AlgolitAlgolitAlgolit

Data Workers

Mundaneum, Mons, March 2019
https://www.algolit.net/index.php/Data_Workers

Data Workers was an exhibition at the Mundaneum from 28th March till 28th April 2019, coproduced by Constant. It showed algoliterary works, stories told from an ’algorithmic storyteller point of view’. Companies create artificial intelligences to serve, entertain, record and know about humans. The work of these machinic entities is usually hidden behind interfaces and patents. In the exhibition, algorithmic storytellers leave their invisible underworld to become interlocutors.
The exhibition was a creation by members of Algolit. Some works were by students of Arts² and external participants to the workshop on machine learning and text organised by Algolit in October 2018 in Mundaneum. With the support of La Communauté française/Arts Numériques and Arts².

Algoliterator

Muntpunt, Brussels, Public Domain Day, May 2018
https://algolit.net/index.php/Algoliterator

The Algoliterator is trained using a neural network and the full works of Belgian author Felix Timmermans, who entered the public domain in 2018. The Algoliterator helps you to write a text in the style of Timmermans. You can choose a start sentence from his oeuvre, you can also choose whether you want the Algoliterator to produce the following sentences based on primitive training, intermediate training or final training. The machine proposes a paragraph that you can edit. If you are happy with the result, you can send it to Zora, the house robot of Muntpunt. She will read out the text for you.
Sources: https://gitlab.constantvzw.org/algolit/algoliterator.clone

Algoliterary Encounter

La Maison du Livre, Brussels, Nov 2017
https://algolit.net/index.php/Algoliterary_Encounters

In the framework of Saison Numérique the Maison du Livre opens its space for Algolit during three days in a row. The group presents lectures, workshops and a small exhibition about the narrative perspective of machine learning models. These selflearning algorithms are based on algebra and statistics. They often function as opaque ’blackbox’ algorithms, while they shape applications that are daily used on a worldwide scale, like search engines on the web, translation machines, advertising profiling, facial recognition etc.
Because machine learning models are so present, the members of Algolit felt the need to distillate reading and writing experiments from it. By executing parts of their construction process in a literary context, they become more legible. It is way to experience a few moments that are usually hidden in the making of the model, and that co-design contemporary stories in the way they influence the organisation of information.

CharRNN

Algolit @ HS-la Senne/ de Zenne, Brussels, Nov 2017
Learning to write with Jules Verne, Shakespeare and Felix Timmermans, the CharRNN text generator is a machine that produces new texts using a CharRNN model. The generator is trained on works by Jules Verne, Shakespeare and Timmermans. You can explore the different stages of the learning process through the installation in the space. Selected fragments based on the work of the three authors will be read by Fred Bernier, Gijs de Heij and An Mertens.
The model is based on a script by Justin Johnson. This script is an improved version of the original script by Andrej Karpathy.

On Journey with Hovelbot

Constant_V, Brussels, Oct 2016
https://constantvzw.org/site/On-Journey-with-Hovelbot,2661.html

New landscapes are in the making by how we interact with our telephones and computers. Two hundred years after Mary Shelley wrote her novel Frankenstein, an Algolit Extended group reinterpreted part of the novel using contemporary artificial intelligence, ’On Journey with Hovelbot’ in the window of Constant shows one of the chat-bots made for this publication in the window of Constant.

Frankenstein Revisited

Mad Scientist Festival, Bern, Sept 2016
https://www.algolit.net/frankenstein/

Artificial intelligence can be as complex as it can be simple. With a small group of literary Python lovers we organised a workshop in the festival Mad Scientist in Bern. 200 years after Mary Shelly’s publication of ’Frankenstein; or, the Modern Prometheus’, we invited writers, artists, designers, programmers, and all people interested in literary chatbots based on or inspired by this gothic novel for three days of editing, writing and rewriting.

Markov Chain

Transmediale, Berlin, Jan 2015

List of all past Algolit meetings with link to the notes

https://algolit.net/index.php/Agenda#Past_Meetings

Algologs

Varia, Rotterdam, March 2018
http://varia.zone/en/algologs.html
Algologs = a 1 + 1 day dialog with algorithmic practices. This two-day event is an extention of Algolit, a workgroup where text-based practices and algorithms meet. Algologs is part of a series of occurances meant to open up the usually Brussels-based Algolit meetings by inviting external speakers, presenters and participants to join the conversation. During Algologs, we will increase the % of engagement with everyday algorithms.

A Story

Witte de With, Rotterdam, March 2016
https://algolit.constantvzw.org/index.php/A_story
https://www.fkawdw.nl/en/our_program/events/contested_tongues_presented_by_class_of_16
Manetta Berends, Gijs de Heij & An Mertens presented some algoliterary experiments.

Cqrrelations

Constant worksession, DeBuren, Brussels, Jan 2015
https://www.cqrrelations.constantvzw.org/1x0/

’Cqrrelations’ (pronounce crummylations, querylations…) was a Constant worksession in Brussels, inspired by the work of Algolit. The worksession invited data travellers, writers, numbergeeks, programmers, artists, mathematicians, storytellers, and other tech creative souls to explore the world of digital non-relations, desanalysis, blurry categorisations and crummylations in the Big Data that shapes our daily reality and language. They experimented amongst other things with databases on potatoes, and text analysis tools developed by researchers from CLiPS at the University of Antwerp. The session used datasets built from large corpora of natural language to build systems that can accurately guess the gender and age of writers and even find probable lies in written text.

Close-Reading Kenneth Goldsmith’s Uncreative writing, Brussels, June 2015

[en]As part of the residency of artists Billy Bultheel & Enad Marouf in Brussels, Constant/Algolit co-organised a close-reading session of Kenneth Goldsmith’s Uncreative Writing on 5, 6, 7th June 2015. The idea was to create a space for presentation, exchange, discussion on new ways of reading and writing.
Read a little more on this book: http://www.brainpickings.org/2013/02/13/uncreative-writing-kenneth-goldsmith/
A logbook was made with the collective notes of the session. Each participant received it as a printed and binded copy.[nl]Als deel van de Brusselse residentie van kunstenaars Billy Bultheel & Enad Marouf, co-organiseerde Constant op 5, 6 en 7 juni 2015 een close-reading van Kenneth Goldsmith’s Uncreative Writing. Met het boek als leidraad creërden we een ruimte voor presentaties, uitwisseling en discussie over nieuwe manieren van lezen en schrijven.
Meer over dit boek: http://www.brainpickings.org/2013/02/13/uncreative-writing-kenneth-goldsmith/
We maakten een logboek met de collectieve notities van de sessie. Elke deelnemer ontving het als een gedrukt en gebonden exemplaar. [fr]Dans le cadre d’une résidence des artistes Billy Bultheel & Enad Marouf à Bruxelles, Constant co-organisait une session de lecture et de travail autour du livre de Kenneth Goldsmith’s Uncreative Writing. On a créé un espace pour des présentations, des échanges et des discussion autour de nouvelles pratiques de lecture et d’écriture.
Plus d’information sur le livre : http://www.brainpickings.org/2013/02/13/uncreative-writing-kenneth-goldsmith/
Un journal de bord a été créé à partir des notes collectives de la session. Chaque participant l’a reçu sous forme d’un exemplaire imprimé et relié.

