<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: zhaoq</title>
    <description>The latest articles on DEV Community by zhaoq (@zhaoq0103).</description>
    <link>https://dev.to/zhaoq0103</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4210959%2F9f23e0b1-07f3-48a3-a3e9-f0fed551e1bc.jpg</url>
      <title>DEV Community: zhaoq</title>
      <link>https://dev.to/zhaoq0103</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/zhaoq0103"/>
    <language>en</language>
    <item>
      <title>Writing content that survives patch notes: version stamps for static sites</title>
      <dc:creator>zhaoq</dc:creator>
      <pubDate>Fri, 30 Oct 2026 23:46:23 +0000</pubDate>
      <link>https://dev.to/zhaoq0103/writing-content-that-survives-patch-notes-version-stamps-for-static-sites-5g7</link>
      <guid>https://dev.to/zhaoq0103/writing-content-that-survives-patch-notes-version-stamps-for-static-sites-5g7</guid>
      <description>&lt;p&gt;A game guide has a property no ordinary article has: it can become wrong without anyone editing it.&lt;/p&gt;

&lt;p&gt;The page does not change. The game does. A balance patch ships, an item gets nerfed, and a page that was accurate in March is misleading in June — with no diff to show for it.&lt;/p&gt;

&lt;p&gt;For a site that relies on search traffic, that is the worst kind of decay: it is invisible in every tool that looks at your own repository.&lt;/p&gt;

&lt;h2&gt;
  
  
  Stamp the version, not the date
&lt;/h2&gt;

&lt;p&gt;The lazy fix is to slap "Last updated: June 12" on the page. It is worse than nothing, because it implies verification that may not have happened. I have updated a page's typography and watched the "last updated" date move without a single fact being checked.&lt;/p&gt;

&lt;p&gt;What I do instead is record the version the page was verified against, in the frontmatter:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight yaml"&gt;&lt;code&gt;&lt;span class="nn"&gt;---&lt;/span&gt;
&lt;span class="na"&gt;title&lt;/span&gt;&lt;span class="pi"&gt;:&lt;/span&gt; &lt;span class="s2"&gt;"&lt;/span&gt;&lt;span class="s"&gt;Builds&lt;/span&gt;&lt;span class="nv"&gt; &lt;/span&gt;&lt;span class="s"&gt;and&lt;/span&gt;&lt;span class="nv"&gt; &lt;/span&gt;&lt;span class="s"&gt;artifacts"&lt;/span&gt;
&lt;span class="na"&gt;verifiedAgainst&lt;/span&gt;&lt;span class="pi"&gt;:&lt;/span&gt; &lt;span class="s"&gt;1.4.2&lt;/span&gt;
&lt;span class="na"&gt;verifiedOn&lt;/span&gt;&lt;span class="pi"&gt;:&lt;/span&gt; &lt;span class="s"&gt;2026-09-18&lt;/span&gt;
&lt;span class="na"&gt;stillValidFor&lt;/span&gt;&lt;span class="pi"&gt;:&lt;/span&gt; &lt;span class="s"&gt;1.4.x&lt;/span&gt;   &lt;span class="c1"&gt;# optional: which patch series this is expected to hold&lt;/span&gt;
&lt;span class="nn"&gt;---&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;"Verified against 1.4.2" is a claim I can defend. "Updated today" is not.&lt;/p&gt;

&lt;h2&gt;
  
  
  Make staleness a query
&lt;/h2&gt;

&lt;p&gt;The point is not the field, it is what the field lets you do. One grep answers the only question that matters after a patch:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;&lt;span class="c"&gt;# every page that claims a version older than the one that just shipped&lt;/span&gt;
&lt;span class="nb"&gt;grep&lt;/span&gt; &lt;span class="nt"&gt;-l&lt;/span&gt; &lt;span class="s2"&gt;"verifiedAgainst:"&lt;/span&gt; content/guides/&lt;span class="k"&gt;*&lt;/span&gt;.md &lt;span class="se"&gt;\&lt;/span&gt;
  | xargs &lt;span class="nb"&gt;grep&lt;/span&gt; &lt;span class="nt"&gt;-H&lt;/span&gt; &lt;span class="s2"&gt;"verifiedAgainst:"&lt;/span&gt; &lt;span class="se"&gt;\&lt;/span&gt;
  | &lt;span class="nb"&gt;grep&lt;/span&gt; &lt;span class="nt"&gt;-v&lt;/span&gt; &lt;span class="s2"&gt;"1.4.2"&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;That list is your re-verification queue. Not "everything, someday", but "these eleven pages, this week".&lt;/p&gt;

