Reports
* Sophie interested in Collaborative storytelling & data retrieving (add topic to the main pad
* Catherine: EXPO ANNA K on 19-11
spreading the fiction in another format - printed book published beginning 2016
hybrid publishing: printed book, online archive, local artwork (to spread the fiction)
the local network (TPLink/Piratebox) as a read-write interface
post-connectivity era, place/time, implies new practises (the wifi-router as political object, or social space)
-> relation to 'Network as a book' - reflection during Kenneth Goldsmith's reading session June 15
-> thinking of collectively written ToS
trying to look for ways to combine infrastructure provided by Library/Pirate box (messy) (http://librarybox.us/whatis.php ) and Michael's Open WRT (https://openwrt.org/ ) proposal
* Cristina
Media Design Master in Piet Zwart.
Interest in alternative means of distributing info.
interested in PirateBox & OpenWRT
uses Raspberry Pi
*(also see Loraine's project uses RPi with infrastructure of Piratebox: http://www.imal.org/en/connectingcities2015/xmlab )
*--> a tour through Molenbeek, walking from RPbox to RPbox, discussing site-specific narratives
*Loraine will publish soon the documentation of the project, open to share/discuss these issues
http://spreektu1060.be/p/creations/#s20
pirate social communities, like: ... ... look at power structure, social organization & currency
Tempory Autonomous Zones, TAZ Hakim Bey - freedom from social rules / allow access to member of a community, tension between open acces & closed (power) structures amongst them, close structure.
https://en.wikipedia.org/wiki/Temporary_Autonomous_Zoneref Offline Art in Transmediale 2015
---forgot to mention why interested in algolit: interested in the way the form of a language influences the content. eg: patent structure. how ambiguities in language are exploited.
---plus previous work with OCR: recording OCR mistakes. reading together with the algorithm
* Manetta
Media Design Master in Piet Zwart.
continuing the last relearn session, but from her own perspective How the common sense is constructed / feminist point of view explored together with Femke
her point of interest: found file of adjectives rates by polarity (positive/negative)
'amazing' listed twice: 0.8 + 0.4 -> 2 different descriptions on semantic levl, when running the program mathematically merged into 0.6
-> how human natural languages are being systemized to make them computable / automate nl-processes
txt of Florian Cramer abt languages, in the book Software Studies, Matthew Fuller
programming languages control the machines
ambiguity of the word code : encoding / decoding processes, not really operating anymore
looking at the infrastcrtures of these systems, interest in Wordnet/lexicon, use to org. semantic systems (semantic+math), project : build a publishing platform
design information processes, how does it influence the information
publishing about natural language processes --- (first step (collecting material) in online here: http://pzwart1.wdka.hro.nl/~manetta/i-could-have-written-that/ )
ref sent by Silvio : a serialized novel written in real time by Joshua Cohen : http://www.pckwck.com/ & http://uselesspress.org/about/
Post Digital Archive project by Silvio: http://p-dpa.net/
*An
studies computational linguistic statistics at the - Univ Antwerpen
statistics is called: "de kunst van het benaderen" / "the art of approximation" --> translating language into numbers. '0' is described as 0.00000000001, a way to avoid calculating with '0', universe of conventions: how to cimplify things? how to make a model of reality. The approximity of the values is commonly 'forgotten' about. (what is a good hypothesis, what is the error-margin?) it's difficult to sometimes 'just' accept certain conventions
--> used to built classifiers, like the sentiment classifiers
temporary publication, made by people in a space; rewriting a book by using a local network, and people customize a book while they share a space
using techniques in terms of time / space / type of research you want to do (anxiousness, positiveness, racism)
An is interested to classify/analyse anxiousness, linked to the ideology of creating anxiousness in the news (risk/certainty)
*book: http://www.degruyter.com/view/product/203826
statistic new in Literature studies : need to define their argument (why statistic is interesting? important? : (not about creating a truth) to create a contxt to look a t priorities)
research in linguistic based on intuition vs build digital model/pattern.
baysian statistics - accept subjective position, talks about 'belief', 'more certain of your belief'
*Hans
libgen, Hans' favorite download website for books (the link changes from time to time), http://gen.lib.rus.ec/
About libgen (abstract from Radical Tactics of the Offline Library, Warwick 2014).