Désert Numérique #5, La Drôme (FR), July 2014

http://www.paramoulipist.be/algolit/Radio_algolit_1.wav
Catherine, Brendan, Nicolas and An created this first performative lecture in the framework of Désert Numérique #5. It was recorded and broadcasted on Radio Mulhouse MNE, Radio Panik in Brussels, Reboot FM in Berlin and Radio Libertaire in Paris.

https://next.liberation.fr/culture/2014/07/08/desert-numerique-festival-fertile_1059175
Algolit at Désert Numérique #5, by Marie Lechner
‘Ce jeu de traduction est également pratiqué par le collectif Algolit, cinq artistes et programmeurs disséminés en Europe, qui explorent les correspondances entre la littérature et le code informatique. Leur lecture performance à la salle des fêtes s’est faite sous l’égide de Nicolas Bourbaki, prête-nom d’un groupe mythique de mathématiciens français, qui a inspiré les écrivains de l’Oulipo. Lors de cette séance, elle-même très oulipienne, ils se sont appliqués à décortiquer le fonctionnement de l’algorithme de Markov, utilisé entre autres dans l’écriture automatique des spams. Cet algorithme doit son nom à un mathématicien russe passionné par les statistiques, qui a conçu un modèle de probabilités (la chaîne de markov) à partir non pas de chiffres, mais des vers d’un chef-d’œuvre de Pouchkine Eugène Onéguine. Algolit détourne ainsi les règles de l’algorithme à des fins de création littéraire, et reproduit le procédé sans ordinateur, via un jeu à base de Post-it et de lancers de dés. Une manière de détricoter ces technologies toujours plus opaques qui nous modèlent à notre insu.’

Chercher le texte

Paris, October 2013
Program:http://chercherletexte.org/fr/performance/textnfx/
Report by An:https://www.paramoulipist.be/?p=1243

Constant Variable, Brussels, July 2013

Spam poetry using Neel Doff, a residency with Olivier Heinry & An Mertens

Residency 2

Nantes, 27-29th May 2013
Looking into the notion of ‘algorithm’
Catherine Lenoble, Nicolas Malevé, Olivier Heinry, An Mertens

Masterclass FIGURA: Literature and Media Innovations - the question of genre transformations

Musée Royal de Mariemont, Feb 2013
Program:https://pad.constantvzw.org/p/Algolit_masterclass_figura_20130215
Notes:https://pad.constantvzw.org/p/Algolit_masterclass_figura_130215_notes

Residency 1

Variable, Brussels, Oct 2012
Exploring the space between media art and literature
Catherine Lenoble, Nicolas Malevé, Olivier Heinry, An Mertens, Stéphanie Vilayphiou

In October 2012 Catherine and An arranged a first residency of 4 days in Brussels, together with the poetic programmers Nicolas Malevé and Olivier Heinry. Stéphanie Villayphiou was the guest on the first day. They exchanged practises, looked into existing works, talked extendedly about Oulipo. In the end the name Algolit was born, with a mailinglist to continue the stream of ideas.

Notes:https://pad.constantvzw.org/p/Algolit_121010
Report by An:https://www.adashboard.org/?p=917
Report by Olivier:http://www.heinry.fr/olivier/index.php/category/Humeurs
Report by Catherine:http://litteraturing.wordpress.com/2012/10/25/litteraturing-in-brussels-part-one/
https://litteraturing.wordpress.com/2012/10/25/litteraturing-in-brussels-part-two/

2 Python scripts were created:
by Nicolas:https://www.algolit.net/scripts/residency1_121010/pyilit.py
[en]by Catherine/An: [nl]door Catherine/An: [fr]bpar Catherine/An: https://www.algolit.net/scripts/residency1_121010/startNLTK.py

LABtoLAB

From 2009 till 2011 LABtoLAB was a network of media labs set up by 4 European organizations that are active in the field of new media : Constant (Brussels, BE), Kitchen Budapest (Budapest, HU), Medialab-Prado (Madrid, ES), PiNG (Nantes, FR).
It aims was to create a platform for sharing experiences and ways of doing, exploring the role of the lab in offering spaces for collaborative learning and knowledge exchange, studying specific cases and examining the possibilities of life-long learning in the context of the lab. From 2009 to 2011, meetings, workshops and seminars took place in Budapest, Madrid, Brussels and Nantes. Each encounter opened up to a wider community of labs and individuals, and extended beyond European borders. This website documents the process and its outcomes.
Catherine Lenoble, initiator of the project, was interested in the practise of An Mertens, as both are writers in the classical sense of the word, and also experimenting in the media arts world. Catherine and An promised each other that they would get together on a quiet moment to discuss their personal work.
http://www.labtolab.org/

Verbindingen/Jonctions 9

In 2005 Constant organised the meeting days Verbindingen/Jonctions around the Language of Sharing.
An Mertens was invited to ‘creatively’ translate the interventions of the guests. In this image you see her translating Oulipo-member Jacques Jouet. It was An’s first encounter with Oulipo. The collective was introduced from their algorithmic point of view: potential literature based on conditions that could be lines of code. An was fascinated. This encounter resonated with her for years.

http://constantvzw.org/w/?u=http://archive.constantvzw.org/events/vj9/

2012

http://datapride.wordpress.com/
“J’y ai rencontré plusieurs universitaires spécialistes en TAL, c’est à dire traitement automatisé de la langue, et que les manipulations
auxquelles nous nous sommes livrés relèvent de la génétique textuelle.” Olivier Heinry

http://www.cnrtl.fr/
"Créé en 2005 par le CNRS, le CNRTL fédère au sein d’un portail unique, un ensemble de ressources linguistiques informatisées et d’outils de traitement de la langue. Le CNRTL intègre le recensement, la documentation (métadonnées), la normalisation, l’archivage, l’enrichissement et la diffusion des ressources. La pérennité du service et des données est garantie par l’adossement à l’UMR ATILF (CNRS – Nancy Université), le soutien du CNRS ainsi que l’intégration dans le projet d’infrastructure européenne CLARIN.

http://gutenbergzuckerberg.blogspot.com/
“à propos, voici un expérience poétique assez particulière que j’ai faite il y a peu, ça a fait pas mal jaser sur ce réseau en question.” Robin Decourcy

http://bookotron.com/agony/news/2011/12-26-11-podcast.htm#podcast122611
Interview with Jonatham Lethem on his ‘collage’ book The Extasy of Influence.

https://boingboing.net/2012/11/29/the-many-stages-of-writing-a-r.html
Timelapse view of writing a research paper

http://www.internetactu.net/2012/11/27/le-pudibond-internet-qui-controlera-les-algorithmes
“La prolifération de la fonction de saisie semi-automatique sur les sites web les plus populaires en est un autre exemple (“Les résultats qui s’autocomplètent dans les moteurs de recherche soulignent le caractère privé des conversations que les gens croient avoir leurs ordinateurs” rappelait récemment Sean Gourley du moteur de recherche Quid dans un article du New York Times). En théorie, tout ce que fait la saisie automatique est de compléter votre requête avant que vous ayez fini de la taper en utilisant un algorithme pour prédire ce que vous allez inscrire. …
“Morozow suggère une sorte d’audit comme pour les banques car il n’y a aucun moyen de forcer ces compagnies à ouvrir leurs algorithmes. C’est leur core business. Mais même une telle mesure on en est loin.” Nicolas Malevé

2013

https://longstreet.typepad.com/thesciencebookstore/2012/10/750-daguerreotype-portrait-mosaic-1840-1860.html
“Nicolas, j’ai pensé à toi et ta code poetry à la vue de cette page de daguerreotypes”

http://africaunchained.blogspot.com/
AfroSF | African SciFi anthology: via Boing BoingAfroSF is the first ever anthology of Science Fiction by Africa

http://p.xuv.be/2013-marques-pages
Un spécial pour Catherine:

An, bon anniversaire!
nicolas@peripheral:~$ sudo apt-get install an
nicolas@peripheral:~$ man an

http://blog.fperez.org/2012/01/ipython-notebook-historical.html
“vous vous souvenez sans doute en octobre dernier de la console python
ainsi que d’une autre console un brin plus avancée baptisée iPython?
Via la liste OKFN, j’ai appris que cette console abrite depuis fin 2011
un notebook permettant d’intégrer formules, images vidéos etc dans un
journal ensuite lisible par n’importe quel navigateur.
Tous les détails sont dans cet article.”

