Where in the web is thefarrelly.com?

I've been slowly adding to this website over time. Recently I implemented analytics, and configured a weekly analytics email (related post) that I receive at 5pm every Friday. It includes the overall amount of traffic my website experiences, along with a breakdown per page.

It's been great getting to understand use of this website. Analytics are an interesting paradigm unto themselves. I've begun understanding the pages which are frequented more than others, how many pages the average user traverses, and which page users normally enter the website through.

It's only made me wonder more about this website. Nowadays I can paint a picture of this website via the relation users have with it's pages. However, that only describes this website. There's an entire web out there. How does this website relate to the rest of it? Where can you find this website referenced? Which pages are referenced?

To find my website referenced across the web would require a web crawler. A bot going from web page to web page, collecting links and references.

That's a fun project itself, but how could I then understand my website in relation to others? How could I visualise it? Naturally a web crawler would create a JSON interpretation of the data. But, JSON isn't a very compelling storyteller. Nor does it piece together a story particularly well.

Graph nodes are a great tool to visualise relationships between vast swathes of information like this. Since every node in the network can link to any other node, there's an implicit or even explicit relation between nodes. A visualisation like this reveals the hidden relationships each node has to others in the network.

I thought about creating a tool which could create this representation, but thankfully vasturiano on Github had already created a 3D Graph node representation library, react-force-graph. Now we're cooking with gas. This library unlocks an answer to a question I hadn't yet thought of, what shape does my website take?

This 1.0 launch of graph-web ties all of this together. In order to create the graph node representation I had to make a web crawler. Then with the trawled data, I can show off its representation and answer the question, what shape does my website take? Currently it's a funnily branched tree with one branch showing off many more leaf (or end) nodes than the other. Which makes sense, I have over 30 posts, but only 3 projects.

If you're interested in seeing it's shape, you can see it live at projects/web-graph. You're presented with two options, watch each node being added to the network, or skip to the end and see the shape this website takes.

My future plans are to answer the unresolved questions above:

• How does this website relate to the rest of the internet?
• Where can you find this website referenced?
• Which web pages are being referenced?