rss.xml

<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>technicalwriting.dev</title>
    <link>https://technicalwriting.dev</link>
    <description>
      Field notes from the frontier of technical writing.
    </description>
    <language>en-us</language>
    <pubDate>Wed, 29 Oct 2025 21:47:07 -0700</pubDate>
    <atom:link href="https://technicalwriting.dev/rss.xml" rel="self" type="application/rss+xml"/>
    <item>
      <title>word2vec-style vector arithmetic on docs embeddings</title>
      <link>https://technicalwriting.dev/embeddings/arithmetic/index.html</link>
      <description>
        Does word2vec-style vector arithmetic (e.g. vector("King") -
        vector("Man") + vector("Woman") = vector("Queen")) work in technical
        writing contexts?
      </description>
      <pubDate>Wed, 29 Oct 2025 21:47:07 -0700</pubDate>
      <guid>https://technicalwriting.dev/embeddings/arithmetic/index.html</guid>
    </item>
    <item>
      <title>Docs for AI agents</title>
      <link>https://technicalwriting.dev/ai/agents/index.html</link>
      <description>
        What are docs for AI agents? How are they different than internal eng docs? Do
        we really have to maintain the agent docs and eng docs as separate docs sets?
        This page contains my notes on these questions.
      </description>
      <pubDate>Wed, 25 Jun 2025 16:35:58 -0700</pubDate>
      <guid>https://technicalwriting.dev/ai/agents/index.html</guid>
    </item>
    <item>
      <title>Understanding task types in the Gemini Embedding API</title>
      <link>https://technicalwriting.dev/ml/embeddings/tasks/index.html</link>
      <description>
        The models.embedContent method of the Gemini Embedding API supports a
        taskType parameter. This parameter lets you specify what kind of task the
        embedding will be used for. Let's try to deduce how this "task type" thing
        works.
      </description>
      <pubDate>Thu, 15 May 2025 15:34:40 -0700</pubDate>
      <guid>https://technicalwriting.dev/ml/embeddings/tasks/index.html</guid>
    </item>
    <item>
      <title>Automating code deletion with Gemini (and a little Python)</title>
      <link>https://technicalwriting.dev/ml/gn.html</link>
      <description>
        I used Gemini 2.0 Flash and a little Python to automate the process of
        removing code from over 200 GN build files. Here's how it went.
      </description>
      <pubDate>Tue, 29 Apr 2025 17:17:17 -0700</pubDate>
      <guid>https://technicalwriting.dev/ml/gn.html</guid>
    </item>
    <item>
      <title>Export Google Analytics data to Sheets via Apps Script</title>
      <link>https://technicalwriting.dev/analytics/sheets.html</link>
      <description>
        This tutorial shows you how to use Google Apps Script to
        export your Google Analytics data to Google Sheets. The data automatically
        updates every night. There’s also a custom menu item within Sheets to update
        the data on-demand. You can also optionally also expose the data over HTTPS
        to the public internet.
      </description>
      <pubDate>Fri, 25 Apr 2025 15:00:12 -0700</pubDate>
      <guid>https://technicalwriting.dev/analytics/sheets.html</guid>
    </item>
    <item>
      <title>First impressions of Pocket Flow's tutorial generator</title>
      <link>https://technicalwriting.dev/ml/pocketflow/index.html</link>
      <description>
        Pocket Flow's tutorial generator describes itself as an AI agent
        that analyzes GitHub repositories and creates beginner-friendly
        tutorials for codebase contributors. With its default settings,
        the output from TCK was frankly unusable. It did not produce a
        tutorial, and the writing was not geared towards codebase contributors.
        BUT! With very little tweaking, I was able to get content that is very
        well-suited for codebase contributors. I was not able to get it to
        produce veritable tutorial content, though.
      </description>
      <pubDate>Mon, 21 Apr 2025 17:12:09 -0700</pubDate>
      <guid>https://technicalwriting.dev/ml/pocketflow/index.html</guid>
    </item>
    <item>
      <title>The good, the bad, and the ugly of managing Sphinx projects with Bazel</title>
      <link>https://technicalwriting.dev/sphinx/bazel/context.html</link>
      <description>
        My experience of managing Sphinx projects with Bazel.
