diff --git a/content/posts/shape-of-knowledge/death-note-l.jpg b/content/posts/shape-of-knowledge/death-note-l.jpg
new file mode 100644
index 0000000..3a17ee9
Binary files /dev/null and b/content/posts/shape-of-knowledge/death-note-l.jpg differ
diff --git a/content/posts/shape-of-knowledge/index.md b/content/posts/shape-of-knowledge/index.md
new file mode 100644
index 0000000..2efbccd
--- /dev/null
+++ b/content/posts/shape-of-knowledge/index.md
@@ -0,0 +1,44 @@
+---
+title: "The shape of knowledge"
+description: ""
+date: 2024-12-11
+tags: []
+---
+
+Recently, I've felt a bit lost about where to go in my career. I pride myself as a generalist and love to explore new domains and I felt like web/backend didn't have anything new to offer.
+
+I've been wondering how `specialists` found their domain and what types of profiles exist in the wild.
+
+> We value “T-shaped” people. That is, people who are both generalists (highly skilled at a broad set of valuable things—the top of the T) and also experts (among the best in their field within a narrow discipline—the vertical leg of the T.)  _ Valve's handbook
+
+When first hearing this, it felt logical, and then I wondered: do people really just pick a single domain to become an expert in? Shouldn't we value V shaped people more?
+Those questions began to sit in the back of my mind for a while and this blog is my answer.
+
+![](death-note-l.jpg "Painting of an L-shaped person???")
+
+My first observation was that no domain lives in isolation from the others. To give a simple example using webdev, no backend exists without its frontend. No one can be a backend expert without knowing how your API is going to be used and what could be its pain points.
+
+{{< alert "note" >}}
+
+If you feel hurt by this sentence, I hope it can serve as motivation for you to learn a bit about a new domain. fullstack is not a mediocre back & a mediocre front, it is a necessary path to become a better webdev.
+
+{{< /alert >}}
+
+The second observation I made is obvious: you do not know the depth of a domain without being an expert in it. You can use the Dunning Kruger graph to illustrate.
+
+Domains can be hidden behind another unknown to you. For example, you wouldn't know about [NaN boxing or tagged pointers](https://piotrduperas.com/posts/nan-boxing) if you only do [webdev](https://x.com/mitchellh/status/1599117365622767616).
+
+![](niho2-skill-tree.jpg)
+
+Seizing new knowledge (aka learning new stuff) is tiring and takes time. Learning everything at once is impossible. Knowing when to learn a new domain and when to focus on the current ones is an important skill.
+
+Learning is a permanent dance between seizing new knowledge and settling on existing ones.
+As long as you don't settle in every domain, you are still exploring the knowledge map. 
+
+![](map-hxh.png)
+
+I still don't know how specialists define their subjects (do they just toss a coin and never change subject??), but I'll continue to explore new domains while working on my current projects. I recently moved to the DevOps/platform team of my company and I plan on learning computer graphics in 2025.
+
+I don't have much to say, keep exploring the knowledge map ; I'm sure there's an island or two that are unexplored.
+
+<!-- vim: set wrap: -->
diff --git a/content/posts/shape-of-knowledge/map-hxh.png b/content/posts/shape-of-knowledge/map-hxh.png
new file mode 100644
index 0000000..e183ece
Binary files /dev/null and b/content/posts/shape-of-knowledge/map-hxh.png differ
diff --git a/content/posts/shape-of-knowledge/niho2-skill-tree.jpg b/content/posts/shape-of-knowledge/niho2-skill-tree.jpg
new file mode 100644
index 0000000..89d3a46
Binary files /dev/null and b/content/posts/shape-of-knowledge/niho2-skill-tree.jpg differ