The bot I like is the ASCII Art bot. I’m always into computationally generated art and event though it doesn’t say how it is made, since an art piece is posted every 30 minutes, I’m assuming it was computationally generated. One thing this bot does extremely well is creating character and emotion. It’s not like these bots are just stick figures. They are overwhelmingly personified by the simplest of characters.
Although I really enjoy this bot, I’d like if it played with font or boldness or italicization or formatting more. I think it could stand to use the full range of text editing options.
Pentametron ( @pentametron) by Ranjit Bhatnagar is a twitter bot that searches twitter for posts that happen to be in iambic pentameter and retweets them. Iambic pentameter is a line of verse with five metrical feel, each consisting of one short or unstressed syllable followed by one long (or stressed). There is something wonderful and poetic about this, as each tweeter did not know (I am assuming) that their tweet was iambic pentameter. It is a simple acknowledgment of older ways of organizing language in a highly contemporary way of writing. Ranjit Bhatnagar works with interactive and sound installations, with scanner photography, and with internet-based collaborative art.
This Twitter-bot is a unique take on the conversation of gender equality, gender fluidity and gender identity that is taking over the political and social spectrum for last few years now. This bot generates and pairs random words to create unique, unheard of genders and pairs them with a symbol and a Braille translation. The result is formatted to look like a bathroom sign that you would hang in a corridor or bathroom door. The background colors, genders, symbols and braille are all generated randomly.
— Restroom Genderator (@RestroomGender) July 3, 2016
I personally love the completeness and absurdity of the project. In terms of completeness, I respect a project that goes to round out all the edges to an idea, so in this case he did not just generate genders, but a fully constructed sign with all the required parts to it. And the results look clean and good looking, all of them.
— Restroom Genderator (@RestroomGender) June 30, 2016
The absurdness comes from the sometimes non-existent link to human behavior or appearance (such as “minerals” in the gender name), the often un-relatable symbols to human form (ordinarily you see a generic man or woman stick figure, not an alien symbol), and the last of which is the fact that, in the end, this bot is sort of segregating genders even more. Instead of promoting a single bathroom for all to use, it is generating an infinite amount of segregated bathrooms, one for each gender. I believe whether you are for or against gender equality in bathrooms, Restroom Genderator is an enjoyable piece for all to enjoy and remark about.
I first considered writing about the catholic mom twitter bot at first because of its ability to incite and continue a heated argument without even having an actual sense of reason (I’d love to see an Atheist version), but I decided to talk about @thricedotted’s soft-spoken twitter bot: *.
In my art, I always try to achieve a certain kind of mood. That could vary depending on the art piece, but there always need to be some direction that I try to push people towards thinking about. @thricedotted’s twitter bot isn’t quite poetry, but it’s more than just random quips. The bot constructs these short sentences in a way that directs our imagination into a certain realm of thought.
Looking through A Brief History of the Future of Twitterbots, I am finding a some of these bots to be quite arbitrary, and mainly exist to show that it is possible for them to exist…if that makes sense. There are some that definitely may serve an important purpose, and I do like a couple of these-like the random sandwich recipe bot, that’s pretty handy. Followers would possibly accumulate based on a shared interest in collecting these random sandwich recipes, or to find inspiration for their packed lunches. Anything related to random food gets a thumbs up from me. But what I see from these is that in the way they might be random but relate to a specific topic, they kind of serve as a catalyst for creative thinking. For example, the art assignment bot would be a nice way for an individual to have a new exercise for the mind every day. Arbitrarily generated content may seem like a fun way to get “users” or viewers of the content to try new things, expand their mind.
I’m really moved by the yourevalued bot, or bots similar to those, in which the bot responds to an individual using the twitter interface and is depressed. To think that a twitter bot could literally save a life is amazing, and it shows how powerful something that could potentially post sandwiches or random colors and words might be. It is a tool that can literally save humanity!!!!
Bots like this^ as well as bots that can provide twitter users with valuable information are the bots that I personally find myself to be most inspired by. However, I think the idea of using bots to generate content purely for the imagination is interesting. (I mean who am I to say if anything can be ‘purely for the imagination’, people use what they make from that and bring it into the real world, can anything ever really be ‘purely for imagination’?), but I would be curious to see more examples of a twitterbot be used as a tool to create narratives, I also like the idea of having this twitterbot’s output function as something like a performance.
Overall, definitely seeing twitterbots to be a great way to connect with others, a way to really acknowledge our similarities or be there for one another despite differences. Using twitterbots, we can stand in solidarity.