http://sens-public.org/
Labo et la revue sens public à Lyon. Un des boulots récents sur le site est le recueil “Pures Données” (http://sens-public.org/spip.php?article964) consacrés aux rapports entre écriture et compilation de données, une collection de textes sont présentés ayant pour objet commun d’investir la forme de la liste en résonance avec les pratiques computationnelles.

http://w3.erss.univ-tlse2.fr/textes/pagespersos/tanguy/HDR/HDR.pdf
Complexification des données et des techniques en linguistique : contributions du traitement automatique des langues aux solutions et aux problèmes, Ludovic Tanguy, 13 décembre 2012
Ce mémoire traite entre autres des différentes approches du traitement informatique du langage, ainsi que de ses impasses. C’est plutot rédigé à la première personne, pas du tout dans un ton scientifique désincarné, et il y a pleins de choses à piocher on dirait.

http://vimeo.com/43759584
Conférence ‘L’auteur numérique’ : avec Pierre Ménard, Catherine Lenoble, Cécile Portier.

http://en.wikipedia.org/wiki/Babel-17
The language is a code, it is not based on a code. It explores this idea of
http://en.wikipedia.org/wiki/Linguistic_relativity

https://www.chronicle.com/article/its-not-plagiarism-in-the-digital-age-its-repurposing/
by Kenneth Goldsmith, créateur d’ubuweb, entre autres

https://carinekrecke.wixsite.com/kreckesisters
Les deux soeurs jumelles Carine & Elisabeth Kreché. Leur
présentation était raffraîchaissante, puisqu’elles étaient les premières
et les seules (si je ne me trompe pas) à expérimenter avec la
fiction/les récits d’un nouveau terrain: Google Street View. Elles
considèrent le mapping par photo comme un grand archive de photos
‘machinales’ prises par une machine, tout comme les images de
caméras de
surveillance.
Elles se posent les questions suivantes:

• quel est le statut de ces images produites par une machine, sans
  auteur & sans regard?
• jusque quel point racontent-elles la vérité?
• comment peut-on/doit-on interagir avec cette collection immense, voire
  universelle, elle veut reporduire le monde, qui appartient à des
  monopoles d’exploitation?
  Elles proposent de pousser les frontières de l’art vers cette enquête
  critique et prendre position. On le fait par:
• toucher à cet espace ‘universel’
• y constituer un itinéraire qui introduit une subjectivité, le récit ->
  on regarde le monde à travers l’écran, grâce à cet outil on peut visiter
  des lieux ‘dangereux’

https://carinekrecke.wixsite.com/kreckesisters/dakotagate
Le point de départ de cette création était une photo d’un homme avec un
kalashnikov devant un supermarché, photo qui a été enlevée après par
Google; elles sont allées sur place pour découvrir les images absentes
et ont fait un récit de voyage sur ce qu’elles y ont vécu, notamment un
grand conflit local avec le sindiens sioux…

https://carinekrecke.wixsite.com/kreckesisters/404-not-found
Oeuvre composée de captures d’écran de Jerez (Mexique), la ville qui est
célèbre pour ses narcotraficants et les violences contre les femmes
Elles sont allées voir qu’on trouve des traces de cette violence sur
Google Street View; Google Street View devient ‘porte-parole’ par cette
perspective
Elles décrivent de façon ‘neutre’ ce qu’elles voient, comme le genre du
roman documentaire; elles deviennent les excécutrices littéraires du
GSTreetView

Très chouette tout cela, sauf qu’il n’y avait pas de prise de position
particulière sur:

• comment ‘regarder des images sans contexte’: dans Recorded Futures
  elles montrent une photo d’une grande tâche rouge sur le sol entre deux
  rues, qui pourrait induire l’idée qu’il s’agit du sang, mais qui
  pourrait très bien aussi être une tâche chimique. L’interprétation est
  au spectateurs… elles ne font que montrer
• idem pour leurs descriptions ‘neutres’, soi-disant
  Cette attitude qui se réflète dans le ‘récit’ opaque de leurs
  publications artistiques m’a énérvée un peu, sinon le propos est
  chouette: écrire des romans policiers avec Google Street View ;-)

http://www.codecademy.com
Python course for beginners

https://www.liberation.fr/ecrans/1997/06/06/queneau-interdit-de-web-le-fils-de-l-ecrivain-gagne-un-proces-pour-contrefacon-en-ligne_207696
concernant l’affaire “Queneau interdit de Web”
Décision de justice complète ici :
http://www.legalis.net/spip.php?page=jurisprudence-decision&id_article=108

On aimerait étiqueter le texte suivant
http://www.legalis.net/spip.php?page=jurisprudence-decision&id_article=108
pour remplacer chaque substantif par le 7ème substantif qui suit (un texte S+7 récursif pour reprendre la terminologie de l’Oulipo).
Tu connais un set de Part of speech francophone à installer dans nltk?

Thomas Girault:
Si tu cherches un Part of Speech Tagger pour le français, il y en a plusieurs :
TreeTagger : http://www.cis.uni-muenchen.de/~schmid/tools/TreeTagger/
J’ai pas testé mais il semble que qq’un a écrit un wrapper compatible NLTK
https://github.com/miotto/treetagger-python/blob/master/treetagger.py
J’ai aussi entendu parler du Stanford Part-Of-Speech Tagger qui semble
s’interfacer avec NLTK.
http://nlp.stanford.edu/software/tagger.shtml
https://code.google.com/p/nltk/source/browse/trunk/nltk/nltk/tag/stanford.py?r=8712
http://stackoverflow.com/questions/9663918/how-can-i-tag-and-chunk-french-text-using-nltk-and-python
Tu devrais sans doute pouvoir étiqueter tes textes avec ces outils. Tu pourras déjà travailler sur les noms communs (les substantifs c’est une autre histoire !).

Fnyhg Nytbyvg!
comment: !/usr/bin/env python
comment: -*- coding: utf-8 -*-
import codecs  
print(codecs.encode("Salut Algolit!","rot13"))

http://twitter.com/pickover
Cliff Pickover @pickover
Shiver in ecstasy. This is a palindromic prime:

http://www.openbeelden.nl/media/615463/Schrijfmachine_collectie
type writer collection

https://larlet.fr/david/biologeek/archives/20080511-bonnes-pratiques-et-astuces-python/
passage sur les lists comprehensions très clair

http://fabienpoulard.info/post/2011/01/09/Python-et-Tree-Tagger
Tree Tagger

http://fabienpoulard.info/post/2011/01/04/Petit-script-pour-d%C3%A9couvrir-les-expressions-rationnelles
découvrir les expressions rationelles (même source que TreeTagger)

ROT 13 in LOOP
comment:!/usr/bin/env python
comment: -*- coding: utf8 -*-
from codecs import encode, decode
print("\n###########################################################\n")
print("#\n#                ROT13 is NOT fun!                       #\n#")                                                        
print("#\n###########################################################")
def rotatif(source):
    source = unicode(source,"utf8")
    #print "type de saisie : "
    #print type(source)
    sortie = encode(source, "rot13")
   #bug en cas de caractères non-ASCIIsans la ligne suivante
    sortie = unicode(decode(sortie, "iso-8859-1"))
    #print "type de sortie : "
    #print type(sortie)
    return sortie
while 1:
    print(rotatif(raw_input(u"\nSaisissez une phrase : ")))

http://inforef.be/swi/download/apprendre_python.pdf
M. Swinnen, Apprendre à programmer avec Python
le petit exercice ROT13 posté précedemment m’a pris pas mal de temps à
cause de problèmes de conversions de mode string en mode unicode et
vice-versa.
M. Swinnen en a donné une explication très claire, que je vous recommande:
chapitre ‘Le point sur les chaînes de caractères’
M. Swinnen ne connait pas OSP manifestement.

