Sep 27, 2019 · newsletter

Other Orders: Re-contextualizing the Familiar

My Twitter feed sorted by semantic similarity to “crudely understood Marxism” using Other Orders.

To illustrate how categories shape our thoughts, the philosopher Michel Foucault cites Jorge Luis Borges’ (made-up) quotation from a “certain Chinese encyclopedia," where it is supposedly written that:

Animals are divided into: (a) belonging to the Emperor, (b) embalmed, (c) tame, (d) suckling pigs, (e) sirens, (f) fabulous, (g) stray dogs, (h) included in the present classification, (i) frenzied, (j) innumerable, (k) drawn with a very fine camelhair brush, (l) et cetera, (m) having just broken the water pitcher, (n) that from a long way off look like flies.

I thought of the Borges quote while reading through the list of Tweet sorting mechanisms in “Other Orders” - a project by Sam Lavigne:

Alphabetical; Alphabetical by Username; Antisemitism as It Is Understood by the Right; Apocalyptic; Approximate Quantity of Shame Expressed; Chronological; Cop-Like; Crudely Understood Marxism; Density of Adjectives; Density of Nouns; Density of People, Places, Brands, Monetary Values and Dates; Density of Verbs; Eroticism as an Approximation of Similarity to a Phrase by Anaïs Nin; Exclamatory; Gothness; Kafkaesque-ness; Length of Text; Neoliberalism as Determined by Proximity to Famous Neoliberals; Number of Numbers; Percentage of Words Which Are Filler Words; Quantity of Gendered Words; Questioning; Similarity to @dril; Similarity to Values Expressed in TED Talks; Total All-Time Posts from User; Total Emoji; Total Favorites; Total Hashtags; Total Retweets; Use of Language Similar to Language Used by Corporate Social Media Accounts Such as Amazon.

Both projects use alternate orders to make us think more critically about the ones we take for granted (the defaults). Lavigne wants to push back against recommendation systems that privilege engagement (views, likes, retweets, replies). Such systems can end up promoting content based on how controversial it is, regardless of its accuracy. Even the reverse-chronological sort, a default that feels ‘natural’ for our social media feeds, is a specific ordering choice with consequences. It means newness is privileged, and discourages making connections between Tweets across time.

On sorting

We sort by what is quantifiable. Engagement and chronology can be determined and sorted through metadata that is already numbers (add together views, likes and retweets for engagement; compare time for chronology). Part of the fun of the Other Orders list is how it moves between easily countable metadata, more advanced counting of types of words (nouns, verbs, numbers, emoji, filler words) and word similarity as calculated through word embeddings. It reminds me of building our prototypes, where you explore the options a new technology can open: “Ok, we can identify nouns; let’s see what it looks like if we sort by density of nouns… hmmm…. Ok, interesting. What else can we do?"

Sorting by “similarity to the values expressed in TED talks” got me a top tweet of “[sparkle emoji] machine learning [sparkle emoji]".

My favorite categories are the “similarity” ones, including: Similarity to Values Expressed in TED Talks, Similarity to @dril, and Kafkaesque-ness. All of these categories involve looking at similarity between the tweets and an example text using word embeddings. Each of these categories contains a recognizable style. One drawn from literature (Kafka), one from weird Twitter (@dril), and one from a particular genre of speaking (TED talks).

On the loss of context and making connections

One of the difficulties of fitting techniques like word embeddings into a product is the model’s lack of a sense of context. It can get you a similar sentence, but it may pluck that sentence out of a paragraph on a different topic. This makes using it for things like summarization a difficult (but not insurmountable) challenge (see our summarization prototype Brief). The cleverness of the similarity groupings in Other Orders is that they turn this lack of context to their advantage, sorting Tweets in a way that enables you to spot connections between people and topics that you would not otherwise have recognized. Looking at the top Kafka and @dril-like Tweets in my feed reminds me of looking at the rabbit-duck illusion: shift my head one direction and I can read the Tweet like a non-sequitur from @dril, shift it the other and I can figure out that in its original context it was a fragment from a larger thread.

Just knowing that Tweets could be sorted in these ways shifts my experience of regular Twitter. Having been exposed to the lens of “is this Tweet @dril-like?", I can start spotting @dril-like Tweets on my own. The @dril stuff is mostly for fun, but I can also see these alternate categories having a more substantial effect on how I understand my feed. Sorting by “Crudely Understood Marxism” (“Items are ordered relative to their semantic similarity to the phrase ‘workers must seize the means of production.'") in my feed shows mostly web developers, writers, and game developers (the main groups I follow) talking about structural frustrations with their work. I read and understood these Tweets before, but since seeing the pattern exposed through the ordering, I now recognize them as part of a larger thread. “Oh, there’s another Tweet about burnout or the need to unionize," I now think to myself, connecting it to that new category in my head.

Try it out

I recommend trying Other Orders for yourself. If you don’t have Twitter or don’t want to connect yours, play with sorting the sentences of works of literature on the about page; it will give you a good feel for how these sorting mechanisms can re-contextualize the familiar.

