[-empyre-] Welcome to Week 3 on -empyre-: Internet/Online Representation

[B. Bogart]

Hello Anna,

Thanks for the clarification. I had in mind "talking down to me" versus
"talking me down". Such a subtle change of words with such a huge
difference in meaning.

Indeed recommender systems are strictly designed to provide options the
user may like based on what they have showed interest in / previously
chosen. The behaviour / prediction formalization I described is a more
general case. Predicting what we would like is technically similar to
predicting what we will do.

I agree that these systems are not value neutral. Thanks for the
article, I was largely familiar with these systems in the abstract
rather than in terms of the details. I was struck by one detail that I
had assumed recommender system's did, but don't seem to: model the
similarities between products in order to learn not only what product
the 'customer' is expected to want, but more importantly why they would
want that product. If the systems learned this, then they could provide
suggestions for items before they were ever bought/rated. In short, it
seems the current recommender approaches are much more about the
customer and not very much at all about the products. I expect this
addition is an active research topic.

The section in the article on "Customer clusters" is interesting in
relation to "identity clusters" and the relation between individual and
group identities. I think it is interesting and useful to consider
traditional and accessible media representations in cinema and TV along
with the veiled models and clusters of identity used in social media and
online commerce systems. The traditional media that becomes our shared
culture, by virtue of its accessibility, can be much more easily
reflected upon than the veiled systems that may have serious cultural
impact.

You wrote: "Your remark also that most of our engagements online are in
'private spaces' not 'public' ones, thereby eroding the 'commons' is
generative. And, I think it is partially true because social media is
both public and private..."

I'm thinking of private and public here not in terms of access, but in
terms of legality. A private space (like a mall) has different rules
around privacy than a true public space. We visitors have the right (in
some jurisdictions) to take photographs in public space, but not in
private space. In private space, the rights of the visitor can be
constrained by the owner of that space.

I think the problem is that most online spaces (twitter, facebook, etc.)
appear to be public because they are accessible, but are actually
private as there are constraints on the rights of the users determined
by the owners of the space. I would go so far to say they are not public
in any significant way, they only appear to be public because of our
ease of access, and the scale.

Thanks for the quick identity theory primer!