http://hawaiicowboy.free.fr/info.html
WHAT ARE YOU ? by Stéphane Degoutin, Marika Dermineur and Gwenola Wagon, november 2005.
http://whatareyou.net
Ce projet me fait penser à un 100000 milliard d’associations de modes

https://fr.flossmanuals.net/initiation-a-python
Cédric Gémy, FLOSS manuals francophone

https://fr.wikipedia.org/wiki/Jean_V%C3%A9ronis
Linguiste et spécialiste du Web

http://difdepo.hypotheses.org/bibliographie
blog d’un groupe de chercheurs dédié à l’oulipo, ils ont entre autres
établis une bibliographie complète (ou presque)

http://archive.olats.org/livresetudes/basiques/litteraturenumerique/basiquesLN.php
Les basiques: La Littérature numérique

http://www.formules.net/index.html
La revue Formules qui se consacre à la littérature contrainte: on
peut pas tout consulter mais on a accès aux éditoriaux et et aux
sommaires.

https://readingclub.fr/info
Project of online reading and writing by Annie Abrahams and Emmanuel Guez

https://writingmachines.org
website of Emmanuel Guez

http://fr.wikipedia.org/wiki/Spam#Origine_du_terme_.C2.AB_spam_.C2.BB
https://www.google.be/search?q=spam+can&oe=utf-8&rls=org.mozilla:en-US:official&client=firefox-a&channel=fflb&gws_rd=cr&um=1&ie=UTF-8&hl=fr&tbm=isch&source=og&sa=N&tab=wi&ei=bIMUUufQBrTy7Ab2zoDoBg&biw=1276&bih=635&sei=b4MUUoLANMqqhQfk2IDQCw
Origine du terme spam

http://www.wintermute.org/brendan/?p=exquisite-code
Project by Brendan Howell who did performance at Chercher le Texte in Paris.
In his work he’s looking at different ways of reading and writing using
Python and a collection of texts. A book that came out of a
collective writing session was part of the exhibition’ Book with an
attitude’ during LGRU Porto.

http://www.clips.ua.ac.be/pages/pattern
Just came across this, from the university of Antwerp, not too far from Brussels
yahoo, there is an imap spam reader given as an example here:
http://www.clips.ua.ac.be/pages/pattern-web

http://taalschrift.org/editie/99/schrijf-en-u-wordt-ontmaskerd
I read this interview with Walter Daelemans who is leading the department of CLIPS
I translated some interesting fragments on their work:
What if they analyze the content of the article with their software?
Walter Daelemans: ‘Then we can state with certainty that you’re older
than 25 and for 80% certainty you’re a man, with a university degree. We
can state things about your personality: whether you’re extravert or
intravert, think rationally or rather intuitively. If you would chat, we
can also define your dialect or regional accent…’
What do you need in order to make these “predictions”?
Walter Daelemans: ‘We need at least 5000 words of text by you. … We
look for two kinds of information: objective information, like facts;
and subjective, like opinions. Next, we see how language has been used
for this: choice of words, construction of sentences… Stylometrie.’
Do women write that differently than men?
(An: hold your horses ;–))
Walter Daelemans: ‘Women use more personal pronouns, more verbs, more
relational words than men. Men use more articles and nouns, more
prepositions and more words that describe quantity. It doesn’t matter
whether they write fiction or non-fiction, the differences remain the
same. Male writers will use more a non-fictional descriptive style.
Women who write scientific texts, will do that - against all
expectations - rather in fictional style.Mannelijke schrijvers zullen
bovendien meer in de non-fictionele, beschrijvende stijl schrijven.’
And in order to detect lies on fora, chatrooms:*
‘A man who wants to make believe he’s a teenager or a child, will use too
much ’fashionable’ words of that generation, and not in the right context…
someone who writes a false review (of a hotel he’s never been to), will
put too much irrelevant comments.’
Comment Algolit: Indeed we should look at this non-naively.

http://www.internetactu.net/2012/03/08/lift12-ecrire-avec-les-machines/
« La fin est un problème avec les histoires numériques », s’amuse James Bridle. « Comment les terminer ? J’ai grandi dans Geocities qui a été tout simplement effacé. Je ne sais pas comment on termine une histoire. Je ne sais pas comment terminer Shipadrift. Je fais des projets sans en connaître la fin, mais ils sont souvent plus intéressants après avoir été fait », sourit Bridle. « L’important est surtout de comprendre comment les personnes apprécient leurs propres histoires personnelles dans cet espace codé, dans ce monde diffus. »

http://www.christian-faure.net/2012/01/31/le-devenir-algorithmique-4-les-jeux-decriture/

2014

http://p-dpa.net/
Post-Digital Publishing Archive
The aim of P-DPA is to systematically collect, organize and keep trace of experiences in the fields of art and design that explore the
relationships between publishing and digital technology. The archive acts as a space in which the collected projects are confronted and juxtaposed in order to highlight relevant paths, mutual themes, common perspectives, interrelations, but also oppositions and idiosyncrasies.

http://seminaire.sens-public.org/
Really good program here.

http://www.jeudepaume.org/index.php?page=article&idArt=1887
Recording of Conference by Stéphanie Vilayphiou at Jeu de Paume, with Alessandro Ludovico.
I really enjoyed your comment on how you had to ‘mask’ your script, so Google Books would not identify it as the same ‘visitor’ each time you would launch a request in Google Books. Changing between Firefox, Chrome, Safari & also changing between locations.
How curious that requests from Belgium are much quicker blocked than requests from France.

http://eliterature.org/2014/01/call-for-1st-time-e-lit-artists-21514/
yearly applications since

http://docs.behat.org/quick_intro.html
https://github.com/cucumber/cucumber/wiki
Anyone used behat (php) or cucumber (ruby)?
With BDD, you write human-readable stories that describe the behavior of your application. These stories can then be auto-tested against your application. For example, imagine you’ve been hired to build the famous ls UNIX command. A stakeholder may say to you:
Feature: ls
In order to see the directory structure as a UNIX user I need to be able to list the current directory’s contents
Scenario: List 2 files in a directory
Given I am in a directory “test”
And I have a file named “foo”
And I have a file named “bar”
When I run “ls”
Then I should get:
"""
bar
foo
"""
In this tutorial, we’ll show you how Behat can execute this simple story as a test that verifies that the ls commands works as described.

http://haccslab.com/?p=62
Mark Marino and others do a kind of online working group where they do “close reading” of the software used in works in electronic literature. It’s mostly academics but the conversations can be interesting. Today is the last day to apply to participate.

http://pixelmonkey.org/pub/nlp-training/
These are slides from a presentation but they do walk through some nice examples using NLTK with lots of code.

http://www.bldgblog.com
Articles read from Beyond the Beyond (Sterling’s blog, category : web semantics)
The Writer as Meme Machine by Kenneth Goldsmith
http://www.newyorker.com/online/blogs/books/2013/10/the-writer-as-meme-machine-how-has-the-internet-altered-poetry.html
“Quality is beside the point–this type of content is about the quantity of language that surrounds us, and about how difficult it is to render meaning from such excesses. In the past decade, writers have been culling the Internet for material, making books that are more focussed on collecting than on reading. These ways of writing–word processing, databasing, recycling, appropriating, intentionally plagiarizing, identity ciphering, and intensive programming, to name just a few–have traditionally been considered outside the scope of literary practice.”
From one language to another : from translation to conversion :
https://www.technologyreview.com/2013/09/25/15029/how-google-converted-language-translation-into-a-problem-of-vector-space-mathematics/
“This is an important clue. The new trick is to represent an entire language using the relationship between its words. The set of all the relationships, the so-called”language space“, can be thought of as a set of vectors that each point from one word to another. And in recent years, linguists have discovered that it is possible to handle these vectors mathematically. For example, the operation ‘king’ - ‘man’ + ‘woman’ results in a vector that is similar to ‘queen’.It turns out that different languages share many similarities in this vector space. That means the process of converting one language into another is equivalent to finding the transformation that converts one vector space into the other.This turns the problem of translation from one of linguistics into one of mathematics.
Words Matter vs Clinical English Written Standard
http://techcrunch.com/2014/01/11/such-dfw-very-orwell-so-doge-wow/
”This is the weird thing the Internet has done to language: Standard Written English – or, at least, its most fundamentalist form, Clinical Standard Written English – has actually become incorrect in most online contexts. Before he wrote 1984, George Orwell wrote a famous essay called “Politics and the English Language http://en.wikipedia.org/wiki/Politics_and_the_English_Language,” which decried: In our time it is broadly true that political writing is bad writing. Where it is not true, it will generally be found that the writer is some kind of rebel, expressing his private opinions and not a “party line.” Orthodoxy, of whatever color, seems to demand a lifeless, imitative style."
Tree of Codes
https://2005.visual-editions.com/our-books/tree-of-codes
code façon cut-up by Jonathan Safran Foer published by Visual Editions (printed in Belgium)