&lt;p&gt;Because the whole site is a folder of Markdown files, this is a &lt;code&gt;grep&lt;/code&gt;, not a database migration. The cheapest content-management system I have ever used is a filename convention plus a field.&lt;/p&gt;

&lt;h2&gt;
  
  
  Where it changes the writing
&lt;/h2&gt;

&lt;p&gt;Once a page carries a version, you start writing differently — for the better:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Prefer rules over numbers.&lt;/strong&gt; "Artifact stacks multiply the same stat" ages better than "artifact stacks give +37%", which will be wrong in a month.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Say what the number is for.&lt;/strong&gt; If I do quote a value, it gets a version and a source. A number without those is a liability.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Separate durable structure from volatile tuning.&lt;/strong&gt; The page's shape — what to consider, in what order, why it matters — survives patches. The specific picks do not, so they live in clearly marked sections that are easy to replace.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The &lt;a href="https://sephiriawiki.vercel.app/guides/builds-guide" rel="noopener noreferrer"&gt;Sephiria build and artifact notes&lt;/a&gt; on my &lt;a href="https://sephiriawiki.vercel.app" rel="noopener noreferrer"&gt;Sephiria Wiki&lt;/a&gt; are the page I use to test this: the parts that age fast (which stat beats which) and the parts that do not (what you are optimising for, and when) should not be interleaved. Builds get rebalanced; the decision behind them usually does not.&lt;/p&gt;

&lt;h2&gt;
  
  
  Rendering it honestly
&lt;/h2&gt;

&lt;p&gt;Render the stamp as one small line, not a badge. It should say exactly what it means:&lt;/p&gt;

&lt;blockquote&gt;
&lt;p&gt;Verified against 1.4.2 on 2026-09-18.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;Not "up to date". Not "updated regularly". A specific claim with a specific boundary, so a reader on 1.5 knows how much to trust it.&lt;/p&gt;

&lt;p&gt;That honesty is the whole value proposition on a small site where you cannot re-verify every page on every patch. There is no shortage of guides. There is a shortage of guides that tell you when they were last true.&lt;/p&gt;

&lt;h2&gt;
  
  
  The cost
&lt;/h2&gt;

&lt;p&gt;Ten minutes per page per patch cycle, and the discipline to re-run one grep. In exchange, you can answer "which of my 200 pages are stale?" in a second — and you never again have to guess whether a page is wrong or just old.&lt;/p&gt;

</description>
      <category>webdev</category>
      <category>nextjs</category>
      <category>content</category>
    </item>
    <item>
      <title>The page nobody links to: finding orphan pages in a small content site</title>
      <dc:creator>zhaoq</dc:creator>
      <pubDate>Tue, 27 Oct 2026 23:30:27 +0000</pubDate>
      <link>https://dev.to/zhaoq0103/the-page-nobody-links-to-finding-orphan-pages-in-a-small-content-site-8ka</link>
      <guid>https://dev.to/zhaoq0103/the-page-nobody-links-to-finding-orphan-pages-in-a-small-content-site-8ka</guid>
      <description>&lt;p&gt;Every "SEO checklist" article tells you to build internal links. Almost none tell you how to find out that you failed.&lt;/p&gt;

&lt;p&gt;I run a set of small game wiki sites. Most have 15–25 guide pages. Small enough that I assumed internal linking took care of itself: there is a &lt;code&gt;/guides&lt;/code&gt; index, every page is on it, done.&lt;/p&gt;

&lt;p&gt;Then I looked at the analytics, and one page had traffic while another page on the same site — similar length, published the same week — had almost none. Not because it was worse. Because nothing pointed at it except the index.&lt;/p&gt;

&lt;h2&gt;
  
  
  Orphans are a graph problem, not a content problem
&lt;/h2&gt;

&lt;p&gt;An orphan page is a page with no inbound internal links. It is not broken, it is not thin, it just sits there. Crawlers find it through the sitemap eventually; users basically never do.&lt;/p&gt;

