A bot is an automated application, usually on social media, which performs a specific, repetitive task. Doing a cursory search on the Internet provided me with some important context for Mark Sample and Steven Lubar’s blog posts. There are a lot of different bots out there, many of which give bots a bad reputation. They can be used for spamming purposes, in automated network attacks, political campaigns, or to post comments designed to deliberately inflame users. But they can also be used for more benign purposes: to order pizza from Dominos, to answer questions on commercial websites, and to find discounts on eBay.
Sample and Lubar’s posts explore two types of benign twitter bots. For those of you who are twitter-illiterate (like me), twitter bots are automated twitter accounts that can perform simple actions like tweeting, retweeting, and messaging.
Museumbots
In Museumbots: An Appreciation, Lubar shares his thoughts on the value of museumbots. These bots randomly select and post objects from a museum’s collection several times a day. While seeing cool historical objects on your news feed is interesting, it is not what makes these bots so illuminating. It is what these objects say about choice.
The seemingly randomness of the objects shown by the bots are actually more representative of the museum’s collection than the objects displayed for public consumption. Museums make choices that influence which objects in the collection the public sees. They have to consider what is appropriate to the public or dealers; what the curator and conservator expect; what is in the budget; what fits in the space. The public also engages in choice. They decide which exhibits to visit based a map of the museum; what is advertised; what catches their attention.
“[I]t does an excellent job of making clear the differences between what’s at the museum, and what I see on display.”
It is the randomness of the bots that reveal choice, something that we take for granted. These museumbots disclose SOME of those choices. But wouldn’t it be cool if a bot existed that could reveal the choices curators make when they purchase new collections? Or reveal items that have been removed from the museum?
Protest Bots
Mark Sample explores another type of bot: the protest bot or “a computer program that reveals the injustice and inequality of the world and imagines alternatives.”
What makes a twitterbot a “bot of conviction” with a message “so specific you can’t mistake it for bullshit”? Sample lists the five characteristics shared by all bots of conviction.
- Topical – they speak to recent news stories and current events.
- Data-based – they rely on reliable research, statistics, etc.
- Cumulative – their true message is revealed in the aggregate.
- Oppositional – they take a stand.
- Uncanny – they reveal the hidden or hide the obvious.
What a Bot of Conviction is Not: @TwoHeadlinesand @TheHigherDead
@TwoHeadlines is topical and data-driven, but not cumulative or opposition. There is no theme inherent in the accumulated tweets and they do not reflect or take a stance on the news itself.
@TheHigherDead is oppositional, uncanny, and topical, but is not data-driven as it doesn’t use actual news of ed-tech blogs.
What does a bot of conviction look like in practice?
@ClearCongress is a great example of a protest bot. It retweets members of Congress and redacts parts of their tweet based on their current congressional approval rating. It’s topical, gathers information from polling data and congressional accounts, and hides what should be visible. When seen together, the tweets reveal the disconnect between Congress and their constituents and the uselessness of Congress in general.
@congressedits, which has recently been suspended, and @NSA_PRISMbot are two other examples.
Sample himself created “a bot of consolation and conviction” called @NRA_Tallyin response to a 2014 shooting near the UC-Santa Barbara campus. It creates headlines for imagined mass shootings, followed by a fictional NRA response.
Each tweet contains a number between 4 and 35, victims drawn from historical record, a location based on historical sites of mass shootings, a type of firearm that has been used in mass shootings in the US, and a response from the NRA which mimics rhetorical statements made after mass shootings.
@NRA_Tally is an example of what Sample calls tactical media “that engages in a ‘micropolitics of disruption, intervention, and education.’” It strips the NRA’s of their main tactic for shutting down the gun control debate, namely accusing those who talk about it of politicizing the victim’s deaths. In this case, there are no real victims. It momentarily unsettles the gun control debate and instead focuses on the weapon itself.
For @NRA_Tally and protest bots in general, it is the persistence of the bot that makes it powerful. “[T]his is a bot that doesn’t back down and cannot cower and will tweet for as long as I let it.”
Both Lubar and Sample demonstrate two ways that bots can be used to draw attention to certain issues: the way we collect and exhibit historical objects and they way we use can use bots in social protests. What do you think are the benefits of turning over our interpretations and exhibitions to machines? What are the downsides?
Thank you, Erica! I really enjoyed these readings. While the persistence and imperviousness of bots may be productive, I think if people realized that bots were behind it all, they might not find it as persuasive or personal. For instance, I get emails that make it sound like the editor is writing to me personally in attempts to invite me to either pay attention to an issue or to join an activist cause. Yet, I think over time people who receive such messages become desensitized. Having said that, I also think that in many ways, bots like journalism are ultimately created by people and for people and have political undertones. Both bots and journalism are ways to convey information, and so, it really depends how it is done and perceived.
I also resent Mark Sample’s romanticization of so-called “journalism of conviction” of the nineteenth century. It ignores how censored many of those newspapers have been, the history of it being used for propaganda purposes, not to mention who was able to read and buy the newspapers.
While I was reading this, I was thinking about Twitter’s policies, especially given the use of Russian, or other foreign countries’ usage of bots to spam accounts and influence perceptions. I fear that any potential changes to policies to prevent the use of third party platforms, including automation, would also affect Museums, or other accounts, that use bots in a helpful way. Looking into Twitter’s policy some more, I found that the main “ground-rules” that would be applicable here are “Build solutions that automatically broadcast helpful information in Tweets” and “Don’t spam or bother users, or otherwise send them unsolicited messages.” Both of these seem fairly vague, but hopefully, they don’t affect the bots doing the great work that you described!
Neither of the articles really discuss twitter policies in depth but I think it’s an important dynamic to consider. One bot, “Drop the ‘I’ Bot,” retweets messages that use the term “illegal immigrant” and replies “People aren’t illegal. Try saying “undocumented immigrant” or “unauthorized immigrant” instead. This bot was created by two journalists who wanted to draw attention to people’s use of the word. The bots received attention in the press from outlets such as BBC news, who applauded the bots’ message that actions can be illegal but not human beings. However, Twitter suspended the account in 2015. This brings up an important question about twitter regulations. To what degree do they silence voices of protest?
I’m really intrigued by Lubar’s discussion of how institutions can use digital collections to reveal parts of the history making process that are often invisible. I wonder how social media can be harnessed to achieve this goal, both in terms of museumbots and not. Haley mentioned in her comment on Sara’s blog post “Absences and Opportunities” the idea of institutions sharing authority and it made me curious how institutions can share authority to reveal the collecting process. Maybe bringing to the public conversations between museums about acquiring pieces (buying/selling, bidding, loaning) or the search for objects that may or may not exist or the negotiation over using one object over another in an exhibit? Doing work like this takes away a lot of the mysticism and sparkle of institutions (which many may not like), but it also is important to the public’s understanding of how history is made in terms of what is displayed, but also what exists in archives.