resources for the chatbots presentation on Tuesday
- artists bots (Piero)
- some elements of the history of chatbots (Piero)
- critical approaches of chatbots (Anne)
- bot that passed the Turing test - goostman (Anne)
Links sent by Marie Lechner http://pad.constantvzw.org/p/LiensML
critical approaches of bots :
https://www.incapsula.com/blog/bot-traffic-report-2015.html
https://en.wikipedia.org/wiki/Loebner_Prize
https://howwegettonext.com/the-bot-power-list-2016-d9b40ea4ce0a#.9qh1yhfl3
https://howwegettonext.com/if-you-talk-to-bots-youre-talking-to-their-bosses-cd8e390c242f#.bchjbtg0z
Notes 07/09/2016
Presentation on Contemporary Bots by Anne
A bot , also known as Internet bot, is a program that runs automated tasks over the Internet. Typically intended to perform simple and repetitive tasks, Internet bots are scripts and programs that enables their user to do things quickly and on a scale. src : https://www.incapsula.com/blog/understanding-bots-and-your-business.html
In the book "Bot, the origin of new species in the late 90s", bots define as a small intelligence program with a personality performing an online service.
Today bots much more sinister vs playful.
Bots proliferate in service where one could expect humans.
Feeling cheated.
See :
https://www.incapsula.com/blog/bot-traffic-report-2015.html
Global Bot Report Traffic in 2015 that makes distinction between human, good bots & bad bots.
- 51.5 % humans
- 29 % bad bots
- 19.5 % good bots (among them scrappers, spammers, hacking tools, impersonators/malicious bots)
- Why scrapper bots are bad bots ?
Net Neutrality VS Cyber Criminalities.
Decrease of good bots due to less RSS feeds use.
Decline in Spam bots too related to Google's Penguin algorithm for spamdexing in 2012.
Loebner Prize annual competition in AI ; the "Loebner Test".
https://en.wikipedia.org/wiki/Loebner_Prize
Eugene Gootsman is a chatter bot portrayed as a 13 year old Ukrainian bot.
https://en.wikipedia.org/wiki/Eugene_Goostman
If you talk to bot, you're talking to their bosses.
https://howwegettonext.com/if-you-talk-to-bots-youre-talking-to-their-bosses-cd8e390c242f#.4zcem3yho
2 new trends emerging :
- First, the sophistication with which bots interact is increasing. They are able to respond to (or more accurately, be triggered by) a greater variety of situations.
- Second, the mode by which they frame that interaction?—?how the machine addresses you?—?is also shifting. (...) The new breed of bots, however, are dropping non-personal words like “the” in favor of mysterious first-person pronouns like “I” and “me,” or possessive pronouns like “my.”
Bots as assistants => as "butlers"... well usually view as female (Gender issue reproduced also in the world of virtual beings).
Bots are there to serve you BUT they would never be your "friend" :
"The one persona that is likely to always be missing is the Honest Bot, the one that clearly tells you its agenda, like a true friend who drops their façade and lets you know their dark secrets. (...) The Honest Bot does not currently exist. The new wave of bots are?—?to use a term popularized by the existentialist misfit Holden Caulfield from The Catcher in the Rye?—?the ultimate phonies. They’ll pretend to be friendly, to be cool, to be serious, to be insightful, and even to be self-reflective. But they’ll never just be themselves. Because, in the end, there is no “I.” There is only a company, its shares held by who knows who, possibly registered in Panama."
Demystifying IRC by James
Hexchat, Lime, Konversation, Pidgin, Adium.
Internet relay chat
IP protocol : packets
From UDP, broadcast protocol to TCP protocol.
With worldwide web : HTTP & HTTPS.
$ nc -ul 127.0.0.1 -p 8000
irc built on top on irc protocol
exercice : telnet to irc
telnet 10.1.1.1 6667
NICK xxxx
USER xxxx 8 * : xxxx
JOIN #frankenstein
PRIVMSG #frankenstein :hello world
changes of IRC
EFNet 1990 - way to establish nick name
Anarchy net (A-net) wanted to stay wild ....
"the great split" of IRC : arguments about the roles of OPs (operators) : US vs Europe. freenode.net on the european state
why the choice of IRC for making bots ?
you can tweak it. stitch things together and create one's space.
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Draft of presentation:
ART BOTS HISTORY AND ARGUMENT ABOUT THEIR EARLY CONNECTION WITH SPAM
I would like to rapidly speak about the history of Internet bots and how artists entered this field.
Before I begin, I would lie to define how I am going the use the word bot in this presentation.
When I speak about a bot here, I refer to a computer program written to automatically carry out some tasks within a network.
Even though bots are as old as the networks where they were put to work, the dramatic spread of them is linked to spam.
I will argue that it is because of spam that artists did or did not see bots as a potential medium for their expression.
As some of you might know, the term “spam” spread in the early 90s on Usenet, a decentralised network of computers that preceded the World Wide Web.
It is here that the first commercial message was automatically sent to a significant number of users.