      </description>
      <pubDate>Tue, 1 Apr 2025 16:43:36 -0700</pubDate>
      <guid>https://technicalwriting.dev/sphinx/bazel/context.html</guid>
    </item>
    <item>
      <title>A tutorial on managing Sphinx projects with Bazel</title>
      <link>https://technicalwriting.dev/sphinx/bazel/tutorial.html</link>
      <description>
        This tutorial shows you how to manage core Sphinx workflows through
        Bazel. You'll learn how to: set up the Bazel build system, build the
        Sphinx docs, inspect the built docs, spin up a local server to preview
        the docs, add a Sphinx extension to the Bazel build, and deploy to
        GitHub Pages.
      </description>
      <pubDate>Tue, 1 Apr 2025 16:42:35 -0700</pubDate>
      <guid>https://technicalwriting.dev/sphinx/bazel/tutorial.html</guid>
    </item>
    <item>
      <title>2024 Machine Learning Review (For Technical Writers)</title>
      <link>https://technicalwriting.dev/ml/reviews/2024.html</link>
      <description>
        A review of how much (or little) the ideas from GenAI Outlook 2023 have panned
        out, and a discussion of potential future trends in 2025.
      </description>
      <pubDate>Fri, 3 Jan 2025 12:41:04 -0700</pubDate>
      <guid>https://technicalwriting.dev/ml/reviews/2024.html</guid>
    </item>
    <item>
      <title>The intractable challenges of technical writing</title>
      <link>https://technicalwriting.dev/strategy/challenges.html</link>
      <description>
        There are 3 intractable challenges in technical writing.
        I do not believe we will ever be able to completely solve these challenges
        using only the practices and technologies of the 2010s.
      </description>
      <pubDate>Fri, 3 Jan 2025 12:41:04 -0700</pubDate>
      <guid>https://technicalwriting.dev/strategy/challenges.html</guid>
    </item>
    <item>
      <title>Embeddings are underrated</title>
	    <link>https://technicalwriting.dev/embeddings/overview.html</link>
	    <description>
        Machine learning (ML) has the potential to advance the state of the art
        in technical writing. No, I'm not talking about text generation models.
        The ML technology that might end up having the biggest impact on technical
        writing is embeddings. What embeddings offer to technical writers is the
        ability to discover connections between texts at previously impossible scales.
      </description>
	    <pubDate>Mon, 21 Oct 2024 09:09:03 -0700</pubDate>
	    <guid>https://technicalwriting.dev/embeddings/overview.html</guid>
    </item>
    <item>
      <title>Exploring the intertwingularity of a docs site</title>
	    <link>https://technicalwriting.dev/data/intertwingularity.html</link>
	    <description>
        I’m building a web crawler so that I can track how pages in my docs
        site link to each other and to the outside web more broadly. If a lot of my docs pages link to
        some particular page, then that page is probably important. PageRank Lite, basically, except
        with much more focus on intra-site backlinks.
      </description>
	    <pubDate>Wed, 16 Oct 2024 22:02:44 -0700</pubDate>
	    <guid>https://technicalwriting.dev/data/intertwingularity.html</guid>
    </item>
    <item>
      <title>Focus on decisions, not tasks</title>
	    <link>https://technicalwriting.dev/strategy/decisions.html</link>
	    <description>
        A quote from Every Page Is Page One that has deeply changed how I
        approach technical writing.
      </description>
	    <pubDate>Tue, 15 Oct 2024 16:16:16 -0700</pubDate>
	    <guid>https://technicalwriting.dev/strategy/decisions.html</guid>
    </item>
    <item>
      <title>You can deeplink to a specific PDF page</title>
	    <link>https://technicalwriting.dev/www/pdf.html</link>
	    <description>
        Just append #page=X to your URL, where X is the page you
		    want to deeplink to.
      </description>
	    <pubDate>Thu, 11 Jul 2024 09:01:07 -0700</pubDate>
	    <guid>https://technicalwriting.dev/www/pdf.html</guid>
    </item>
  </channel>
</rss>