&lt;p&gt;The fix is mechanical if you can see the graph. So I wrote the dumbest possible version of that graph:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight javascript"&gt;&lt;code&gt;&lt;span class="c1"&gt;// walk content/guides/*.md, collect [text](/guides/slug) links&lt;/span&gt;
&lt;span class="kd"&gt;const&lt;/span&gt; &lt;span class="nx"&gt;pages&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="nx"&gt;fs&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;readdirSync&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="dl"&gt;'&lt;/span&gt;&lt;span class="s1"&gt;content/guides&lt;/span&gt;&lt;span class="dl"&gt;'&lt;/span&gt;&lt;span class="p"&gt;).&lt;/span&gt;&lt;span class="nf"&gt;filter&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="nx"&gt;f&lt;/span&gt; &lt;span class="o"&gt;=&amp;gt;&lt;/span&gt; &lt;span class="nx"&gt;f&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;endsWith&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="dl"&gt;'&lt;/span&gt;&lt;span class="s1"&gt;.md&lt;/span&gt;&lt;span class="dl"&gt;'&lt;/span&gt;&lt;span class="p"&gt;));&lt;/span&gt;
&lt;span class="kd"&gt;const&lt;/span&gt; &lt;span class="nx"&gt;inbound&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="k"&gt;new&lt;/span&gt; &lt;span class="nc"&gt;Map&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="nx"&gt;pages&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;map&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="nx"&gt;p&lt;/span&gt; &lt;span class="o"&gt;=&amp;gt;&lt;/span&gt; &lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="nx"&gt;p&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="mi"&gt;0&lt;/span&gt;&lt;span class="p"&gt;]));&lt;/span&gt;

&lt;span class="k"&gt;for &lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="kd"&gt;const&lt;/span&gt; &lt;span class="nx"&gt;p&lt;/span&gt; &lt;span class="k"&gt;of&lt;/span&gt; &lt;span class="nx"&gt;pages&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt; &lt;span class="p"&gt;{&lt;/span&gt;
  &lt;span class="kd"&gt;const&lt;/span&gt; &lt;span class="nx"&gt;md&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="nx"&gt;fs&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;readFileSync&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="s2"&gt;`content/guides/&lt;/span&gt;&lt;span class="p"&gt;${&lt;/span&gt;&lt;span class="nx"&gt;p&lt;/span&gt;&lt;span class="p"&gt;}&lt;/span&gt;&lt;span class="s2"&gt;`&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="dl"&gt;'&lt;/span&gt;&lt;span class="s1"&gt;utf8&lt;/span&gt;&lt;span class="dl"&gt;'&lt;/span&gt;&lt;span class="p"&gt;);&lt;/span&gt;
  &lt;span class="k"&gt;for &lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="kd"&gt;const&lt;/span&gt; &lt;span class="nx"&gt;m&lt;/span&gt; &lt;span class="k"&gt;of&lt;/span&gt; &lt;span class="nx"&gt;md&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;matchAll&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="sr"&gt;/&lt;/span&gt;&lt;span class="se"&gt;\]\((\/&lt;/span&gt;&lt;span class="sr"&gt;guides&lt;/span&gt;&lt;span class="se"&gt;\/[&lt;/span&gt;&lt;span class="sr"&gt;a-z0-9-&lt;/span&gt;&lt;span class="se"&gt;]&lt;/span&gt;&lt;span class="sr"&gt;+&lt;/span&gt;&lt;span class="se"&gt;)\)&lt;/span&gt;&lt;span class="sr"&gt;/g&lt;/span&gt;&lt;span class="p"&gt;))&lt;/span&gt; &lt;span class="p"&gt;{&lt;/span&gt;
    &lt;span class="kd"&gt;const&lt;/span&gt; &lt;span class="nx"&gt;target&lt;/span&gt; &lt;span class="o"&gt;=&lt;/span&gt; &lt;span class="s2"&gt;`&lt;/span&gt;&lt;span class="p"&gt;${&lt;/span&gt;&lt;span class="nx"&gt;m&lt;/span&gt;&lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="mi"&gt;1&lt;/span&gt;&lt;span class="p"&gt;].&lt;/span&gt;&lt;span class="nf"&gt;split&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="dl"&gt;'&lt;/span&gt;&lt;span class="s1"&gt;/&lt;/span&gt;&lt;span class="dl"&gt;'&lt;/span&gt;&lt;span class="p"&gt;).&lt;/span&gt;&lt;span class="nf"&gt;pop&lt;/span&gt;&lt;span class="p"&gt;()}&lt;/span&gt;&lt;span class="s2"&gt;.md`&lt;/span&gt;&lt;span class="p"&gt;;&lt;/span&gt;
    &lt;span class="k"&gt;if &lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="nx"&gt;inbound&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;has&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="nx"&gt;target&lt;/span&gt;&lt;span class="p"&gt;))&lt;/span&gt; &lt;span class="nx"&gt;inbound&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;set&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="nx"&gt;target&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="nx"&gt;inbound&lt;/span&gt;&lt;span class="p"&gt;.&lt;/span&gt;&lt;span class="nf"&gt;get&lt;/span&gt;&lt;span class="p"&gt;(&lt;/span&gt;&lt;span class="nx"&gt;target&lt;/span&gt;&lt;span class="p"&gt;)&lt;/span&gt; &lt;span class="o"&gt;+&lt;/span&gt; &lt;span class="mi"&gt;1&lt;/span&gt;&lt;span class="p"&gt;);&lt;/span&gt;
  &lt;span class="p"&gt;}&lt;/span&gt;
&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Thirty lines, no dependencies, runs in under a second. It told me what I wanted to know: on a 20-page site, the bottom three pages had zero inbound links from other guides. Only the index pointed at them.&lt;/p&gt;

