you want AI to create? How big is the problem that AI is supposed to solve?</p>
|
||
|
||
<p>You can also more generally consider if you’re working on a use case that needs a high level of “correctness”, or not. E.g., building a screen exactly based on a design, or drafting a rough prototype screen.</p>
|
||
|
||
<h3 id="be-aware-of-the-available-context">Be aware of the available context</h3>
|
||
|
||
<p>Probability isn’t only about the model and the tool, it’s also about the available <strong>context</strong>. The context is the prompt you provide, plus all the other information the agent has access to via tool calls etc.</p>
|
||
|
||
<ul>
|
||
<li>
|
||
<p>Does the AI assistant have enough <strong>access to your codebase</strong> to make a good decision? Is it seeing the files, the structure, the domain logic? If not, the chance that it will generate something unhelpful goes up.</p>
|
||
</li>
|
||
<li>
|
||
<p>How effective is your tool’s <strong>code search strategy</strong>? Some tools index the entire codebase, some make on the fly <code>grep</code>-like searches over the files, some build a graph with the help of the AST (Abstract Syntax Tree). It can help to know what strategy your tool of choice uses, though ultimately only experience with the tool will tell you how well that strategy really works.</p>
|
||
</li>
|
||
<li>
|
||
<p>Is the <strong>codebase <a href="https://www.thoughtworks.com/radar/techniques/ai-friendly-code-design">AI-friendly</a></strong>, i.e. is it structured in a way that makes it easy for AI to work with? Is it modular, with clear boundaries and interfaces? Or is it a big ball of mud that fills up the context window quickly?</p>
|
||
</li>
|
||
<li>
|
||
<p>Is the existing <strong>codebase setting a good example</strong>? Or is it a mess of hacks and anti-patterns? If the latter, the chance of AI generating more of the same goes up if you don’t explicitly tell it what the good examples are.</p>
|
||
</li>
|
||
</ul>
|
||
|
||
<h2 id="2-impact-if-ai-gets-it-wrong-and-you-dont-notice-what-are-the-consequences">2. Impact: If AI gets it wrong and you don’t notice, what are the consequences?</h2>
|
||
|
||
<p>This consideration is mainly about the <strong>use case</strong>. Are you working on a spike or production code? Are you on call for the service you are working on? Is it business critical, or just internal tooling?</p>
|
||
|
||
<p>Some good sanity checks:</p>
|
||
|
||
<ul>
|
||
<li>Would you ship this if you were on call tonight?</li>
|
||
<li>Does this code have a high impact radius, e.g. is it used by a lot of other components or consumers?</li>
|
||
</ul>
|
||
|
||
<h2 id="3-detectability-will-you-notice-when-ai-gets-it-wrong">3. Detectability: Will you notice when AI gets it wrong?</h2>
|
||
|
||
<p>This is about <strong>feedback loops</strong>. Do you have good tests? Are you using a typed language? Does your stack make failures obvious? Do you trust the tool’s change tracking and diffs?</p>
|
||
|
||
<p>It also comes down to your own familiarity with the codebase. If you know the tech stack and the use case well, you’re more likely to spot something fishy.</p>
|
||
|
||
<p>This dimension leans heavily on traditional engineering skills: test coverage, system knowledge, code review practices. And it influences how confident you can be even when AI makes the change for you.</p>
|
||
|
||
<h2 id="a-combination-of-traditional-and-new-skills">A combination of traditional and new skills</h2>
|
||
|
||
<p>You might have already noticed that many of these assessment questions require “traditional” engineering skills, others</p>
|
||
|
||
<p class="full"><img src="overview_three_dimensions.png" alt="A summary of the 3 dimensions, as described in the article text. They are color coded in two different colors, marking factors like knowing your tool and context as “New AI skills and experience to develop”, and others (like maintainability, safety net, …) as “Traditional skills and experience”" /></p>
|
||
|
||