Libgen is a massive online (book) library. Libgen, short for Library Genesis, is also based in Russia, and is even more shadowy than Avax. It was briefly offline in 2011, but soon returned bigger than ever. At the time it had about 800,000 books. It now has well over a million volumes and occupies approximately twelve terabytes of storage. Many e-books in libgen are not optimized – hundreds of them are well over 100 MB in size. For example, Middle Egyptian: An Introduction to the Language and Culture of Hieroglyphs by James P. Allen weighs in at over 107 MB and is not even downloadable via their web portal. Then how to acquire this book? Well, either by downloading the entire 9 TB libgen.info library which is available via torrents (linked in the Tor network), or finding someone who has already done that, so as to acquire it from them. Libgen is a kind of hybrid that is too big to be a convenient Personal Portable Library and so by definition, like archive.org, is an online entity. Yet its survival is predicated on an offline strategy. People have developed their own online libraries on the ground provided by libgen. An example is bookfi.org, based on the original Library Genesis data set, and grown in a different direction. In February 2013, libgen branched out into libgen.org. This has the entire libgen library, and allows people to add books. My research indicates that, around mid-2013, it had an upload rate of approximately 100-150 items per day.
Memory of the world, Marcell Mars: https://www.memoryoftheworld.org/
Link to aaaaarg.org: https://www.memoryoftheworld.org/blog/2014/10/28/aaaaarg-org/
juridical research on big data, how big-data influences interactions between organizations and individuals, in the context of a legal framework
txt generating follow-up (after Relearn) ex : Click-o-Tron http://www.clickotron.com "click-bait articles generated by neural networks"
http://larseidnes.com/2015/10/13/auto-generating-clickbait-with-recurrent-neural-networks/
completely computer generated - neural networks based on words
-> combine with preformatted models using linguistic structures
-> looks for common word combinations, based on what is there (not on a 'history', 'previous model')
possible to show different steps of neural network processing? are there forms of illustrating/visualising the process? But that's not the same as understanding it.
ref: mario "gaining consciousness" with the help of NN https://www.youtube.com/watch?v=qv6UVOQ0F44
Deep learning is a form of neural networks, but other mathematical models are used that improved results a lot. (Geoffrey Hinton: Boltzmann models - https://www.youtube.com/watch?v=AyzOUbkUf3M)
Hans is working on his thesis (on the topic of developing statistics for social networks that has to do with agent-based-modeling, making a lot of simulations and taking the averages out of it --- for statistic studies?) and has access to the supercomputer in Gent.
< monte carlo methods in statistics
mathematical models which create new visibilities - how visibilities are changing / linked to political thinking.
intervening at nodes of transmission - identifying
transmissions could be: HIV, mechanisms of power -> needed in order to develop new political thinking, what works/not anymore in legal framework
*Gijs
Past work : BotOpera (Public Domain Day, iterative perfo at Wiels) bots & Diva the plotter.
Question of the audience - how you bring bots & algolrithms & readwrite machines on stage ? - - > performing arts / another stage ...
audience did not interfere with the chat.
Question of the setting & contxt.
Most interesting were discussions along the way, for example about IRC as a social space. There is a paradox: it's social for the people in it, but not for the audience.
During the opening at Constant, the audience was introduces personally to the installation, and people started to play around with it.
Social reading, writing, watching space. A chatroom. (history, where does it comes from, interesting space between physical & virtual world, & subject to explore.
installation: writing machine, taken on by plotter immerdiately
performance: 1 screen, closed
can be terminal/interface into archive, conversation with people who died
Gijs' interest is in regarding the installation as drawing machine, and regarding the plotter as a performative device
... there can be a chat, bots, the terminal / another form of displaying text?, information being displayed at the other side of the room ...
different groups globally
another project: Text generation project (Obama text generation created during Relearn)
for exhibition in Groningen
writing machines & signatures
*book: Thomas Jefferson and his copying machines
Obama text generation project: working with Markov-chains, using Videogrep.py, using subtitles & video as output format
using video's (from youtube) as an interface for a text-generation project, next to written generated text
in the 80s: Max Headroom, an actor playing a computer generated figure
https://upload.wikimedia.org/wikipedia/en/7/7b/MaxheadroomMpegMan.jpg
https://en.wikipedia.org/wiki/Max_Headroom_%28character%29
Live book workshop with Loraine during Mondothèque workshop
html to print (OSP) + epubs
iframe in html: in your page you can load another page & display it
iframe into epub: import webpage into epu & analyse content
le livre sur le livre : what a book was at present & future, brings these questions now.