&lt;h2&gt;
  
  
  The index is not a link
&lt;/h2&gt;

&lt;p&gt;That is the part I got wrong for months. Being listed on &lt;code&gt;/guides&lt;/code&gt; counts as a link in the technical sense, but it is a link from a page that exists only to list links. Every page gets exactly one of those. It carries about as much weight as a sitemap entry.&lt;/p&gt;

&lt;p&gt;What actually moves a page is a contextual link from a page that is already getting traffic, with anchor text that describes the destination.&lt;/p&gt;

&lt;p&gt;So the rule I now apply, per page:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Every guide must have &lt;strong&gt;at least two inbound links from other guides&lt;/strong&gt;, not just the index.&lt;/li&gt;
&lt;li&gt;The linking paragraph must make sense if you deleted the link — a reference, not a "see also".&lt;/li&gt;
&lt;li&gt;Anchor text is the page's actual subject, not "click here" and not the brand name over and over.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  What the fix looks like
&lt;/h2&gt;

&lt;p&gt;Take a page like the one on &lt;a href="https://waterparksimulatorwiki.vercel.app" rel="noopener noreferrer"&gt;Waterpark Simulator Wiki&lt;/a&gt; that covers &lt;a href="https://waterparksimulatorwiki.vercel.app/guides/hidden-content-easter-eggs-guide" rel="noopener noreferrer"&gt;where the hidden content in Waterpark Simulator is&lt;/a&gt;. A long, specific page — exactly the kind that never gets linked, because no other guide &lt;em&gt;needs&lt;/em&gt; to mention it to make its own point.&lt;/p&gt;

&lt;p&gt;The fix is not to edit that page. It is to find two existing pages where the link is genuinely useful, and add it inside a sentence that would still make sense without it:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;a page about early-game progression, mentioning it as "the list of things you can find without advancing the plot";&lt;/li&gt;
&lt;li&gt;a page about map or area content, mentioning it where the reader would otherwise have to go looking.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Two links, two different anchors, both inside a real sentence. The destination page does not change at all — only who points at it.&lt;/p&gt;

&lt;h2&gt;
  
  
  What this does and does not do
&lt;/h2&gt;

&lt;p&gt;To be clear about expectations: this does not make a page rank by itself. Internal links redistribute whatever authority the site already has; on a young domain there is not much to redistribute. What it does do is make sure the pages you spent the most time on are not the ones nobody can reach.&lt;/p&gt;

