Finding insight in my Social Media Comments
I mentioned in a previous post that I use LinkedIn as a publishing medium. Ever since big tech took over specialized communities, it feels like no one really reads personal blogs anymore, and I am sometimes forced to go where the audience is.
On LinkedIn, there are a few different contexts in which I write.
- Long Form Writing: These are typically articles, for which I do a fair bit of research and go through many rounds of editing. Producing long form articles of this sort takes me a few days at least.
- Short Form Writing: These are Short pieces, typically around 500 words at a 3000 characters limit. This also involves some research and editing, and it takes me between 30 mins to an hour to write one of these.
- Comments on Other People’s Posts: These range from a single sentence to a paragraph or two. It takes only a few minutes to compose, because, I’m building on someone else’s thinking. This 3rd type is what this post is about, and I’ll get to it in a bit.
There are other types of content – such as LinkedIn’s new expert contribution section, in which you are asked to “contribute to a topic, as an expert“. I largely ignore these because I don’t think this is a community resource and suspect that LinkedIn is using free labor to fine-tune A.I. models.
Why my comments are valuable to me
Unlike Long and Short form writing, the theme and context of comments is set by the owner of the original Post (The O.P). This means that whatever my comment is, it builds on top of, and extends, someone else’s ideas. This inevitably means that I am prompted to think in new ways and different contexts than I’m generally prone to, and as a result my own comments surprise me sometimes.
With Long and Short form writing, I keep a local record of what I have posted. I store them as text files, with a date and a link to the actual post on LinkedIn. Comments on the other hand are harder to store because I can never predict what I’m going to comment on and when.
I’ve thought about this before. It really would be nice to have a searchable database of my Linkedin comments, available locally. Last night I figured out how, and it was no work at all.
Creating a local, searchable database of my LinkedIn comments
It turned out that almost no effort was required on my end. LinkedIn has a surprisingly nice user data export system. Almost all the data that you have generated within LinkedIn is presented to you as CSV files. (I could not find an export of my Posts. It seems this is not offered as an export, so good thing I have my own local copies of these from 2019 onwards.)
Among all those CSV files is one called comments.csv. This file is supposed to contain every comment you have ever made on Linkedin. The comments file offers just 3 fields. A date stamp, a link to the Original Post and my comment. This means to see the context of my comment, I have to visit the link, which I guess is fair. The original post is not my property, only my comments are.
The CSV format is nice because you can open it in your spreadsheet software and run a search.
I took it a step further. I converted the comments.csv file into both a Treeline Database format and an Org mode file. This lets me do additional things such as linking comments that are similar or adding additional metadata and notes to specific comments if I wanted to.
The Good, Bad and Ugly in my comments
Now I can find useful patterns in my comments. For instance, over the past 15 years or so, I have left 3347 comments. Some are pithy, just a single word. Others are full-blown sermons. I can now use full-text search and it works beautifully. I can easily search for a variety of subjects that I’m interested in and have commented on.
There are also plenty of comments that make me cringe when I read them today. I found comments that run the gamut of bitter, vindictive, uninformed and plain ignorant. Some are mystifying, and I don’t know what I was thinking when I wrote them. On the other hand, I also found comments that seem to have been written during periods of lucidity – these are actually informed and useful. I also found interesting links that I had posted as comments, which I’ve completely forgotten about (15 years is a long time.) as well as people I engaged with heavily in the past but have lost touch with.
This database is not going to see heavy use or anything, but it’s nice to know that it is on hand if I ever need it.
I’ve now created a recurring task to download my comments file every 6 months or so, to keep my LinkedIn Comments Database up to date. All in all, it’s nice to add yet another dimension to my personal information infrastructure.