<h2 id="combining-the-three-a-sliding-scale-of-review-effort">Combining the three: A sliding scale of review effort</h2>
|
||
|
||
<p>When you combine these three dimensions, they can guide your level of oversight. Let’s take the extremes as an example to illustrate this idea:</p>
|
||
|
||
<ul>
|
||
<li><strong>Low probability + low impact + high detectability</strong> Vibe coding is fine! As long as things work and I achieve my goal, I don’t review the code at all.</li>
|
||
<li><strong>High probability + high impact + low detectability</strong> High level of review is advisable. Assume the AI might be wrong and cover for it.</li>
|
||
</ul>
|
||
|
||
<p>Most situations land somewhere in between of course.</p>
|
||
|
||
<p class="full"><img src="extreme_scenarios.png" alt="An illustration showing the two extreme cases of the 3 dimensions: Low probability + low impact + high detectability is the perfect case for vibe coding; High probability + high impact + low detectability is the case that needs the most human scrutiny" /></p>
|
||
|
||
<h2 id="example-legacy-reverse-engineering">Example: Legacy reverse engineering</h2>
|
||
|
||
<p>We recently worked on a legacy migration for a client where the first step was to create a detailed description of the existing functionality with AI’s help.</p>
|
||
|
||
<ul>
|
||
<li>
|
||
<p><strong>Probability of getting wrong descriptions</strong> was medium:</p>
|
||
|
||
<ul>
|
||
<li>
|
||
<p><strong>Tool:</strong> The model we had to use often failed to follow instructions well</p>
|
||
</li>
|
||
<li>
|
||
<p><strong>Available context:</strong> we did not have access to all of the code, the backend code was unavailable.</p>
|
||
</li>
|
||
<li>
|
||
<p><strong>Mitigations</strong>: We ran prompts multiple times to spot check variance in results, and we increased our confidence level by analysing the decompiled backend binary.</p>
|
||
</li>
|
||
</ul>
|
||
</li>
|
||
<li>
|
||
<p><strong>Impact of getting wrong descriptions</strong> was medium</p>
|
||
|
||
<ul>
|
||
<li>
|
||
<p><strong>Business use case:</strong> On the one hand, the system was used by thousands of external business partners of this organization, so getting the rebuild wrong posed a business risk to reputation and revenue.</p>
|
||
</li>
|
||
<li>
|
||
<p><strong>Complexity:</strong> On the other hand, the complexity of the application was relatively low, so we expected it to be quite easy to fix mistakes.</p>
|
||
</li>
|
||
<li>
|
||
<p><strong>Planned mitigations</strong>: A staggered rollout of the new application.</p>
|
||
</li>
|
||
</ul>
|
||
</li>
|
||
<li>
|
||
<p><strong>Detectability of getting the wrong descriptions</strong> was medium</p>
|
||
|
||
<ul>
|
||
<li>
|
||
<p><strong>Safety net:</strong> There was no existing test suite that could be cross-checked</p>
|
||
</li>
|
||
<li>
|
||
<p><strong>SME availability:</strong> We planned to bring in SMEs for review, and to create a feature parity comparison tests.</p>
|
||
</li>
|
||
</ul>
|
||
</li>
|
||
</ul>
|
||
|
||
<p>Without a structured assessment like this, it would have been easy to under-review or over-review. Instead, we calibrated our approach and planned for mitigations.</p>
|
||
|
||
<h2 id="closing-thought">Closing thought</h2>
|
||
|
||
<p>This kind of micro risk assessment becomes second nature. The more you use AI, the more you build intuition for these questions. You start to feel which changes can be trusted and which need closer inspection.</p>
|
||
|
||
<p>The goal is not to slow yourself down with checklists, but to grow intuitive habits that help you navigate the line between leveraging AI’s capabilities while reducing the risk of its downsides.</p>
|
||
|
||
</div>
|
||
|
||
<div class = 'series-nav'>
|
||
<div class = 'latest'>
|
||
<p class = 'label'>latest article (Sep 02):</p>
|
||
|
||
<p><a href = 'an-accidental-blackboard.html'>An Accidental Blackboard</a></p>
|
||
</div>
|
||
|
||
<div class = 'prev'>
|
||
<p class = 'label'>previous article:</p>
|
||
|
||
