From d97433279b83f921955bf0418aae92a75caf56aa Mon Sep 17 00:00:00 2001
From: Sage Vaillancourt <sagev9000@tutanota.com>
Date: Thu, 8 Sep 2022 00:30:56 -0400
Subject: [PATCH] Add more to what-is-kafka.md

A few appearance tweaks. Mostly for headings.
---
 src/app.css                         |  1 -
 src/routes/+page.svelte             |  6 +--
 src/routes/blog/[post]/+page.svelte |  1 -
 src/routes/blog/what-is-kafka.md    | 62 ++++++++++++++++++++++++++++-
 4 files changed, 64 insertions(+), 6 deletions(-)

diff --git a/src/app.css b/src/app.css
index 9cca01d..f49645c 100644
--- a/src/app.css
+++ b/src/app.css
@@ -184,7 +184,6 @@ p {
 
 h3 {
 	font-family: 'Montserrat', sans-serif;
-	margin-top: 0;
 }
 
 p {
diff --git a/src/routes/+page.svelte b/src/routes/+page.svelte
index cf97639..7ec6b8e 100644
--- a/src/routes/+page.svelte
+++ b/src/routes/+page.svelte
@@ -121,8 +121,8 @@ JSON.parse(`{
 </div>
 
 <style>
-    h1 {
-        padding-top: 1em;
+    h3 {
+        margin-top: 0;
     }
 
     .standalone-data-display pre {
@@ -199,7 +199,7 @@ JSON.parse(`{
     }
 
     .eye-catch {
-        margin-bottom: 1em;
+        margin: 1em;
     }
 
     .eye-catch h2 {
diff --git a/src/routes/blog/[post]/+page.svelte b/src/routes/blog/[post]/+page.svelte
index 7410c39..7ca9829 100644
--- a/src/routes/blog/[post]/+page.svelte
+++ b/src/routes/blog/[post]/+page.svelte
@@ -19,7 +19,6 @@
         max-width: 90vw;
         display: flex;
         flex-direction: column;
-        align-items: center;
         flex-wrap: wrap;
     }
 
diff --git a/src/routes/blog/what-is-kafka.md b/src/routes/blog/what-is-kafka.md
index d27ff57..8937113 100644
--- a/src/routes/blog/what-is-kafka.md
+++ b/src/routes/blog/what-is-kafka.md
@@ -9,6 +9,66 @@ headerImageAlt: "Kafka's logo"
 Initially built by LinkedIn, Apache Kafka is a "distributed event store" and "stream processing" platform.
 But what does that mean exactly?
 
+## Basics of Kafka
+
+The core principles of Kafka are very simple. There are three key components: producers, consumers, and the kafka cluster.
+
+### Producers
+
+Producers are responsible for creating new messages for your Kafka system. Every time an event occurs that the system
+should know about, a producer will generate a message describing this event.
+
+For example, a bookstore may have an application that _produces_ a message every time someone buys a new book:
+
+```json
+{
+  "action": "purchase",
+  "title": "The Metamorphosis",
+  "price": 9.99,
+  "buyerId": "84694fc7",
+  "storeNum": "FL-231",
+}
+```
+
+The rest of the system decides what exactly to _do_ with this new information. Those parts of the system are called
+**Consumers**.
+
+### Consumers
+
+Whenever a message is produced in the Kafka system, a consumer can act in response to that new information.
+
+Returning to the previous example: this particular bookstore may have a service that informs authors and publishers
+every time one of their books has been purchased. This service may work as a consumer. Then, every time a book is
+purchased, the service _consumes_ the new message, and alerts the creators however it sees fit.
+
+One of the main benefits of this structure is an inherent loose coupling. Producers do not care how consumers work. If a
+consumer wants to send a text, rent a billboard, or order an ice cream, the producer simply is not interested. This
+allows each part of the system to focus on its own piece of the domain, each staying smaller and more self-contained.
+
+### Kafka Cluster
+
+The Kafka cluster, in short is what facilitates the connection between producers and consumers. An indefinite number of
+producers and consumers can connect to the cluster, and the cluster itself can be powered by many servers.
+
+The cluster decides exactly which messages need to go where, and a key part of this process is its use of **Topics**.
+
+#### Topics
+
+"Topics" are Kafka's way of dividing up messages into different categories. A simple topic for our bookstore might be
+`book_purchases`, or `coffee_orders` (for the requisite café inside). So, when a book is sold, our producer would push a
+new message into the `book_purchases` topic. And of course, when someone orders a fancy latte or something hot and
+bitter, a message is instead put into the `coffee_orders` topic.
+
+Importantly, consumers do not blindly receive every message produced in the system. Instead, they subscribe to
+individual topics. Our author-notification service might subscribe just to the `book_purchases` topic, not being
+very interested in messages from `coffee_orders`.
+
+Thus, every time the Kafka cluster receives a newly-produced message with a given topic, it only forwards it along to
+consumers that have intentionally subscribed.
+
+## TODO: 
 ## Distributed Event Store
 
-## Stream Processing
\ No newline at end of file
+Essentially
+
+## Stream Processing