http://terribleminds.com/ramble/2014/02/18/baboon-fart-story-is-now-an-actual-thing/
Also on a goldsmithian tip, today’s amazon self-publishing scandal / prank

http://p-dpa.tumblr.com/post/67140118292/il-nulla-luca-fadda-2013-a-free-350
Related to the previous link, there’s the story of this Italian guy who published an empty book and it got to the top of Amazon chart.
The “empty” book (it actually had a blurb) was set free to download for 4 days and it was downloaded 250 times.
The author says: “the fact that with 250 copies you get to the top of the chart, should make you think. Maybe the algorithm that generates the chart doesn’t work properly and maybe it should be the case to reprogram it in base of the reviews.”
My guess is that the Italian store is still so fragmented and small, that is not that hard to get to the top. Also interesting to notice that the author doesn’t have any consciousness of the microhistory from which the book derives: no mentions of the pletora of other (actual) empty books produced and a -somewhat pathetic- use of the project as an excuse to moralize on the ethics of self-publishing and literature: <<the books isfully ironic, fully stupid, fully simple. So simple that anyone could have
made it. I didn’t want to show off my intelligence with this little thing,
and who wanted to understand this, is simply superficial.>>

In brief, a missed opportunity.

Projects by Juan Julián Merelo who presented his git-novel at the latest FOSDEM. He’s head of the Office for Free Software at the
University of Granada. The blog of his Dpt: http://geneura.wordpress.com/
http://jj.github.io/hoborg
Novel developed in GitHub, open source, and published on Amazon.
http://jj.github.io/HashSlash
Other novel on its way (in Spanish)
https://github.com/JJ/PiBook
Script to create a book composed of the 9999 first digits of Pi. Translated from the original in Spanish by Psicobyte specially for FOSDEM’14.

https://weise7.org/book/
Unlike other books, this book acts as an Internet independent wireless server, running from a tiny, custom designed computer inside the book.

https://bibliotecha.info/
this project, made during an hackaton by PZI students, made me think of it.

Martin Howse
http://www.1010.co.uk/org/
His Black Death and Dark Interpreter noise-synthesizers are inspired / based on literature.

Kathrin Günther
http://katier.org/
Star Shots and Spirit Photography

http://www.occultomagazine.com/

http://www.jimcampbell.tv/portfolio/miscellaneous_references/
Formula for computer art

http://litteraturing.net/
2014, nouvelle année, nouveaux archivages, et nouveaux essais…

http://lourdesiglesias.net/susurros.php
Working with a friend writer, Lourdes Iglesias, we made an experiment to browse her new novel by concordances.
It is inspired by the concordance function of the nltk toolkit: http://www.nltk.org/book/ch01.html
(look for concordance in the page, there is no direct link), source is here: http://www.nltk.org/_modules/nltk/text.html
For this, I am trying to translate the original function in a regular expression to make it more portable.

http://scifi2scifab.media.mit.edu/2014/01/07/2h2k-lawyer-artificial-labor-and-ubiquitous-interactive-machine-learning
artifical labour & ubiquitous interactive machine learning and part of the website have been finished re-using LOREM GIBSON, a Lorem Ipsum generator based on Gibson’s texts : )

https://github.blog/2012-06-12-github-data-challenge-winners/
https://github.com/blog/1476-get-up-to-speed-with-pulse
Yes, it’s really crazy how Github goes into turning itself as a CV manager.
Git and Github even more is really a pharmakon: it’s useful to read the process of a project but it can turn easily into a control tool. When we present OSP’s website, which visualizes our Git commits, we try to address this question by focusing on the process than on the fact that clients can control our work. But indeed, because we know it’s public, we try to avoid anger messages targeted to our clients ;) we turn them into private jokes.
Therefore, I did like this one: “Percentage of Commit Messages with Expressions of Anger” on https://github.com/blog/1162-github-data-challenge-winners. Commit messages, because we commit, can really become an outlet, an “exutoire” in French (“outlet” doesn’t seem strong enough)

http://www.lrec-conf.org/proceedings/lrec2012/pdf/330_Paper.pdf
“In this paper, we present the newly established Danish speech corpus PiTu. The corpus consists of recordings of 28 native Danish talkers (14 female and 14 male) each reproducing (i) a series of nonsense syllables, and (ii) a set of authentic natural language sentences. The speech corpus is tailored for investigating the relationship between early stages of the speech perceptual process and later stages. We present our considerations involved in preparing the experimental set-up, producing the anechoic recordings, compiling the data, and exploring the materials in linguistic research. We report on a small pilot experiment demonstrating how PiTu and
similar speech corpora can be used in studies of prosody as a function of semantic content. The experiment addresses the issue of whether the governing principles of Danish prosody assignment is mainly talker-specific or mainly content-typical (under the specific experimental conditions). The corpus is available at http://amtoolbox.sourceforge.net/pitu/”

https://www.ny-web.be/artikels/interview-kenneth-goldsmith/
“I think of both foreign human languages and computer languages in the same way: both are tools which provide me with the opportunity to create new texts, albeit in different ways. For example, I am currently making recordings of me reading the works of Wittgenstein in German, a language I neither read nor understand. I so horribly mispronounce the words that most German speakers who hear my work don’t recognize it as German. So what is it, then? It’s these sorts of questions that fascinate me.”
“Computer languages are another story. The Canadian poet Christian Bök has predicted that in the future, the poet will have to learn Perl http://www.perl.org/ in order to write poetry. It’s a good point and brings to mind one of the reasons for the sorry state of electronic poetry. It’s hard enough for a poet to write a good poem. Now she must also be a programmer and a designer. These are great demands and I only know of one poet who can write poems, program and design equally well. The financial hurdles of getting design or programming help are insurmountable for most poets, hence the pitiful state of the art today.”

http://www.spacecaviar.net/
Led by former Domus editor Joseph Grima, Space Caviar has developed a real-time publishing algorithm called Fear of Missing Out (FOMO) a piece of software that will automatically create written articles from live speech and social media streams during a three-day programme of talks held in the Nike http://www.nike.com/Aero-static Dome at Palazzo Clerici this week.
Source: http://www.dezeen.com/2014/04/08/space-caviar-fomo-algorithmic-journalism-machine-free-newspaper-milan-2014/

http://jonaslund.biz/works/the-fear-of-missing-out/
do you know if this has any relation to the Jonas Lund project “The Fear of Missing Out”? It also used generative algos to generate mash-up descriptions of installation artworks.
I’m pretty sure there is no connection. Also on algopop, Jonas’ piece was mentioned: http://algopop.tumblr.com/post/82086104487/fomo-algorithmic-journalism-machine-via-dezeen

The Pale King, David Foster Wallace
“I went through a sudden period where I couldn’t read. I mean that I could actually read - my mother knew that I could read from when we’d read children’s books together. But for almost two years at Machesney, instead of reading something I’d count the words in it, as though reading was the same as just counting the words. For example, ‘Here came Old Yeller, to save me from the Hogs’ would equate to ten words which I could count off from one to ten instead of its being a sentence that made you love Old Yeller in the book even more […] I still sometimes lapse into counting words, or rather usually the counting goes on when I’m reading or talking, as a sort of background noise or unconscious process, a little like breathing.”

http://www.shakespeare-navigators.com
POLONIUS
What do you read, my lord?
HAMLET
Words, words, words.
POLONIUS
What is the matter, my lord?
HAMLET
Between who?
POLONIUS
I mean, the matter that you read, my lord.

http://bengrosser.com/projects/scaremail/scaremail-generator/
Based on text of Fahrenheit 501
code includes markov chain :-)

http://p-dpa.tumblr.com/post/89460001882/almanacco-bompiani-1962-le-applicazioni-dei
Almanacco Bompiani 1962 - Le applicazioni dei calcolatori elettronici alle scienze morali e alla letteratura, 1962, pp. 87-188.
Arguably the first Italian survey of the employment of computers in the field of literature and moral science. Including, among many others, Giovanni Anceschi, Nanni Balestrini, Silvio Ceccato, Umberto Eco, Karl Gerstner, Bruno Munari, Diter Rot.

