Since “OK Boomer” emerged as a mainstream thing during fall 2019, I have been tempted to engage with it. I talked about it with friends, and got progressively convinced that a stupid idea could actually be fun, and interesting. Finally, during the Christmas holidays, I put aside my PhD work and made the OK-Boometer website, with the help of some friends and notably OuestWare.
On OK-Boometer you can look at the tweets and profiles who get the most ok-boomed. You can also look at tweets getting ok-boomed live; it looks like they receive an avocado. You can also search a username to see if it had been ok-boomed. You may find it amusing. But beyond the fun, there are a few interesting things I would like to expose.
OK boomer is multiple things, but I narrow it down to one practice: when a Twitter user replies to another tweet with “OK boomer” and that’s it. Are people actually doing that? I was unsure, but my experiment showed that indeed, it is something they do.
As a practice it has multiple facets, that I just observed by spending some time with the website. Some tweets receive one or two OK boomer (e.g. private jokes) and some attract a lot. Those getting ok-boomed are of different kinds.
Some ok-boomed tweets are the direct expression of “ideals of past generations” or whatever qualifies as boomer-speech – think offensive old white male ideological positions. Example:
Other ok-boomed tweets are reporting the boomer-speech of someone else. In this case the ok-boomed Twitter account (e.g. a media) is not the actual target. For instance in the tweet below, it is Biden who gets ok-boomed, not The Hill.
If those two kinds qualify as finger-pointing, the third kind does not. Those tweets are often self-aware, and there OK boomer looks like a social game, or a running joke. It is as varied as cultural practices can be. See for instance the tweet below – and if you want the full story, look at Neil deGrasse Tyson’s record on OK-Boometer.
So obviously, the ok-booming on Twitter is a cultural practice. OK-Boometer establishes that. Is it interesting? In itself, probably not. But OK-Boometer does more than establishing facts. It acts on the practice – or at least it may act.
Bottom-up innovations were more common during the early years of the web, following the ideals of the hippy culture where it emerged. People’s practices could become embodied in technology. For instance, the tweet below is the actual point of origin of the hashtag (story here); it was a practice first, and later became implemented in the Twitter infrastructure.
I see the practice of ok-booming as a possible form of finger-pointing; as a specific, narrow angle on Twitter content that may capture something unique. But for that to happen, it needs to be embodied in technology.
Embodying a practice in technology
A key inspiration for OK-Boometer was the N-word bot. It is a bot operating on a different platform, Reddit. You can call it on someone in a public thread. It comes and exposes publicly how many times that person has employed the “N-word” (“nigger” or other variants) on Reddit. Read this for an explanation, find an example there.
The N-word bot embodies a certain form of finger pointing. In terms of data harvesting, it is quite trivial. But in terms of social engineering, it is brilliant – kudos to its author ogkushmonster. Redditors have been using it a lot. The typical use case is to reveal the ideology of other Redditors to delegitimize them. It is particularly efficient when the comment itself is not technically racist (or homophobic, or…), or ambiguous to some degree, but still supports that kind of thinking. For instance this comment:
In this case, by revealing that the author has used the “N-word” many times, it tells the story that this person is a racist in disguise that just got unmasked. That, of course, is not necessarily true. But it is what the bot makes possible, a specific form of finger pointing.
In the current state of OK-Boometer, you can do something similar, but manually. You can search for a specific Twitter user and get to know if he or she was ok-boomed, and for which tweets. Note that if the tweet has been deleted, it will appear as a blank space (we only store the tweet ids, not their content).
For instance, has the author of Harry Potter, J. K. Rowling, been ok-boomed? Yes, 406 times so far, and only for this tweet:
Has the French president, Emmanuel Macron, been ok-boomed? Not much, just 29 times, and for 3 different tweets. Good to know.
The most ok-boomed French personality is Raphaël Enthoven, receiving 621 OK boomer, for 16 different tweets. The most ok-boomed of his tweets (490) is the following:
By aggregating information in a specific way, OK-Boometer has the potential to transform a practice into something else. I don’t know what exactly, and I do not know if it will be good of bad. Let’s see and act in consequence when the time comes – if it comes, as there is a real risk that nobody uses it and it remains a useless experiment.