&lt;p&gt;Two practical notes from doing this across a fleet:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Run it after every batch of new pages.&lt;/strong&gt; Orphans are created by growth, not by neglect. New pages link outward to old ones far more often than the reverse.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Do not automate the anchor text.&lt;/strong&gt; The script can tell you a page needs a link. Only you can write the sentence it belongs in — and a sentence that exists only to hold a link reads exactly like what it is.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;The whole diagnostic is one file and a &lt;code&gt;Map&lt;/code&gt;. The expensive part is writing the two paragraphs afterwards.&lt;/p&gt;

</description>
      <category>seo</category>
      <category>webdev</category>
      <category>javascript</category>
    </item>
    <item>
      <title>Running 48 Next.js sites from one laptop: what the tooling actually costs</title>
      <dc:creator>zhaoq</dc:creator>
      <pubDate>Tue, 27 Oct 2026 13:37:22 +0000</pubDate>
      <link>https://dev.to/zhaoq0103/running-48-nextjs-sites-from-one-laptop-what-the-tooling-actually-costs-3mkf</link>
      <guid>https://dev.to/zhaoq0103/running-48-nextjs-sites-from-one-laptop-what-the-tooling-actually-costs-3mkf</guid>
      <description>&lt;p&gt;I did not set out to run 48 websites. It was supposed to be one.&lt;/p&gt;

&lt;p&gt;A year ago I started a small wiki for an indie game — game guide pages, the kind of thing where "how do I beat this boss" gets 200 searches a month and nobody has written it down. That one site worked well enough that the obvious next move was a second, then a fifth, then a template and a script.&lt;/p&gt;

&lt;p&gt;At 48 sites the problems stopped being about writing pages and started being about tooling. Here is what it actually costs, in case you are tempted to do the same thing.&lt;/p&gt;

&lt;h2&gt;
  
  
  The stack is boring, on purpose
&lt;/h2&gt;

&lt;p&gt;Every site is the same Next.js app: static export, a &lt;code&gt;content/guides/*.md&lt;/code&gt; folder, and three config files (&lt;code&gt;lib/seo-config.ts&lt;/code&gt;, &lt;code&gt;lib/adsterra-zones.ts&lt;/code&gt;, &lt;code&gt;lib/site.ts&lt;/code&gt;). No database, no API routes, no server state.&lt;/p&gt;

&lt;p&gt;That is the whole point. When 48 sites share one codebase shape, a fix is written once and applied 48 times. When they diverge, maintenance becomes a full-time job. The number of times I have been glad a site had a custom layout is zero; the number of times a shared &lt;code&gt;components/&lt;/code&gt; folder saved me is every week.&lt;/p&gt;

&lt;h2&gt;
  
  
  pnpm is not a preference, it is disk space
&lt;/h2&gt;

&lt;p&gt;With npm, 48 sites with a shared dependency set came to &lt;strong&gt;18.7 GB&lt;/strong&gt; of &lt;code&gt;node_modules&lt;/code&gt;. Same versions, copied 48 times.&lt;/p&gt;

&lt;p&gt;Switching to pnpm's content-addressable store brought that to &lt;strong&gt;479 MB&lt;/strong&gt; — one real copy, hard-linked into every project. Same &lt;code&gt;node_modules&lt;/code&gt; layout from the app's point of view, 39× less disk.&lt;/p&gt;

&lt;p&gt;If you run more than a handful of similar projects on one machine, this alone is worth the migration.&lt;/p&gt;

&lt;h2&gt;
  
  
  The deploy quota is the real bottleneck
&lt;/h2&gt;

&lt;p&gt;Vercel's free tier allows 100 deployments per day. Sounds like plenty until a bulk change touches every site — renaming one metadata field across 48 projects is 48 deployments, and they do not roll up.&lt;/p&gt;

&lt;p&gt;So deployments have to be queued. Mine live in a plain text file, one site per line, and a script drains it:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;&lt;span class="c"&gt;# drain the queue, respecting today's remaining quota&lt;/span&gt;
scripts/deploy-pending.sh &lt;span class="nt"&gt;--limit&lt;/span&gt; 8
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Two rules came out of getting this wrong:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;The local deploy log is not the quota.&lt;/strong&gt; Dashboard deploys, git-integration deploys and anyone clicking "Redeploy" all count. My log said 65 used when the API said the limit was gone.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;A quota error is not a bug.&lt;/strong&gt; It exits zero, leaves the queue untouched, and the next cron run picks up where it stopped. Treating it as a failure just produces noise every day.&lt;/li&gt;
&lt;/ol&gt;