This tomb holds Diaphantus. Ah, what a marvel! And the tomb tells scientifically the measure of his life. God vouchsafed that he should be a boy for the sixth part of his life, when a twelfth was added, his cheeks acquired a beard; He kindled for him the light of marriage after a seventh, and in the fifth year after his marriage He granted him a son. Alas! late-begotten and miserable child, when he had reached the measure of half his father’s life, the child grave took him. After consoling his grief by this science of numbers for four years, he reached the end of his life.(A riddle written by a friend of Diophantus of Alexandria, author of the Arithmetica, 3rd century. Found in The Annotated Turing by Charles Petzold, Wiley Publishing.)

http://www.in-vacua.com/cgi-bin/tapemark.pl
an online generator of the algorithmic poem by Balestrini
Source code: http://www.in-vacua.com/cgi-bin/tapemark.pl
Tape Mark 1 is a poem by Nanni Balestrini. It dates from 1961. It was exhibited in Cybernetic Serendipity, ICA Gallery, 1968.
It is sometimes credited as the first computer poem. It is a significant early example of computer poetry.
This version has been programmed using the description in the Cybernetic Serendipity catalogue.
The original code is unavailable. The present program produces verses similar to the English translations that are still available.
There is the occasional oddity in syntax, but according to the catalogue of Cybernetic Serendipity there were, post-computer,
‘a few editorial changes in points of grammar and punctuation’. It is stated in the catalogue that ten elements were selected. However, on my count the number is nine (see below). Then this raw text was divided up into six lines of metrical units. The program emulates this approximately.
This is an example verse from the Cybernetic Serendipity catalogue. I have indicated where phrases start and finish:
Hair between lips, || they all return
to their roots || in the blinding fireball ||
I envision their return || until he moves his fingers
slowly, || and although things flourish ||
takes on the well known mushroom shape || endeavouring
to grasp || while the multitude of things come into being.
Here’s another:
In the blinding fireball || I envisage
their return || when it reaches the stratosphere ||
while the multitude of things come into being || head pressed
on shoulder || thirty times brighter than the sun ||
they all return to their roots || hair
between lips || takes on the well-known mushroom shape.
Balestrini sources:
Lao Tzé’s Tao Te Ching: ‘While the multitude of things comes into being, I envisage their return. Although things flourish, they all return to their roots.’
Michihito Hachiya’s Hiroshima Diary: ‘The blinding fireball expands rapidly, thirty times brighter than the sun. When it reaches the
stratosphere, the summit of the cloud takes on the well-known mushroom shape.’
Paul Goldwin’s The Mystery of the Elevator: ‘head pressed on shoulder, hair between lips, lay motionless without speaking, till he moved his fingers slowly, trying to grasp.’

https://eld2015.wordpress.com/
the following seminar might be of interest:
‘Digital Literary Studies’ is an international conference exploring methods, tools, objects and digital practices in the field of literary studies. The digitization of artifacts and literary practices, the adoption of computational methods for aggregating, editing and analyzing texts as well as the development of collaborative forms of research and teaching through networking and communication platforms are three dimensions of the ongoing relocation of literature and literary studies in the digital medium. The aim of this two-day conference is to contribute to the mapping of material practices and interpretative processes of literary studies in a changing media ecology.

http://kavan.land/
algoliterary creation by Catherine, Stéphanie, Alex

https://github.com/dariusk/NaNoGenMo-2014
NaNoGenMo 2014 or National Novel Generation Month: Spend the month of November writing code that generates a novel of 50k+ words.
The only rule is that you share at least one novel and also your source code at the end.

2015

http://rwet.decontextualize.com/pdfs/witz71.pdf
“Grep: A Grammar” by Loss Pequeno Glazier
His grep works: http://epc.buffalo.edu/authors/glazier/dp/appendices.html
It’s not clear if he reorganizes his sentences after the grep, because his texts can be read extraordinarily well.

http://rwet.decontextualize.com
Allison Parrish teaches electronic writing, word games, etc. in the U.S.
Her website has a lot of resources with courses on Unix, Python, Markov chains…

http://lav.io/2014/05/transform-any-text-into-a-patent-application/
A nice work from one of her students: Transform The Communist Manifesto or a Kafka novel into a patent.

https://www.huffingtonpost.fr/2014/10/18/rappeurs-vocabulaire_n_6003668.html?utm_hp_ref=musique
comparing rap vocabulary: rappeurs français vs wordlists

http://chairnerd.seatgeek.com/fuzzywuzzy-fuzzy-string-matching-in-python/
nice explanations

https://github.com/ocropus/ocropy
All that is solid, like books, melts into typos

https://code.google.com/p/python-tesseract/
https://github.com/madmaze/pytesseract
https://code.google.com/p/pytesser/
How does it differ from the python libraries using tesseract? Is it using other machine learning algorithms?

Rob Fitterman & Vanessa Place “Notes on Conceptualisms”
The idea is to read this from our (Jara Rocha) shared southern present european engendered sensibilities, to see if that makes any sense of scratch in the political, ethical and aesthetical issues proposed by this branch of the new conceptualist poetics (a brief intro by Craig Dworkin and some selected works can be found in UBUweb: http://www.ubu.com/concept/ – and Kenneth Goldsmith srives on it along Uncreative Writing as well)

http://median.newmediacaucus.org/art-infrastructures-information/seeing-the-sort-the-aesthetic-and-industrial-defense-of-the-algorithm/
algolit-related material on the issue of representing the algorithm ; you’ll see romanian folk dancers choreographing a Quick-Sorting task, you’ll read a solid article, going from educational attempts and commercial depictions of algo, to claims for a “Countervisuality to Algorithms”.
Quicksort dance: https://vimeo.com/114911241

http://applymagicsauce.com/
This demo predicts your psycho-demographic profile from digital footprints of your behaviour. It reveals how you might be perceived by others online and provides academically robust insights on your personality, intelligence, leadership, life satisfaction and more. We think every citizen has a right to understand their data, but most big tech companies would rather not reveal what is predictable (or profitable) about you. Fortunately, you can now download your social media data and analyse it directly using our tool.

https://zenodo.org/record/18166?ln=en#.VW1vKrzhlC1
Designed by Loraine Further, written by Simon Worthington
The Book Liberation Manifesto is an exploration of publishing outside of current corporate constraints and beyond the confines of book piracy. We believe that knowledge should be in free circulation to benefit humankind, which means an equitable and vibrant economy to support publishing, instead of the prevailing capitalist hand-me-down system of Sisyphean economic sustainability. Readers and books have been forced into pirate libraries, while sales channels have been monopolised by the big Internet giants which exact extortionate fees from publishers. We have three proposals. First, publications should be free-at-the-point-of-reading under a variety of open intellectual property regimes. Second, they should become fully digital in order to facilitate ready reuse, distribution, algorithmic and computational use. Finally, Open Source software for publishing should be treated as public infrastructure, with sustained research and investment. The result of such robust infrastructures will mean lower costs for manufacturing and faster publishing lifecycles, so that publishers and publics will be more readily able to afford to invent new futures. For more information on the Hybrid Publishing Consortium see http://consortium.io/

http://www.joshmillard.com/markov/calvin/
Calvin and Markov digests Calvin and Hobbes strips and generates new comics using Markov chains, perl, and Imagemagick.