<p><a href = 'partner-with-ai-and-throw-away-the-code.html'>Partner with the AI, throw away the code</a></p>
|
||
</div>
|
||
|
||
<div class = 'next'>
|
||
<p class = 'label'>next article:</p>
|
||
|
||
<p><a href = 'anchoring-to-reference.html'>Anchoring AI to a reference application</a></p>
|
||
</div>
|
||
|
||
<div class = 'index'>
|
||
<p><a href = '/articles/exploring-gen-ai.html'><img src = 'donkey-card.png'></img></a></p>
|
||
</div>
|
||
</div>
|
||
</main>
|
||
|
||
<nav id = 'bottom-navmenu'>
|
||
<nav class = 'navmenu'>
|
||
<div class = 'nav-head'> <div class = 'search'>
|
||
<!-- SiteSearch Google -->
|
||
<form method='GET' action="https://www.google.com/search">
|
||
<input type='hidden' name='ie' value='UTF-8'/>
|
||
<input type='hidden' name='oe' value='UTF-8'/>
|
||
<input class = 'field' type='text'
|
||
name='q' size='15' maxlength='255' value=""/>
|
||
<button class = 'button' type='submit'
|
||
name='btnG' value=" " title = "Search"/>
|
||
<input type='hidden' name='domains' value="martinfowler.com"/>
|
||
<input type='hidden' name='sitesearch' value=""/>
|
||
<input
|
||
type='hidden' name='sitesearch' value="martinfowler.com"/>
|
||
</form>
|
||
</div>
|
||
|
||
<div class = 'closediv'>
|
||
<span class = 'close' title = 'close'></span>
|
||
</div>
|
||
</div>
|
||
|
||
<div class = 'nav-body'>
|
||
<div class = 'topics'>
|
||
<h2>Topics</h2>
|
||
|
||
<p><a href = '/architecture'>Architecture</a></p>
|
||
|
||
<p><a href = 'https://refactoring.com'>Refactoring</a></p>
|
||
|
||
<p><a href = '/agile.html'>Agile</a></p>
|
||
|
||
<p><a href = '/delivery.html'>Delivery</a></p>
|
||
|
||
<p><a href = '/microservices'>Microservices</a></p>
|
||
|
||
<p><a href = '/data'>Data</a></p>
|
||
|
||
<p><a href = '/testing'>Testing</a></p>
|
||
|
||
<p><a href = '/dsl.html'>DSL</a></p>
|
||
</div>
|
||
|
||
<div class = 'about'>
|
||
<h2>about me</h2>
|
||
|
||
<p><a href = '/aboutMe.html'>About</a></p>
|
||
|
||
<p><a href = '/books'>Books</a></p>
|
||
|
||
<p><a href = '/faq.html'>FAQ</a></p>
|
||
</div>
|
||
|
||
<div class = 'content'>
|
||
<h2>content</h2>
|
||
|
||
<p><a href = '/videos.html'>Videos</a></p>
|
||
|
||
<p><a href = '/tags'>Content Index</a></p>
|
||
|
||
<p><a href = '/fragments'>Fragments</a></p>
|
||
|
||
<p><a href = '/boardgames'>Board Games</a></p>
|
||
|
||
<p><a href = '/photos'>Photography</a></p>
|
||
</div>
|
||
|
||
<div class = 'tw'>
|
||
<h2>Thoughtworks</h2>
|
||
|
||
<p><a href = 'https://thoughtworks.com'>Home</a></p>
|
||
|
||
<p><a href = 'https://thoughtworks.com/insights'>Insights</a></p>
|
||
|
||
<p><a href = 'https://thoughtworks.com/careers'>Careers</a></p>
|
||
|
||
<p><a href = 'https://thoughtworks.com/radar'>Radar</a></p>
|
||
|
||
<p><a href = 'https://www.thoughtworks.com/engineering'>Engineering</a></p>
|
||
</div>
|
||
|
||
<div class = 'feeds'>
|
||
<h2>follow</h2>
|
||
|
||
<p><a href = '/feed.atom'>RSS</a></p>
|
||
|
||
<p><a href = 'https://toot.thoughtworks.com/@mfowler'>Mastodon</a></p>
|
||
|
||
<p><a href = 'https://www.linkedin.com/in/martin-fowler-com/'>LinkedIn</a></p>
|
||
|
||
<p><a href = 'https://bsky.app/profile/martinfowler.com'>Bluesky</a></p>
|
||
|
||
<p><a href = 'https://www.twitter.com/martinfowler'>X</a></p>
|
||
|
||
<p><a href = 'https://boardgamegeek.com/blog/13064/martins-7th-decade'>BGG</a></p>
|
||
</div>
|
||
</div>
|
||
</nav>
|
||
</nav>
|
||
<footer id='page-footer'>
|
||
<div class='tw-logo'>
|
||
<a href='https://www.thoughtworks.com/engineering'>
|
||
<img src='/thoughtworks_white.png'>
|
||
</a>
|
||
</div>
|
||
<div class='menu-button'>
|
||
<div class='icon-bars navmenu-button'></div>
|
||
</div>
|
||
<div class='copyright'>
|
||
<p>© Martin Fowler | <a href="/aboutMe.html#disclosures">Disclosures</a></p>
|
||
</div>
|
||
</footer>
|
||
|
||
<script src = '/jquery-1.11.3.min.js' type = 'text/javascript'></script>
|
||
|
||
<script src = '/mfcom.js' type = 'text/javascript'></script>
|
||
|
||
<script src = 'exploring-gen-ai.js' type = 'text/javascript'></script>
|
||
</body>
|
||
</html>
|