Nora Reed makes a lot of pretty funny twitterbots that sit at the intersection of trolling and satire. Here are a few samples from different bots she’s made:
But the bot that takes the cake is Carol, a bot that tweets ridiculous conservative Christian-like phrases in an attempt to troll stuck up internet atheists.
And people actually took this bot for a human being, got angry with it, and started twitter fights. All the bot could do was tweet back generic responses, which made people even angrier because their points weren’t getting through to her.
As funny as it is, with groups of people like the Westboro Baptist Church and other pretty extreme organizations always spewing stuff like this unironically, Carol is pretty convincing, and as outrageous as the things she says are, the people that interact with her also come across as pretty pathetic. Fights with strangers over the internet can be funny, but no one wants to be the guy who’s actually a part of it.
A pair of bots I really liked were Thricedotted’s The Rhymin’ Riddler, and the Riddler’s Apprentice. The Rhymin’ Riddler posts riddles that always start off with “What do you call a…” and then the other bot the Riddler’s Apprentice will respond with an answer to that riddle that will rhyme. One thing I like about these bots is how funny they are. Often, the answers to the riddles make more sense than I expect them to, all while rhyming. I like that there are two bots that interact with each other. I think the idea of bot interaction is really nice, because even though the twitter accounts are just bots, they seem to take on more personality when there’s plausible interaction between the two, and the tweets seem mysteriously more intriguing automatically. It’s funny how you can almost see the relationship between the riddler and the apprentice.
For my bots investigation, I looked mainly at visual-computational bots that consume an image and reply with a manipulated version. My favourite one of these bots was @pixelsorter (see above results), but I also just Img Rays, Img Shredder, and IMG2ASCII. I particularly enjoy two aspects of these types of bots:
That you send a tweet and then know you will get a reply within a 30-90 seconds, in contrast to how Twitter is normally used (when you don’t know whether or not someone will reply), this is interesting.
But, you don’t know what it will be and the waiting period makes it such that it’s hard to brute force out an understanding of the algorithm.
Together, this created an interesting human-like interaction that a lot like talking to a real person on twitter. checking out Unfortunately, sometimes the bots didn’t reply, which was very depressing.
A few other fun things I came across while exploring:
Jason Ronallo‘s personal site has a nice cross section of bot work.
At first I looked at Every Color (@everycolorbot) and was astonished by the number of followers (90.7K!!!!!! that’s way more than Carnegie Mellon’s 39.6K followers @CarnegieMellon) it had for simply tweeting a few random colors throughout the day. Maybe I am not excited about colors as these followers are. Continuing to investigate, I found what I think is a much more interesting color bot: ColorSchemer @colorschemez
“I’m trying to find colors that go well together. I’m probably not very good at it because I’m a robot with no sense of style.”
Having just a small introduction in first person makes the bot more interesting and makes it seem like there’s more meaning behind these random color choices. Then the additional random adjectives that go with the colors produces some humorous content. Comparatively to the Every Color bot, I find this ColorSchemer bot more entertaining and something I’d subscribe to. The screenshots below are some of my favorite descriptions and/or color combinations. Avocado Sanstone…really? 😉
The art assignment bot generates random art assignments with requirements and due dates. When reading through them imagining what I would probably make for each of the assignments, I found that some of them are actually projects that could be potentially very meaningful.
Since the bot randomly mixes up very different subject, mediums and techniques so the results always sound weird and artsy. I find this kind of a satire of real world art assignments.
The concept is also particularly interesting. Usually, such as in this class, artists give assignments to computers to make art for them. However, the situation is reversed with art assignment bot: artists, who are usually considered the most free and creative people, are now getting assignments from a bot, which is not even capable of understanding art. It really makes me think about what art is and what a machine can do.
I can hardly believe these are actual quotes from the New York Times. They’re all gold tweets. I’m not sure if they’re literal quotes or markov generated text, but it seems like they’re quotes. I’m having a hard time thinking of any context this would fit in. The image of the New York Times and these tweets are so off that they’re pretty fascinating to read through.
I really liked magic realism bot, because It’s weirdly calming and makes an abstract sort of coherence. It makes a sentence or two that describes a scenario, a concept, or just an event. I think the profile picture of this account is what really allows it to excel because it places you in the past, instead of looking at the posts as current, you can see them as sort of abstract literature with a sort of confidence and sense that it has stood the test of time. The tweets are also thought-provoking, and really just paints a picture of a sort of abstraction of the world and they are all very satisfying in terms of content and coherence. This also sort of read as newspaper headlines, or clickbait articles with stature.
