start

1
about: 2012 end of the world / Mayan calander vibes, predictions for 2020
what: part of a text
entry: 1973 MIT Computerprogram World One. Predicts from 2020 a series of catastrophic events & 20y process of slow demise of human civilization.
link: https://www.timeanddate.com/calendar/apocalypse.html

2
about: sponsered content, economic changes, influencers x corona virus
what: title article
entry: Coronavirus Killed the Influencer Market – Maybe for Good
link:https://www.vice.com/en_asia/article/5dmvzz/coronavirus-social-media-influencers-effect

3
about: corona scamming in this case the darkweb, but also people and scalpers on facebook selling groups selling (fake or used) hand santizer or people ringing at doorbells trying to sell soap or a sanitizer
what: title article
entry: Scammers on the dark web offer to sell blood and saliva of someone they claim has coronavirus for $1000 as others create fake vaccines
link: https://www.dailymail.co.uk/news/article-8208887/Scammers-selling-blood-saliva-claim-coronavirus-1000.html

4
about: a random facebook opinion under a government add/post about corona written by “Bayram Turacak” // due to character count it is a snippet but quoted from original
entry: "onzichtbare hoge machten die een nieuwe tijdperk willen " "de nieuwe generatie inwillen meeslepen met de technologie en met dat gechip enzo"
link: facebook comment / screenshot

5
about:
https://fcpablog.com/2020/01/30/did-corruption-cause-the-deadly-coronavirus-outbreak/
entry: Did corruption cause the deadly coronavirus outbreak?

6
about:
entry: bestraft verdwenen of gestorven in China nadat ze de noodklok hebben geluid over coronavirus
https://www.hln.be/nieuws/buitenland/zij-zijn-bestraft-verdwenen-of-gestorven-in-china-nadat-ze-noodklok-hebben-geluid-over-coronavirus~afd449cd/?referer=https%3A%2F%2Fwww.google.com%2F

7
about:
entry: Nu de wereld is afgeleid door de corona-uitbraak, ziet Beijing zijn kans schoon om voor eens en altijd een einde te maken aan de protesten.
https://www.nrc.nl/nieuws/2020/04/19/china-wil-de-klus-in-hongkong-afmaken-a3997211

8 grote boosheid in Afrika over frans tvfragment waar 2 dokters de suggestie doen om in Afrika te experimenteren met anti-coronavaccins.
https://www.ad.nl/buitenland/afrikanen-ziedend-vanwege-chinees-coronaracisme~a4b2ae5a/?referrer=https://www.google.com/

9 had onderliggende klachten
deze vermelding lees ik vaak, maar geeft naar mijn idee weinig toevoegende waarde aangezien er geen waarde van ernst bij word vermeld hoe erg deze onderliggende klachten zijn etc / vermindere of “goedpraten”

10
conspiracy theorists are claiming that Gates somehow created this pandemic
https://www.forbes.com/sites/brucelee/2020/04/19/bill-gates-is-now-a-target-of-covid-19-coronavirus-conspiracy-theories/#ddb332a62277

My chosen direction with collecting data was about the times I got reminded or confronted with any covid-19 related articles or content, when I wasnt looking for it. For example: facebook ad's, email newsletters, tv watermarks or commercials.

SELFSTUDY WEEK
* NOTES
* READING
* SET OF 10 TOPICS

FIRST WEEK OF CLASSES
* MEETINGS
* COLLECTIVE DATABASE
* CONTRIBUTING 50 PIECES OF DATA
* EXCEL EXCERSICE

I choose to collect this because it bothers me that anything now is about the virus, also when it could be unrelated. Like supermarkets and stores make "Christmas like" commercials about eating together and being homey. As well as the more "emotionally loaded" type of tone in commercials that also air around Christmas times.

Pseudocode
excesice to create haiku's out of randomized data

Compare catagories excersice

>>>>>>>>
this button was a quicker way to calculate

Theory text readings and analysis

I think the comparison of the similair taken spot and similair headliner of the same newspaper are very interesting, this also somewhat goes with my corona tvcommercial opinion on how they pick the same type of feelings they do in christmas commercials as well.

I have chosen texts on something I noticed myself, I have been playing every animal crossing since I was 10 years old, but now it has become a "corona" hype, there are shifts in how companies try to hijack the in-game realm. Which I based my later seen draft on aswell.

Week 3
- First concept sharings
- Draft making
- Experiments
- Editing the database

For the first feedback moments with Arjen and Eefje I shared the same concept I had been thinking about.

Refering back to my first 10 corona related topics, the Mayan 2012 end of the world conspiracies and topics reminds me a bit of the current pandemic. And I predict that this will re-appear in some cheesy high budget hollywood movie, like in 2012 there where also 2 similair very typical cinema films

I want to merge the personal tone of the database with the script of a 2012 end of the world movie, and see what could come out of it.

Arjen recommended me the software Runway ML as well the movie Sunspring which is a Sci-Fi scripts genarated and machine learning written short film.