http://motherboard.vice.com/read/this-ai-creates-interactive-fiction-by-reading-other-peoples-stories
AI creates interactive fiction. Sheherazade is coming to life

<unfold.thevolumeproject.com>
/A Library Where The Books Have Melted Into One Another and The Titles
Have Faded Away./
A curatorial experimentation on the digital folder to trigger anti-authorial agency, movement of thoughts, untimely collaborations.
Re-generated by seven guest curators every two months, Unfold is a library hosting shifting constellations of artistic content, books,
found objects and software, both newly commissioned and already existing. Like all libraries, Unfold is a space of copies, copies of
copies, appropriations, heterogeneity and contradictions.
The first Unfold is based on http://thevolumeproject.com/ and follows a contrapuntist arrangement for coded texts, found images,
field recordings, photocopies, notes and videos; time-lapsed and displaced. With contributions by Mounira Al Solh, Stefanie Baumann, Andrew Beccone, Walter Benjamin, Nadia Bou Ali, Pauline Curnier Jardin, Aldis Ellersdottir, Per Hattner, Sami Khatib, Marcell Mars, RYBN.ORG, Walid Sadek (& more).

A few quick things that might be interesting:

http://spacy.io/
Another python NLP library. Documentation is nice. The graphing and syntax parsing features look pretty good. I think it could be cool for more poetic works than Pattern or NLTK. Also the POS tagger for english is supposed to be very good.

The Corpus of American Soap Operas
http://corpus.byu.edu/soap/
I have no ideas but this just seems like a gold mine for algo-literarians who have any sense of irony.

DDAI - Digitale Demenz (Artificial Intelligence)
http://ddai.de
I made a sort of artistic catalog for a gallery show on the topic of AI in popular culture. The site is constantly re-generated mostly using markov chains to munge the text. I also used it as an opportunity to curate some of my favorite music, scientists, kooks and skepticism.

http://adarkroom.doublespeakgames.com/
A textual game online

https://github.com/ryankiros/neural-storyteller
Neural storytelling with Python.
neural-storyteller is a recurrent neural network that generates little stories about images. This repository contains code for generating stories with your own images, as well as instructions for training new models.
https://medium.com/@samim/generating-stories-about-images-d163ba41e4ed
A review about it.

http://socialmediacollective.org/reading-lists/critical-algorithm-studies/
Critical algorithm studies: a readinglist.
Comes from Microsoft… Matthew Fuller is on.

https://www.renpy.org/
Ren’Py is a visual novel engine – used by thousands of creators from around the world – that helps you use words, images, and sounds to tell interactive stories that run on computers and mobile devices. These can be both visual novels and life simulation games. The easy to learn script language allows anyone to efficiently write large visual novels, while its Python scripting is enough for complex simulation games. Ren’Py is open source and free for commercial use.

https://www.wired.com/2015/11/google-open-sources-its-artificial-intelligence-engine/
Google open sourced Tensorflow.

2016

https://www.idpureshop.ch/web/catalogue.aspx?cat=39
Dadabot, publication by Nicolas Nova: Essay about the hybridization of cultural forms (music, visual arts, literature) produced by digital technologies.
http://www.makery.info/2016/01/19/nicolas-nova-voit-du-dada-dans-la-data/
Interview with Nicolas Nova

https://constantvzw.org/site/SICV-Scandinavian-Institute-of-Computational-Vandalism.html
http://sicv.activearchives.org/logbook/
Nicolas Malevé & Michael Murtaugh at Transmediale, Berlin.

http://arxiv.org/abs/1601.03313
https://github.com/valentin012/conspeech
Political Speech Generation
Another interesting article+code on text generation (using topic modeling).

https://keepnotes.com/stanford-university/machine-learning
For those who like to learn more on neural networks and how to implement Tensorflow.

http://nlp.lsi.upc.edu/freeling/node/1
http://www.talp.upc.edu/
FreeLing is a C++ library providing language analysis functionalities (morphological analysis, named entity detection, PoS-tagging, parsing, Word Sense Disambiguation, Semantic Role Labelling, etc.) for a variety of languages (English, Spanish, Portuguese, Italian, French, German, Russian, Catalan, Galician, Croatian, Slovene, among others).
These tools are developed and maintained at TALP Research Center, in Universitat Politècnica de Catalunya. They also benefit from many external contributions from a wide community.

http://passeurdesciences.blog.lemonde.fr/2016/04/04/peut-on-reconnaitre-un-ecrivain-a-sa-ponctuation/
Can a writer be recognised according to the punctuation?
https://neuroecology.wordpress.com/2016/02/15/punctuation-in-novels-2/
Original English post.
https://medium.com/@neuroecology/punctuation-in-novels-8f316d542ec4
The red & blue drawings are interesting too.

http://allpriorart.com/
Algorithmically generated prior art.
Exhausting patent space.

http://robotcomix.com/comix/Catalogue/mobile/
Comics generation by Tony Veale.
The scripts generate a lot of ugly male characters, but some of the questions/thoughts are interesting.
https://horizon-magazine.eu/article/creative-computation-and-what-if-machine.html
The What If Machine, a research project on the automatic generation of ideas for fiction.
“You get a lot of computers that generate stuff that they themselves don’t understand,™ said Dr Veale, who is based at University College Dublin, Ireland.”They generate poetry, but if the computer doesn’t know what the poem means, then the computer is not being creative.™
Tony Veale (same person who launched comics generator above) & Leading Research Group: http://afflatus.ucd.ie/.
He leads https://prosecco-network.eu/, a research network on Computational Creativity.

http://bookworm.benschmidt.org/posts/2015-10-30-rejecting-the-gender-binary.html
Rejecting the gender binary in language.
“What this method actually lets us do is hyperfixate on gender as a category of textual analysis. A world without gender from this
standpoint is interesting because of what it tells us about gender and its strength, not because it lets us tap into some world of
unrestrainedly polymorphous perversity but because it highlights how the many different roles that gender itself plays.”

http://arstechnica.com/the-multiverse/2016/06/an-ai-wrote-this-movie-and-its-strangely-moving/
Algorithmic SF
If it can play Go, it can also write movies?

http://www.i-could-have-written-that.info/
Project by Manetta Berends.
i-could-have-written-that* is a practice based research project about text based machine learning, questioning the readerly nature of the techniques and proposing to represent them as writing machines. The project is built with ie. nltk, Pattern, python, cgi, jinja, git.

https://www.coursera.org/course/nlp
Free online course Natural Language Processing from Stanford University, by the authors of the famous book

Algorithmic surrealism
http://objectdreams.tumblr.com/post/139317698609/here-is-my-predictive-text-rewrite-of-the-elements
Other results at http://objectdreams.tumblr.com/
http://www.theverge.com/2016/7/22/12260406/object-dreams-tumblr-writing
Source code at https://github.com/jbrew
About the funniest algorithmic writing I came across so far.

https://www.technologyreview.com/s/602025/how-vector-space-mathematics-reveals-the-hidden-sexism-in-language/
As neural networks tease apart the structure of language, they are finding a hidden gender bias that nobody knew was there.

http://www.degeneratestate.org/posts/2016/Apr/20/heavy-metal-and-natural-language-processing-part-1/
A very funny way to understand what NLP is about.

