What Topics and Social Platforms Does Mining the Social Web Cover?

Mining the Social Web is a technical manual about extracting and analyzing data from social platforms, and its coverage centers on Gmail and Twitter as the platforms named in its published material, with Python as the working language. It is best understood as a book-and-blog project rather than a hosted tool: the site exists to support the manual and to publish follow-up posts as platform access rules change. If you want a single source that teaches social data mining techniques and also documents how those techniques break when APIs tighten, this is the intended audience.

The two platforms with concrete published examples

The site's own headings point to specific, named workflows rather than a generic platform list.

Both are self-directed, local workflows. The Gmail example is notable because it routes around API access entirely — you take your own data out through Google Takeout and store it in a database you control. That pattern matters if you're trying to analyze your own messages without depending on a platform's developer program.

Python as the assumed toolchain

The project's origin story is explicit about language: the third edition overhaul was framed as modernizing the code to Python 3 and re-testing it. The author came into the project from an astrophysics and machine learning background, and the work was scoped around Python code that readers could run and adapt.

For a reader, that means the techniques are presented as runnable code rather than abstract methodology. You should expect to install libraries, handle credentials or exports, and debug against live or exported data.

The theme that shapes everything: access is the hard part

The most useful thing to understand about this project's coverage is that it was rewritten because platform access changed. The site describes the period after the Cambridge Analytica scandal and major data breaches, when "all the major platforms went on the defensive":

• APIs were changed
• Access to data was severely curtailed
• Certain privileges began requiring approval from the platform's developers

The author notes plainly that the book "was full of examples designed to teach data mining techniques," and that "suddenly a lot of the code no lo[nger worked]." This is why the site frames itself as "Rebooting Mining the Social Web for a Rapidly Changing World."

Practical consequence: if you follow older editions or older blog posts, expect authentication and endpoint steps to be the first thing that fails. The durable content is the analysis technique; the fragile content is the data acquisition step.

What this means for you as a reader

Choose this material if:

• You want to learn social data mining by writing Python, not by using a dashboard
• You're interested in analyzing your own Gmail via Takeout plus MongoDB, which sidesteps third-party API approval
• You want a case study in how social platforms restrict data access, and how to design around it

Look elsewhere if:

• You need a currently maintained, guaranteed-working API client for a specific platform — the site itself documents that this is the moving target
• You want a no-code or hosted analytics product; nothing in the published material suggests one

A concrete way to start

A reasonable first task, matching the site's own example, is to export a slice of your own Gmail through Google Takeout, load the resulting archive into MongoDB, and run a query that answers a question you actually have — for example, which senders you correspond with most over a given year. That exercise teaches the full pipeline (export, ingest, index, query) without touching a restricted API, and it's the same shape of problem the Twitter material addresses with live data.