For example, one tweet reads “A man disguised as a camel is roaming the streets of Manchester. It keeps repeating one word: ‘Telephones.’ ” . This makes sense, like, it probably wouldn’t happen, but it is not unimaginable, and in fact, if that did happen it would either be funny or completely terrifying, but it’s interesting to consider the different implications of the texts and the different ways the events described could be happening. This works as a stand-alone story, but also provides the possibility for more.
This bot takes sentences out of the New York Times completely out of context. And it happens to be a riot. Plenty of people seem to agree – the bot has a loyal following of 131k!
This is just so incredibly fun. I wonder how the Bot was programmed to have such humor or whether the humor is largely self-prescribed knowing that it is a segmented taken from the NY times? Is it inherently funny? Or just funny because we think it’s from the NYT?
Some are just outrageous and I absolutely cannot imagine under what circumstance/context they would make sense under in the NYT:
Perhaps that is part of the allure? Wondering HOW ON EARTH DID THIS COME TO BE?!?!??!?!
I’m so curious – this bot could totally promote articles way better than any ad. I’d want to read this article if only to find out the context behind this outrageous statement.
A twitter bot that I enjoyed was reverseocr. The bot selects a random word and then draws until the ocr library recognizes it as the given word. This process occurs four times a day.
The algorithm used is intriguing to me because it’s based on the probability of the cursor drawing shapes that are similar to letters; the algorithm is based on randomness. I’m interested in learning more about generativity and applying it to my own work (particularly film). I find this algorithm to be a good starting point at developing new, randomly genereated aesthetics/filters that can be applied to the video form.
The bot * (@soft_focuses) is a “poetic experiment” whose tweets evoke an atmospheric and serene theme–I spent considerable time deciding between this and the poem.exe (@poem_exe), as both appealed to me with its introspective and calm sense of literary aesthetic. Plenty of the generated poetry are short, subtle, and perhaps mysterious; not always grammatically correct, the diction and brief eloquence of the vocabulary is enough to present some soft and gentle imagery (as its username suggests). The tweets are non-formulaic and almost cursory reminders to myself to ponder within myself, and to think of the outside world, in a sense of “it’s the little things in life…”–and to laugh at the occasional hilarious outputs. In a way, it’s lovely for such fragments to possibly lead to extensive, deep thoughts.
For this Looking Outwards assignment, I selected Citation Needed (@needsref) by Allison Parrish, which is a bot that crawls through Wikipedia and tweets out sentences that are marked “citation needed.” Seeing a feed full of assertions that we would normally believe in the context of a full-length Wikipedia article, but we actually have no way of actually checking their accuracy, highlights the willingness with which we will accept information. There’s also a certain humorousness to seeing sentences completely out of context, as well as a more subtle humor to seeing sentences that are probably very subjective and up for debate being attempted to be passed off as fact, such as “England is considered the home of the game of football.”
Parrish works primarily in fields of text, and seems to be especially interested in poetry , and words and code.
Reverse OCR is a bot which chooses a word and draws randomly until the OCRad.js library recognizes it as that word. This bot is charming to me because the relationship between the “drawer” and the software “judge.” They are both humorously incompetent. On one hand the bot can’t seem to get the letters right at all and the judge, perhaps out of pity, just accepts the chicken scratch. This bot was created by Darius Kazemi AKA Tiny Subversions (the artist most known for buying random stuff on amazon). Kazemi works in Portland at Feel Train, a creative technology cooperative he founded. Basically this guy just makes a bunch of bots for the lolz.
After looking through many Twitter bots, I really thought this security camera one was the most fascinating. This bot will post random stills from a list of unsecured webcams. According to the Twitter profile, all images are CC0 licensed, and the Quartz article I read explains that the bot will not post photos from particularly invasive cameras, or will try to crop out anything that will show too much information.
I think a signifying detail of good generative art is that it surprises even the owner. Even when the piece is complete, it stays unfinished.
The artist uses the invasiveness of security cameras as the constant, every photo feels like you’re spying on a space, hidden from view—a voyeur. Once you overcome the sense of intrusion, you can focus on the majesty of being able able to see the candid day to day of an enormous array of unknown locations. Looking through these photos brought me back to a story I read last spring from The New Yorker. At the time I felt uncomfortable when reading the story of The Voyeur’s Motel, but after exploring Arnold’s work, I think I can understand it a bit better now. There’s something transfixing about being the fly on the wall, looking out from where no one suspects there to be eyes. Arnold’s piece allows one to have that experience without the guilt. However, in The Voyeur’s Motel, you can begin to understand the adrenalin of seeing what you shouldn’t and nearly being caught.