&lt;h2&gt;
  
  
  Static sites fail in DNS, not in code
&lt;/h2&gt;

&lt;p&gt;Both sites and the tooling that reads them run behind hostile DNS. &lt;code&gt;suggestqueries.google.com&lt;/code&gt; resolves to a corporate IP; &lt;code&gt;*.vercel.app&lt;/code&gt; resolves to different addresses on different resolvers.&lt;/p&gt;

&lt;p&gt;Every verification script therefore does the same two steps — resolve through a public resolver, then connect to that IP explicitly:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;&lt;span class="nv"&gt;ip&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;&lt;span class="si"&gt;$(&lt;/span&gt;dig @1.1.1.1 +short &lt;span class="s2"&gt;"&lt;/span&gt;&lt;span class="nv"&gt;$host&lt;/span&gt;&lt;span class="s2"&gt;"&lt;/span&gt; A | &lt;span class="nb"&gt;head&lt;/span&gt; &lt;span class="nt"&gt;-1&lt;/span&gt;&lt;span class="si"&gt;)&lt;/span&gt;
curl &lt;span class="nt"&gt;-s&lt;/span&gt; &lt;span class="nt"&gt;--resolve&lt;/span&gt; &lt;span class="s2"&gt;"&lt;/span&gt;&lt;span class="nv"&gt;$host&lt;/span&gt;&lt;span class="s2"&gt;:443:&lt;/span&gt;&lt;span class="nv"&gt;$ip&lt;/span&gt;&lt;span class="s2"&gt;"&lt;/span&gt; &lt;span class="s2"&gt;"https://&lt;/span&gt;&lt;span class="nv"&gt;$host&lt;/span&gt;&lt;span class="s2"&gt;/page"&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Without &lt;code&gt;--resolve&lt;/code&gt;, a poisoned answer turns into "the site is down" reports that are not true. With it, verification is boring and trustworthy. If you automate any kind of site checking, build this in from day one.&lt;/p&gt;

&lt;h2&gt;
  
  
  Gates, not vibes
&lt;/h2&gt;

&lt;p&gt;Every content change goes through three checks before it deploys:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Metadata length&lt;/strong&gt; — titles are validated to 40–60 characters, descriptions to 140–160. Sounds pedantic until you measure a batch and find a third of them truncated in search results.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Template residue&lt;/strong&gt; — a grep for placeholder text, TODO markers and copy-paste from the base template.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Live re-verification&lt;/strong&gt; — after deploy, the script fetches the page and checks the &lt;em&gt;rendered&lt;/em&gt; title, not the source file. Source correct and live correct are two different claims, and I have shipped the first without the second more than once.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The third one is the one people skip. The last batch I ran had 21 sites pass in source and one fail live for a reason that had nothing to do with the change — a transient fetch timeout that would have looked like a real failure if the script had not been re-run.&lt;/p&gt;

&lt;h2&gt;
  
  
  What I would tell someone starting
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;Make the first site genuinely good. The template is worthless if the content underneath is not worth reading.&lt;/li&gt;
&lt;li&gt;Automate the checks before you automate the content.&lt;/li&gt;
&lt;li&gt;Keep the deploy queue in a file, not in your head.&lt;/li&gt;
&lt;li&gt;Budget for the boring parts: DNS, quotas, metadata length, dead links.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;One of the sites, &lt;a href="https://nivalisnightswiki.vercel.app" rel="noopener noreferrer"&gt;Nivalis Nights Wiki&lt;/a&gt;, now gets a few thousand pageviews a month, and its &lt;a href="https://nivalisnightswiki.vercel.app/guides/beginner-guide" rel="noopener noreferrer"&gt;beginner guide&lt;/a&gt; is the single most-read page on it. Not a business yet, but the loop works: find a question nobody answered, answer it properly, keep the site from rotting.&lt;/p&gt;

&lt;p&gt;The tooling is the easy part once you accept that the interesting problems are all in the seams — between the build and the deploy, between the source file and the live page, between what the log says and what the API counts.&lt;/p&gt;

</description>
      <category>nextjs</category>
      <category>webdev</category>
      <category>devops</category>
      <category>seo</category>
    </item>
  </channel>
</rss>