https://news.slashdot.org/story/16/11/23/1942237/googles-ai-translation-tool-creates-its-own-secret-language
Google’s AI Translation Tool Creates Its Own Secret Language

Some resources on Deep Learning and neural networks

• Software for Deep Learning and neural networks
  Tensorflow (Google): https://www.tensorflow.org/
  a lot of tutorials, including on language processing: https://www.tensorflow.org/versions/r0.11/tutorials/index.html
  Free course on Deep Learning (Tensorflow): https://keepnotes.com/stanford-university/machine-learning
  Some links to how it is used: https://research.googleblog.com/2016/11/celebrating-tensorflows-first-year.html
  Theano: <http://deeplearning.net/software/theano/index.html />
  http://deeplearning.net/tutorial/intro.html
  other: overview November 2015 http://venturebeat.com/2015/11/14/deep-learning-frameworks/?imm_mid=0dd6b0&cmp=em-data-na-na-newsltr_20151209
  Microsoft Cognitive Toolkit https://www.microsoft.com/en-us/research/product/cognitive-toolkit/
• Free online books
  Deep Learning - Ian Goodfellow and Yoshua Bengio and Aaron Courville: http://www.deeplearningbook.org/
  Neural Networks and Deep Learning - Michael Nielsen http://neuralnetworksanddeeplearning.com
• Blogs and articles
  The Unreasonable Effectiveness of Recurrent Neural Networks http://karpathy.github.io/2015/05/21/rnn-effectiveness/
  Embed, encode, attend, predict: The new deep learning formula for state-of-the-art NLP models: https://explosion.ai/blog/deep-learning-formula-nlp?imm_mid=0eaadb&cmp=em-data-na-na-newsltr_20161116
  Deep Visual-Semantic Alignments for Generating Image Descriptions: http://cs.stanford.edu/people/karpathy/deepimagesent/
• Simulation tools
  Simulation of NN in browser: http://playground.tensorflow.org

https://www.algolit.net/neural_networks_tensorflow/neuralnetworks1_word2vec_film.mov
We looked into word2vec word embeddings, using tensorflow. This is a film of the process by Olivier Perriquet and a (half)
commented script in the same folder.

2017

http://cs.stanford.edu/people/karpathy/densecap/
DenseCap: Fully Convolutional Localization Networks for Dense Captioning

https://www.theguardian.com/technology/2017/feb/23/wikipedia-bot-editing-war-study
Study reveals bot-on-bot editing wars raging on Wikipedia’s pages
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0171774
Even good bots fight: The case of Wikipedia
And chatbots talking to each other is becoming a new trend on Youtube:
https://www.youtube.com/watch?v=WnzlbyTZsQY
https://www.youtube.com/watch?v=X_tvm6Eoa3g

https://arstechnica.com/features/2017/04/the-secret-lives-of-google-raters/
Article on the raters which Google uses to improve its search algorithm.
“Large scale data mining practices like Google Search are not fully automated, but need an almost invisible army of annotators doing the low-level, Mechanical Turk-alike work of rating and building the training and test sets. It is mostly on the labour conflicts, but it tells a lot of the future of low-level jobs.”
It points also to Googles annotating manual: https://static.googleusercontent.com/media/www.google.com/en//insidesearch/howsearchworks/assets/searchqualityevaluatorguidelines.pdf.
Boring to read, but very interesting to analyse from the viewpoint of how Google becomes our guide through the information sphere and how hidden opinions and assumptions influence that.

http://emotions.periscopic.com/inauguration/
Emotional arcs of the past ten U.S. presidential inaugural addresses.
It could be an idea to start creating types of birds wearing these feathers…

http://grail.cs.washington.edu/projects/AudioToObama/
A neural network approach to Obama ventriloquism: Synthesizing Obama: Learning Lip Sync from Audio:
To be clear, this is not about text generation, but video generation matching the audio.
But in theory, and if enough computer power and training data is available, it should be possible to do some text generation, then feed
it into some audio generation tool like https://lyrebird.ai/ and next use this video generation approach. And then everyone can create their fake news …

https://aestheticsforbirds.com/2017/07/20/emergent-poetry-and-google-translate/
Unlike the author of this blog post, I think this is interesting both philosophically and artistically.

https://www.theguardian.com/books/booksblog/2017/apr/28/ebook-every-reader-obliged-to-edit-a-universe-explodes-tea-uglow
Make it your hone: the ebook that you are forced to edit as you read. On each page of A Universe Explodes by Tea Uglow, owners are required to add one word and remove two – which amounts to an odd reading experience.

2018

http://aiweirdness.com/post/171451900302/do-neural-nets-dream-of-electric-sheep
A post on fooling a CNN into seeing sheep where they are not, or failing to spot them because they are the wrong color or in the wrong place.

https://spectrum.ieee.org/the-human-os/robotics/artificial-intelligence/hacking-the-brain-with-adversarial-images
article describing a research into adversarial images fooling both neural nets and humans, with the caveat that the images
were shown ‘Malevé’-style: “Subjects sat in front of a computer screen and were shown an image from a specific group for between 60 and 70 milliseconds … The short amount of time that the image was shown mitigated the difference between how CNNs perceive the world and how humans do.”
Closing with cliché positive and negative future scenarios (both horrible in their own way ?):
“…machine learning techniques could potentially be used to subtly alter things like pictures or videos in a way that could change our
perception of (and reaction to) them…”
“These techniques could also be used in positive ways, of course, and the researchers do suggest a few, like using image perturbations to improve saliency, or attentiveness, when performing tasks like air traffic control or examination of radiology images…”

http://aiweirdness.com/post/159132506927/in-which-a-neural-network-learns-to-tell
Janelle Shane, who does this blog also has an excellent one on knock-knock jokes:“The neural network can be trained to write recipes, invent Pokemon, and invent superhero names. But can it learn to tell a joke?”

https://github.com/mikekestemont/ghent1516
https://github.com/mikekestemont/lot2016
Nice ‘writer’ courses for Python - compared to CodeAcademy, all examples use text, no numbers :-)

https://ulbstic.github.io/workshop/team/
MaSTIC proposes a two-year academic Master in Information Science with a strong focus on both practice and theory. The Master gives access to various professions such as data scientist, functional analyst, document & records manager and projects within the domain of digital humanities.

https://www.washingtonpost.com/lifestyle/can-algorithms-be-funny-veterans-of-clickhole-and-the-new-yorker-team-up-to-find-out/2018/04/04/48b23e1e-36b0-11e8-8fd2-49fe3c675a89_story.html
AlgoLit goes Mainstream?
“At its funniest, this 21st century form of techno-Dadaism is a sublime collaboration ­between humans and machines, merging the creativity of one with the robotic uncanniness of the other into an original comic voice. Welcome to the world of Botnik Studios, a digital company that specializes in artificial intelligence-assisted interactive comedy.”
https://botnik.org/
Botnik is a machine entertainment company run by comedy writers. We use computers to remix text

https://archive.org/details/fifteenthousand00kleigoog
A corpus of “Fifteen Thousand Useful Phrases” that every writing machine should know

http://www.galaxykate.com/pdfs/galaxykate-zine-opulentai.pdf
Nicely made ‘Opulent AI manifesto’

https://twitter.com/GalaxyKate
Kate Compton on Twitter, geologic choreographer, dance breeder, crafter of twitching generative bots.

https://www.haujournal.org/index.php/hau/article/view/hau3.3.003
An anthropology of the Bayesian reasoning, “we are our sieves” by Paul Kockelman.

https://monoskop.org/log/?p=20147
Adrian Mackenzie: Machine Learners: Archaeology of a Data Practice (2017)

https://thegradient.pub/nlp-imagenet/
Computer Vision meets NLP in many ways, how a dataset establishes a paradigm.

https://carolinesinders.com/feminist-data-set/
Feminist dataset by Caroline Sinders

https://pudding.cool/2017/05/song-repetition/
Are Pop Lyrics Getting More Repetitive?