[@FFD8FFDB is an] attempt to profit on the curiosities of casual voyeurs, piquing the curiosity of people that normally wouldn’t be a peeping tom.
In terms of creating @FFD8FFDB Arnold explains that he pulls the stills from a smaller list of his known camera’s, excluding any that are private. He will crop, colour adjust, and use Wordnik to caption the tweets with characters and graphics (just for effect).
This bot–rather, piece, social statement, expression, exploration, study– really struck me. It was the only one that really felt like the creator was trying to share something meaningful through social media and a programmed bot. I haven’t used Twitter in years but I think I may come back to this one every now and again, if not simply just to use the photos in projects.
Robert Yang – Intimate, Infinite
For my looking outwards I looked into the work Intimate, Infinite by Robert Yang. Intimate Infinite is a 3d first person game based on the novel “The Garden of Forking Paths” By Jorge Luis Borges. I read about the story after I played the game, and It made it more interesting because the narrative is abstract, so I wasn’t able to fully understand the work the first time I played it. It was interesting and aesthetically pleasing but the content came across as confusing. The story is about a Chinese professor of English who is a spy for the germans, and he is found out and murdered. I think that the game expects the player to be more knowledgeable about the source story, but that this was not made clear – Unless the game is intended to be abstracted, which it may be because it lends itself well to this through the lagged video feed, the different points of views, and the changing narrator, as well as the discussion of different lives and different timeslines
This set of deliverables has three components, due at the beginning of class on Friday October 28th.
Looking Outwards #06: Bots
Viewing & Prep (for the Book)
A Computationally Generated Book
Looking Outwards #06: Bots
A bot is software that automatically generates cultural content and publishes this content online, generally in a way that others can subscribe to.
In this, your sixth Looking Outwards assignment, you are asked to identify a bot (probably on Twitter, but potentially elsewhere) which interests you. Write a brief Looking Outwards post about it. In your post, please include screenshots of two or three postings from the bot that you think are particularly good.
As usual, please categorize your post with the LookingOutwards06 Category. This makes your posts much easier to find (and grade). Title your post nickname-lookingoutward06.
Write a program (or more likely: a set of programs) which generates a book. Your program(s) should generate the content of the book, as well as execute its layout automatically. Your book should have at least 20 pages, and must be printed and bound.
Learning Objectives
The purposes of this assignment are to prompt students toward:
Further application of generative principles: to generating text, images, layouts, and their comprehensive combination in a complex yet familiar physical object, a book.
Experience combining multiple self-written programs into a multi-stage workflow.
Exposure to a scripting language (Basil.js) for controlling a powerful commercial software application (Adobe InDesign) that would ordinarily be used ‘by hand’.
Exposure to a toolkit (RiTa) for language analysis and synthesis, and text analysis/synthesis concepts such as Markov chains and parts-of-speech taggers.
Awareness of generative text strategies, in the context of artists’ books.
Recommended Tools
It’s likely, though not necessarily guaranteed, that you will use Basil.js for this project. You might also find yourself using RiTa and Temboo.
Basil.js is a scripting library for Adobe InDesign, created by Ted Davis, Benedikt Groß and Ludwig Zeller et al. at the Basel School of Design in Switzerland.
RiTa is a set of software tools for computational literature (for Processing, Node, and p5.js) created by Daniel Howe.
Temboo is a set of libraries that allow you to access a very wide range of APIs, such as Twitter, and much, much more. They provide example code for 10 different languages (including JavaScript, Python, and Processing). Temboo is a powerful commercial product, but essentially free for students to tinker with.
Summary of Deliverables
This checklist won’t surprise you, but it’s nice to be reminded of what’s expected.
Consider books around you.
Develop a program (or more likely: a set of programs) to generate a book. See the section below, Details and Considerations, for more information.
Print and bind your book. Have this physical copy ready for the beginning of class on Friday, October 28.
Write a 100-word description of your book. Publish this description in a blog post on this web site. Title your blog post, nickname-book, and give your blog post the WordPress Category, Book.
Write an additional narrative of 150-200 words describing your development process, and evaluating your results. Please include this in your blog post, too. Include some information about your inspirations, if any.
Upload a photograph of your printed, bound book.
Upload a video or (preferably) an animated GIF of a video of you flipping through your book’s pages.
Upload a PDF of your book.
Upload some photos or scans of your notebook sketches.