Read more

Sep 27, 2019 · newsletter
Sep 27, 2019 · newsletter

Latest posts

Nov 15, 2022 · newsletter

CFFL November Newsletter

November 2022 Perhaps November conjures thoughts of holiday feasts and festivities, but for us, it’s the perfect time to chew the fat about machine learning! Make room on your plate for a peek behind the scenes into our current research on harnessing synthetic image generation to improve classification tasks. And, as usual, we reflect on our favorite reads of the month. New Research! In the first half of this year, we focused on natural language processing with our Text Style Transfer blog series. more
Nov 14, 2022 · post

Implementing CycleGAN

by Michael Gallaspy · Introduction This post documents the first part of a research effort to quantify the impact of synthetic data augmentation in training a deep learning model for detecting manufacturing defects on steel surfaces. We chose to generate synthetic data using CycleGAN,1 an architecture involving several networks that jointly learn a mapping between two image domains from unpaired examples (I’ll elaborate below). Research from recent years has demonstrated improvement on tasks like defect detection2 and image segmentation3 by augmenting real image data sets with synthetic data, since deep learning algorithms require massive amounts of data, and data collection can easily become a bottleneck. more
Oct 20, 2022 · newsletter

CFFL October Newsletter

October 2022 We’ve got another action-packed newsletter for October! Highlights this month include the re-release of a classic CFFL research report, an example-heavy tutorial on Dask for distributed ML, and our picks for the best reads of the month. Open Data Science Conference Cloudera Fast Forward Labs will be at ODSC West near San Fransisco on November 1st-3rd, 2022! If you’ll be in the Bay Area, don’t miss Andrew and Melanie who will be presenting our recent research on Neutralizing Subjectivity Bias with HuggingFace Transformers. more
Sep 21, 2022 · newsletter

CFFL September Newsletter

September 2022 Welcome to the September edition of the Cloudera Fast Forward Labs newsletter. This month we’re talking about ethics and we have all kinds of goodies to share including the final installment of our Text Style Transfer series and a couple of offerings from our newest research engineer. Throw in some choice must-reads and an ASR demo, and you’ve got yourself an action-packed newsletter! New Research! Ethical Considerations When Designing an NLG System In the final post of our blog series on Text Style Transfer, we discuss some ethical considerations when working with natural language generation systems, and describe the design of our prototype application: Exploring Intelligent Writing Assistance. more
Sep 8, 2022 · post

Thought experiment: Human-centric machine learning for comic book creation

by Michael Gallaspy · This post has a companion piece: Ethics Sheet for AI-assisted Comic Book Art Generation I want to make a comic book. Actually, I want to make tools for making comic books. See, the problem is, I can’t draw too good. I mean, I’m working on it. Check out these self portraits drawn 6 months apart: Left: “Sad Face”. February 2022. Right: “Eyyyy”. August 2022. But I have a long way to go until my illustrations would be considered professional quality, notwithstanding the time it would take me to develop the many other skills needed for making comic books. more
Aug 18, 2022 · newsletter

CFFL August Newsletter

August 2022 Welcome to the August edition of the Cloudera Fast Forward Labs newsletter. This month we’re thrilled to introduce a new member of the FFL team, share TWO new applied machine learning prototypes we’ve built, and, as always, offer up some intriguing reads. New Research Engineer! If you’re a regular reader of our newsletter, you likely noticed that we’ve been searching for new research engineers to join the Cloudera Fast Forward Labs team. more

Popular posts

Oct 30, 2019 · newsletter
Exciting Applications of Graph Neural Networks
Nov 14, 2018 · post
Federated learning: distributed machine learning with data locality and privacy
Apr 10, 2018 · post
PyTorch for Recommenders 101
Oct 4, 2017 · post
First Look: Using Three.js for 2D Data Visualization
Aug 22, 2016 · whitepaper
Under the Hood of the Variational Autoencoder (in Prose and Code)
Feb 24, 2016 · post
"Hello world" in Keras (or, Scikit-learn versus Keras)


In-depth guides to specific machine learning capabilities


Machine learning prototypes and interactive notebooks

ASR with Whisper

Explore the capabilities of OpenAI's Whisper for automatic speech recognition by creating your own voice recordings!


A usable library for question answering on large datasets.

Explain BERT for Question Answering Models

Tensorflow 2.0 notebook to explain and visualize a HuggingFace BERT for Question Answering model.

NLP for Question Answering

Ongoing posts and code documenting the process of building a question answering model.

Cloudera Fast Forward Labs

Making the recently possible useful.

Cloudera Fast Forward Labs is an applied machine learning research group. Our mission is to empower enterprise data science practitioners to apply emergent academic research to production machine learning use cases in practical and socially responsible ways, while also driving innovation through the Cloudera ecosystem. Our team brings thoughtful, creative, and diverse perspectives to deeply researched work. In this way, we strive to help organizations make the most of their ML investment as well as educate and inspire the broader machine learning and data science community.

Cloudera   Blog   Twitter

©2022 Cloudera, Inc. All rights reserved.