Of course we could push the logic further, by having an actual Twitter bot for instance. There are other possibilities. I do not have much time, but I’d like to do that with friends, because it’s more fun and productive that way. It’s all about experimenting, and it’s not at all about money – no ads, no cookies, the data can be downloaded (warning: heavy file), and there is even a GitHub repository with the source code.
Re-dramatizing the data
The aesthetic dimension of OK-Boometer is primordial: the design, but also the color, the avocado decorum… All of that plays a role, although it might not be obvious. It provides an angle to the data, if situates them in a certain way, it dramatizes them. But of course, the data were already situated, already dramatized, just in a different way. Digital traces are not in the clouds, their infrastructure is not unsituated, and APIs are major pieces of dramatization.
I did not invent the data. I did some trivial queries on Twitter, and you can basically do the same. The important act was to take these data and re-dramatize them differently. The data themselves are unimportant, it is their re-dramatization that matters, because it allows repurposing them. The data were already there and did nothing for you, because they were in a dormant state – the effect of their previous dramatization. The new dramatization puts them at work in a certain way.
I did not have much time – it was a side project after all – so I did not push the system as far as possible. I tried to find a sweet spot, a “minimal viable product” as they say in the industry. Here I want to share my rationale for the different elements, so that you have an idea of my process.
The central feature is the ability to search, because it is where the finger pointing becomes possible, and it was my starting point. This requires a search feature. But if you search a list, it means that this list preexists. It must be exposed as well, so that the random user can click on something and explore the system. That is why the home page is a list of the most ok-boomed Twitter users, with a search field.
What should happen when one clicks on a Twitter profile? One already knows how much OK boomer that profile received, but one might want to see the tweets in question. But in terms of dramatization, it is not enough. The list does not reward you for clicking on a profile, it does not put the “boomer” at the center of the attention. That is why I implemented this huge, animated, fake LCD screen that rewards you for getting there. You can take a screenshot, and the OK Boomer score is loud and clear. This screen is the meter in “OK Boometer”. The expected list of tweets (and their score) is just below.
To understand which contents get ok-boomed, it is necessary to browse a list of tweets, so I added it.
There is nothing obvious about what “getting ok-boomed” means. But the ok-booming process could have diverted attention from the contents that get ok-boomed, which was my focus. So to clarify this angle, I drew inspiration from Rotten Tomatoes, the famous independent movie rating website, where the metaphor is tomatoes thrown at films – their score system is called the tomatometer. I thought this analogy might work, but with an avocado instead of a tomato. To make that analogy clearer, the live page on OK-Boometer features an animation where an avocado is thrown at the ok-boomed tweet. The avocado throwing represents the ok-booming.
I do not show who throws the ok-boomings, despite having the information. This is intentional. I retained it to not move the focus out of the contents that get ok-boomed. I did not want to glorify it as an individual practice, as I am more interested in its collective dimension.
Since I chose to have the avocado-throwing as the metaphor of ok-booming, I had to enforce it everywhere. That is how the crushed-avocado became the graphical element representing ok-booming everywhere.
To enforce the cultural angle of my dramatization, I wanted a graphical element representing both sides of the social interaction: who is throwing the OK boomer, and who is receiving it. There are no universal symbols for baby boomers. But from my perspective, this story is largely about the millennials. Indeed, the most consistently ok-boomed tweets are those expressing normative views on gender identity, negative views on technology, and climato-negationism (criticizing Greta Thunberg checks the Boomer bingo). Those are part of the millennial identity. I liked that angle, so I searched for a symbol of millennials, and I landed on the avocado (or vice-versa). It might be quite unclear to many, I admit. As for the symbol of baby boomers, Trump was a safe bet – and the OK-Boometer numbers are definitively crowning him as the ultimate boomer.
From the avocado, the general green tone was quite obvious, but there is more to it. I emphasized the dramatization as a generational conflict by using graphical elements that evoke past technology. I picked these old, monochrome LCD screens because they were green like the avocado. In the beginning I wanted a big meter like a gauge or something similar, but I realized that the LCD might as well represent that, and I went for it.
At the end of the day, I hope this experiment will lead to something unexpected. I could very well analyze this data with Tableau and tell you what there is to know about the phenomenon. I will probably do it at some point, if no one else does. That would be engaging with it from the outside. But I wanted to engage with it from the inside. I wanted to dramatize the data differently to foster other people’s imagination. And not only for the Twitter users, not even for them first. Before all, I want to shake the imagination of scholars – of whoever identifies as such.