Embed your code (using the WP-Syntax WordPress plugin to format your JavaScript and/or other code), and include a link to your code on Github. Be careful not to include your user keys or passwords to Temboo!!
Details and Considerations
The content is up to you. Take a look at books around you. How many kinds of books can you think of? Your book may have any combination of text, images, and/or graphics. Examples of possible books include, but are not limited to, things like:
A children’s alphabet book
A nonsense dictionary
A visual encyclopedia
An atlas, full of maps
A coloring book
A screenplay or script for a play
A phonebook for an imaginary location
A collection of letters between two lovers
A 365-day calendar with a horoscope-a-day
A “how-to” book, full of instructions or recipes
A field guide, describing the flowers/insects of a region
A collection of illustrated short stories, poems or sonnets
An illustrated catalogue depicting and describing objects for sale
A book of facts, representing a (daily) snapshot of data from around the world
Write a program (or more likely: a set of programs) to create a book. Your program(s) should generate the content of the book as well as its layout. Try and keep these two programs as cleanly separated as possible.
You will very likely need to write more than one program. For example:
You might write a first program in Python or JavaScript which generates some poems, and saves those poems in JSON format.
You might then write a second program in Processing to illustrate those poems, and save out these illustrations as PNG files.
You might then write a third program in JavaScript to lay out the book containing these texts and images in Adobe InDesign using the Basil.js library.
Note that it is almost certainly not advisable to attempt to do all of this (i.e. poem generation, illustration generation, book layout) using Basil.js! Bad idea. Instead, generate your content using one or more programs, and use Basil just for layout.
Working with Basil.js (SAMPLE CODE, YO)
Most of the complexity in working with Basil.js comes when trying to get your environment set up correctly. If you follow the instructions here VERY CAREFULLY, and I do mean very carefully, you should be alright. In addition to the many fun Basil.js demos that come with the download, I’ve prepared the following readymade example project for you (an illustrated Alphabet Book), which should help you get your own generative book started. It includes a template InDesign file (.indd), Basil.js code (.jsx), all necessary data (.json, .jpg), and a sample PDF output.
Our guest Marius Watz also has created a set of Basil examples for you, including a newly-revised example that lays out a book of Tweets with text frames that size themselves dynamically according to the length of your text. You can find his samples at:
Do some investigation. Check out tools like RiTa for generating rhymes. Check out tools like Temboo for accessing information from online sources, such as tweets or headlines.
Your book should have at least 20 pages. If there is a good reason why your book’s concept calls for fewer pages than this, please consult the professor for permission to make an exception.
Give consideration to the layout of your book. This includes things like your margins, typeface selection, etc., as well as your cover design. You may wish to read Chapter 8 (“Shaping the Page”, pp.143-178) of Robert Bringhurst’s classic Elements of Typographic Style.
Your book may not be laid out ‘by hand’. You are asked to generate your book by scripting Adobe InDesign using Basil.js. (Keep in mind that the units are likely to be points, or 1/72 of an inch.) However, if you are deeply screwed and Basil does not behave, you are permitted to generate a multi-page PDF in Processing. If you do this, you’ll probably have to sacrifice the quality of your typography.
Where Can I Print My Book?
Your book must be printed and bound. Here are five options:
Blurb.com (formats). Good online print-on-demand service. Requires 4-5 days for printing + shipping.
Lulu.com (formats). Good online print-on-demand service. Requires 4-5 days for printing + shipping.
The Espresso Book Machine at the University of Pittsburgh (4000 Fifth Avenue; options & FAQ). Able to do “perfect bound” books in arbitrary sizes (you can design the book dimensions), but only in black-and-white. Very fast. Open until 6pm most weeknights, or 8pm on Wednesdays. Read the requirements carefully! If you want to do this, please read their guidelines carefully!
Fedex/Kinko’s Office Print & Ship Center (3710 Forbes Avenue), open until 11:00pm on weeknights. Able to do “perfect bound” books, but only in a limited range of sizes. Could be expensive, but at least you’re not paying for shipping.
You may bind your book yourself if you’re experienced in bookbinding, and if you know what a bone folder is, and how to do a saddle-stitch. But, please: I will give you a withering look if you try to submit a wad of 8.5×11 sheets stapled at the corner.
A special printing offer. Using funds from our course budget, books completed by 12:00 Noon on Saturday October 22 will be printed and express-shipped from Blurb.com and/or Lulu.com. Books must cost less than $50 to print, not including shipping, and must be uploaded to Blurb or Lulu by Saturday 10/22, without exception. Students who miss the 10/22 deadline must pay for the printing of their own book.