Traité de Documentation, Le Livre sur le Livre, Paul Otlet
index at the end of the book - rewrite index with wikipedia content
describes UDC-universal decimal classification system
if you have structure of your story, grab content form internet/learning algorithms - possible in theory
personal - subjective index
An worked on words frequency, rewrite (+ read) Otlet's book based on selected words from this list
Tracks for this afternoon lunch - until 4pm (then skype)
Discussing, sharing, preparing Algolit Tools Collection
Memory of the World - Calibre pluggin : an account to share books or Libgen/other
Who wants to share what?
- list of books + books/texts themselves; algolit bibliography/library
- links / bookmarks
- scripts
- documentation of existing projects (relearn / Kenneth Goldsmith reading session....)
how?
- memory of the world https://www.memoryoftheworld.org/
*to see mixed collections : https://library.memoryoftheworld.org/#property=authors
*index of collections : https://www.memoryoftheworld.org/blog/cat/catalog/
*ex : https://feminism.memoryoftheworld.org
*to install the plugin: http://imgur.com/a/9ET1C
ask/tell to Marcell : a specific Algolit collection
new plugin available : https://www.memoryoftheworld.org/blog/2015/10/07/indian-summer-release/
how-to : https://www.memoryoftheworld.org/blog/2014/10/28/calibre-lets-share-books/
- http://aaaaarg.fail
- https://www.zotero.org/ browser plugin to manage bibliographic references, organizing links + pdfs + ..., it enables the user also to make annotations to the references
- pads: option to read not to write?
- Ethertoff, Etherpad with an extra layer on top, to control access permissions
- wiki: edited notes
- git repository (on server, interface with GitWeb / GitHub / GitLab - has wiki built in the projects): http://git.constantvzw.org/?p=algolit.git;a=tree
- dedicated wifi-router or Raspberry Pi as portative archive
- oralsite: http://sarma.be/oralsite/ http://activearchives.org/wiki/SARMA:_Oralsite + http://activearchives.org/aaa/recent/
mediawiki extensions:
- https://semantic-mediawiki.org/
- https://www.mediawiki.org/wiki/Extension:EtherpadLite / https://www.mediawiki.org/wiki/Extension:EtherEditor
memory-of-the-world algolit libraries:
https://www30647.memoryoftheworld.org/#property=authors&librarian=Johann+Borges
https://library.memoryoftheworld.org/#librarian=Johann+Borges
https://library.memoryoftheworld.org/#librarian=Makoto+Krajewski
https://library.memoryoftheworld.org/#librarian=Ruiz+Of+Byzantium
meantime Manetta starts on aaaaarg a collection named Algolit / Memory of the world a little bit unstable after variours connexions, log out.
Books tagged under Algolit.
TODO List
* Catherine contacts Marcell on issues of Memory of the World - Algolit collection
* An installs Mediawiki (with Denis) & sends url
* Manetta can look at MediaWiki extensions? :-)
* upload books on aaarg from home
* keep serving books on memoryoftheworld
* write comments/recommendations about your shared books at the algolit MediaWiki
* next meeting for Python Beginners: 15-11
* next meeting general: 4-12 12h-20h
*introduction to uploaded books
*track 1/ writing machines - txt generators
*track 2/ wifi routeur - with Loraine (ask availability)
*writing exercises:
(anyuta)
1. is there any 'homework' before the next meeting? :)
we have been uploading books on Memory of the World (Let's Share Calibre) and aaargh.org with tag 'algolit'
could be nice to add and/or read & comment (like to know what book you should read and why / proposal to do this on the mediawiki, separated from the public library collections)
2. do I understand it right that today you've mostly discussed the possibilities, whats and hows and set the plan of actions?
yezzz
3. is the idea described under 'who wants to share what'? meaning, is it the plan to create a kind of library 'Algolit collectiong'?
yes! there will be a mediawiki with the url www.algolit.net, hopefully from next week onwards; the idea to create a page for each track proposed in the algolit_encounters pad, and start adding links/scripts/documentation....
cool! (jara) hey just answered your email
3-12 20h
http://www.passaporta.be/agenda/schrijvers-op-het-net-ecrire-a-lere-dinternet
if you have interesting work to show, let me know :-) (an)