It was an ad for consultancy services related to the US Green Card lottery.
[show example]
Many instances of automatically sent messages targeting large numbers of users (whether on Usenet, BBSs, IRC, email, etc) can be found even before this example.
There are two reasons why I mention commercial use of bots and the Green Card lottery spam as the real game changer.
The first is purely quantitative: only the commercial potentiality of bots made their presence dramatically escalate in these early 90s networks.
The second is that the commercial aspect of 90s bots in mailing lists and other communication networks was exactly why early net artists, a community known for their political engagement in anti-capitalism movements, regarded them as simply annoying noise.
This bad reputation of bots prevented artists from seeing their artistic potential, which was never really expressed the way the same internet artists did with the website-form for example.
In this sense, the history of internet bots made for explicitly artistic purposes remains somehow incomplete.
Even if it is true that some “purposeless” bots were made in the early times, these were never the work of artists but rather of software engineers who wanted to make their work environment more fun.
If we take IRC for example, the first bots were born just a few months after the protocol was.
It’s the case of [PUPPE] + [BARTENDER] + [GAMES], all dating back to ’89-’91.
[show PUPPE, BARTENDER, GAMES]
This is not to say that artists were not using IRC.
Just like mailing lists and forums, many artists found it natural to build communities around these communication systems, but still somehow considered any automated user as pollution of this environment, even if positioned as a piece of art.
This attitude is even more strange when we consider how much the first net artists cared about fictional identities and characters.
In their unofficial net.art manifesto called “Introduction to net.art”, Alexei Shulgin & Natalie Bookchin even list Fake Identity Construction as one of the typical genres.
[show Introduction to net.art]
Despite all, the intuitively natural step from fake users to automation and artificial intelligent was almost never taken.
Along the bad reputation of bots in those days, more technical reasons for this missed artistic path were scarcity of bandwidth and strict self-control of communication systems used by artists (mainly mailing lists).
With regard with bots, spam and artistic communities online in the mid 90s, the interview between net art theoretician Josephine Bosma and net artist Frederic Madre from 2000 is illuminating as it is controversial.
Madre says: “people wanted to throw you off list if you sent mail bigger than 2k” and continues “I think the often rigid measures providers take against spam [are] basically a fascist thing” and “netiquette laws […] were unjustly enforced by people that pretended to be on 'our' side, you know, for our own good”
Stemming from this polemic view, the artist created a Spam Art Machine in 1999, which can be regarded as one of the first examples of a consistent, automated, explicit artistic intervention within a mailing list.
[show spam machine]
If we were to strictly follow the definition of bots given at the beginning of this presentation, it would not be entirely correct to regard Madre’s Spam Machine as one instance.
In fact, Spam Art Machine was not a fully automated net artwork since the artist (or user) still needed to activate the bot manually.
Spam Machine was rather a web interface that allowed anybody to act as a bot.
If not in strict sense an art bot, Spam Machine got close enough to what some artists begun considering part of their practice many years later.
Fully automated bots created for artistic scopes had to wait for the web 2.0 and contemporary social networks to see the light and be accepted as organic part of an online experience.
To understand this delayed appearance, we could start by mentioning the disappearance of what prevented art bots from being developed in the 90s.
First of all, bandwidth stopped being a problem (at least in Europe and North America), and the concept of netiquette was either forgotten or subcontracted to the corporations that would run the web.
The concept of spam radically changed as well, with commercial offers directly managed by the giant internet platforms through ads.
At the same time, new types of usage and the technical differences between 90s online communication systems and contemporary social networks made the latter ones the right environment for the spreading of art bots.
Briefly listed and analysed:
NEW as opposed to OLD
- individual usership as opposed to enclosed communities (art bots can reach any other user)
- chronological feed-based interface as opposed to separated forum threads (art bots can be generic)
- infinite quantity of content as opposed to limited quantity of content (art bots don’t distract as they can be overlooked)
- variety of users as opposed to homogeneity of users (bots can reach diverse audiences)
- self moderation as opposed to top down moderation (bots can be unfollowed by users as opposed to be silenced by moderators)
For those of you familiar with today’s social networks, it must appear clear from the characteristic just listed that the today’s choice for artist to develop their bots is in fact mostly restricted to Twitter.
For both its ecology of users (large quantity of coders) and the technical facilitation (very laissez faire policy regarding bots in general), Twitter has proven the quintessential habitat of today’s art bots.
Twitter bots have been a real fad of the last two years, to the point that exhibitions were held to present them to a more offline visual art audience.
[list some]
Also media coverage has been extensive.
[show some articles]
Despite the recent success, art bots on Twitter are not a recent phenomenon and most people regard @everyword (2007) as the first instance.
[show everyword]
I will then go through some of what I consider the best art bots on Twitter from the last two years.
[list of best 5 Twitter bots of the last two years]
To conclude, I’d like to give the example of more complex recent art bots that because of the technical and conceptual limitations of social networks had to find their space of existence within the website form or even break into the physical world.
[examples from James Bridle]
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////