I downloaded Runway ML and saw this pre-trained Film Script Generator. I made a txt file of the database and then copied it into the format. For some reason the sample text came out relatifly similair as what I put in so I retried and watched what I thought would be a similair tutorial but didn't work with a pre-trained package.
When I rendered the file I noticed I was using credits so to keep those I decided to look for more similair generators and do some more research after It generated the same text again

I then manually went to the database txt file and removed all numbers, dates or one word only lines.

I noticed really late that the txt file for some reason cut up a lot of sentences, or had hidden lines under the table structure so I will have to redo this copy pasting each line from the database later on probably, or do a mass edit and keep a selection

RANDOM GENERATORS OR COUNTERS WITH THE TEXT OF THE DATABASE.

In case runway ml doesnt work out as I want it to, talktotransformer.com could be a alternative as it's also a GPT-2 model the only difference is that I wouldn't have to feed it data, and I am not sure what text it's trained on, yet it's more accesiable and also works within my concept so far

This textanalyser has a word limit, per generate, but it calculates word use and frequencies efficiant.

I could possible use this If I make the concept more frequency based

This showed my text file had a lot of duplicates but also If ignore those there are reassurances. And the word online was used more then corona for example

Some more text based converter experiments with
text from the database

text to image generator on a trained model

With the variaty of generators and calculators I got curious what would be the machine and internet calculated conslusion of this database, due to the personal nature and input of the content of the data.

So I decided to use all of the text and manually use every converter I come across until I get to one image, sound etc. And try how "calculative" making comes with a surprising end result. And create a algorithm that makes use of a variaty of existing algorhtyms

ALGORITHM // machine-generated making
START: With a text file

1 resoomer.com > summary

2 miniwebtool.com > remove spaces

3 miniwebtool.com > remove line breaks tool

4 used word counter to work around the character limit on resoomer

5 customwriting.org > summarizing tool with a higher character limit

note: some people use punctuation and other's don't so after this step te text already becomes more generative by computers

repeat step 4 and 5 ten times to implement the whole database set

27 count all the summaries in wordcounter.com > now 33.141 characters
28 use number generator in google chrome to decide calculations for:
29 textweaver tool on languageisavirus.com > number of words per line
30 browserling.com > delete duplicate text converter
31 miniwebtool.com > remove linebreaks again
32 resoomer.com > says text seems already summarized enough
33 google browser number generator for:
34 browserling.com > text truncate
35 talktotransformer.com
36 google browser number generator for transformer amount of talking to the a.i > 4 times

repeat step 35 4 times

39 onlinerandomtool.com/shufflelines > nothing happens

Possible branches with the outcome
Image generation:
40 deepai.org/machine-learningmodel > for each generated line convert 1 image

and Sound generation:
40 Langorhythm

According to text generators and text ai model this is the core of the database

According to image generator this is the core of the database

According to sound generator this is the core of the database

I continuously worked on the website
as I was experimenting

Any notes in this font were added after,
to signify what steps I took with me up
until the algorithm

Set 1: Mayan calender 2012 end of the world prediction and 2012 movies > MIT Computerprogram World One, predicting catastrophic events starting in 2012

The database is full of personal feelings and experiences, often in a negative or lonely tone. By making a "static" and organized file of these feelings the context feels different and transforms in "colder" data then the actual message was.

Next step: researching and testing existing generators, randomizers and online transformers or GPU-2 models. With the new text file.

Next step: by researching and testing more generators, you can see how the original message gets more lost through chosen mediums. By over generating I want to experiment in how far I can generate the data into a new message or surprising outcome, with each step resulting in more substance loss then the intended feeling or message someone wrote

Final Concept
The database is full of personal feelings and experiences, often in a negative tone. But by making these feelings "static" and an organized file, the context of the true feeling or message gets lost and transformed in the applied communication medium.

This is also seen in other "the new normal" changes like online communication, presenting during assessments, range of options when making work, video calls, working more digital, SCC talks or talking with friends and family.

The medium of machinal data or message exchange, shapes and transforms communication, feelings and way of making and most importantly context.

With the variety of generators and calculators I got curious what would be the machine and internet calculated conclusion of the database, and how these will distort the personal tone of the communication and personal datasets.

It's also an experiment for myself to see how with calculative making I can get an outcome I couldn't make myself.

About the algorithm

- I only use pre-existing algorithms and calculators.
- Every calculation that asks for a number I ask in google browser's number generator how many I should generate. > No / to minimum free choice.

In the video you can see steps I made and converters I used, the sound of the video is also generated through the narrowed down text by then using langorhytm which is a text to audio converter

How would I reapply this? POST ASSIGNMENT

I was thinking the text generated, could be applied in a corona chatbot time machine, because for most people this is the first large scale pandemic with a lot of consequences, it could be interesting to use the text in a time machine chatbot where in the future you can talk with someone going through a pandemic

OR

I would make an oracle card deck because to me algorithms have a sense of divination to it, and with machine calculated making this could have surprising outcomes that can be used for personal physical divination as well, this would then also reclaim machine-made algorithms

Society POST CORONA

Because there is no clear calculation or prediction to when we will be in the post corona phase, there probably will be a stage of co-corona living. The Dutch minister of infrastructure and water annouched to also after work mostly from home due to traffic jams etc. I also read somewhere before that in other cultures there are alternatives to handshaking because of previous diseases