Ben







On 15-04-22 07:40 AM, Anna Everett wrote:
> ----------empyre- soft-skinned space----------------------
> Hi Ben,
> 
> You may have succeeded in  "talking me down" from the "ledge" of
> indignation a little bit. This is a GOOD thing! It means that you have
> provided me a different lens through which to view my ongoing
> perceptions of what I prefer to call "identity mining" systems, in
> this case. For me, these systems track and thereby recommend sales
> possibilities (and, I ask, "what else is being sold to us--cute cat
> videos on YouTube, based on past viewings; political actors, based on
> whom you followed on Twitter, etc?"). So, yes, you have provided a
> "think piece" for me as I collect images of this tracking of my
> identity online.
> 
> Your comment:
> 
> "I think the majority of our behaviour is habitual and not often
> critically reflected upon. The notion of a system learning these
> habits could certainly encourage a good user experience, good meaning
> efficient and 'natural'. At the same time it seems that a system that
> nudges your habits in incremental ways could be a very insidious
> method of engineering behaviour, and in the context of this month's
> theme, even engineering identity."
> 
> This is rich. As I said, the Amazon recommender system/algorithm can
> and does offer useful benefits (you like that, so you might like
> this). However, such digital panoptic processes suggest identity
> matters that just might warrant our closer attention as I doubt there
> is a value neutrality in such data mining. See this article,"How
> Recommender Algorithms Know What You'll Like."
> <http://www.techradar.com/us/news/internet/how-recommendation-algorithms-know-what-you-ll-like-1078924>
> 
> You also commented:
> 
> "I have not read a lot of theory on identity. Are identities
> considered unmediated, or are they always mediated by a system of
> representation? i.e. is an identity a representation of self, or a
> sense of self? (apologies if this has already been discussed, I did
> not catch all the emails this month)."
> 
> Your remark also that most of our engagements online are in "private
> spaces" not "public" ones, thereby eroding the "commons" is
> generative. And, I think it is partially true because social media is
> both public and private and we enact variations of identity
> performance and politics of the self (and selfie) on Twitter,
> Facebook, Instagram, Vine, Meerkat, and others. But to address your
> question on theories of identity, briefly, it is the case that
> spectatorship theories in media studies and in cultural studies do
> help us understand how individual and group identities tend to be
> mediated through such representational systems as film, TV, video
> games, and other visual culture regimes. Here psychoanalytic, critical
> race, postcolonial, feminist, womanist, reception theories enable us
> to interrogate our relationships, affect, and suturing to particular
> types of identity representations. So our identities are imagined and
> experienced as individual and collective. For example, for myself, as
> an African American woman, I might avail myself of transracial
> identification strategies (willing suspension of disbelief, and trans
> race ego-ideal) to find pleasure in dominant Hollywood films (Manthia
> Diawara and Laura Mulvey), TV, and now online content. For we know how
> these foreground whiteness as normative, heroic, beautiful and
> empowered. Cultural studies enables me to view western films and
> resist their preferred reading of the narratives and thus identify
> with the "enemy" -- indians, or postcolonial theory  to identify with
> the "savages" in Tarzan films, women in action films/TV, etc. Of
> course, Fanon, Stuart Hall, Hazel Carby, and even our panelists:
> Andre, Kishonna, Aphid, and myself, as well as Lisa Nakamura, Tara
> McPherson, Adeline Koh, and many others address such ideas for
> contemporary critical and theoretical demands and approaches in the
> digital era.
> 
> Anyway, it has been a pleasure interacting with you, Ben, and fellow
> panelists, Kishonna, Aphid, and Andre, and you too, dear moderator,
> Soraya.
> 
> Anna
> 
> On Tue, Apr 21, 2015 at 10:05 AM, B. Bogart <ben at ekran.org> wrote:
>> ----------empyre- soft-skinned space----------------------
>> Hello Anna,
>>
>> I had no intent whatsoever to "talk down" at all. Please let me know
>> off-list how that came across and how I can avoid it in the future. I
>> can imagine my interest in using somewhat plain language and making few
>> assumptions may appear over-explanatory. I hope I did not misunderstand
>> your message.
>>
>> My main point regarding identity was not so much the aspect of identity
>> as caricature, but more the notion of whether unconscious
>> characteristics and behaviours are part of identity and what it means
>> for systems to learn those without our explicit consent or knowledge of
>> the process.
>>
>> Indeed there is a cost of admission to many online systems, often in the
>> form of ToS, as you state. This is consistent with the sense that much
>> online discussion happens in 'private spaces', rather than open public
>> spaces (online or offline). Part of the ongoing erosion of 'the commons'.
>>
>> I still need to think about this more, but I think there is something
>> different in agreeing to terms (allowing sites to collect our data) and
>> the use of that data in order to facilitate particular behaviours.
>> Perhaps that is the purpose of any product, not to fulfil a need, but to
>> reinforce an insatiable need.
>>
>> I think the majority of our behaviour is habitual and not often
>> critically reflected upon. The notion of a system learning these habits
>> could certainly encourage a good user experience, good meaning efficient
>> and 'natural'. At the same time it seems that a system that nudges your
>> habits in incremental ways could be a very insidious method of
>> engineering behaviour, and in the context of this month's theme, even
>> engineering identity.
>>
>> Thanks for the link to the article, I had not come across it. I recall
>> reading something about self-objectification leading to a shift of the
>> sense of "pleasure" being less embodied and more related to how one
>> looks to an external observer. A conflation of what feels good and what
>> looks good? This does seem to indicate the media representation of
>> identity (selfie) is conflated with the subject (self).
>>
>> I have not read a lot of theory on identity. Are identities considered
>> unmediated, or are they always mediated by a system of representation?
>> i.e. is an identity a representation of self, or a sense of self?
>> (apologies if this has already been discussed, I did not catch all the
>> emails this month).
>>
>> Again, I apologize for coming off as 'talking down'.
>>
>> Ben
>>
>>
>>
>>
>>
>> On 15-04-21 09:02 AM, Anna Everett wrote:
>>> ----------empyre- soft-skinned space----------------------
>>> Ben, while I am not sure I have been "talked down" completely, your
>>> comment provides fodder as I formulate more clearly my concerns with
>>> recommender algorithms where race matters (not to dredge up that
>>> current Cornel West and Michael Eric Dyson unfriending public split)
>>> online identity, and ideology collide.
>>>
>>> You remarked: "Does identity as a model become a caricature? In order
>>> to be socially engaged online, do we unconsciously internalize these
>>> caricatures and enforce them in order to fit the technologies we use?"
>>>
>>> As I was contemplating your response to me, I was thinking that it is
>>> not necessarily that I/we internalize the caricatures and enforce them
>>> to fit the technologies, but more it has become the price of our
>>> internet participation ticket, like accessing any software update. You
>>> don't get to "discuss" or challenge the terms of agreement, you ether
>>> "agree" or you do not get updated. And besides, who reads the
>>> impenetrable legalese intended to ensure your non-reading of the terms
>>> of agreement. Anyhow, apropos of your observation, I found this Huff
>>> Post article quite pertinent:
>>>
>>> "How Facebook Stalking Could Lead Women To Objectify Their Own Bodies"
>>> <http://www.huffingtonpost.com/2015/04/20/facebook-stalking-makes-you-feel-bad_n_7100832.html?ncid=tweetlnkushpmg00000067>
>>>
>>> Again, Ben, thanks for joining our discussion week.
>>>
>>> Anna
>>>
>>>
>>>
>>> On Mon, Apr 20, 2015 at 3:15 PM, B. Bogart <ben at ekran.org> wrote:
>>>> ----------empyre- soft-skinned space----------------------
>>>> Hello Anna,
>>>>
>>>> Your observation on the apparent identity of target marking materials is
>>>> very interesting. I've been using an add-blocker and opting out of paper
>>>> mailing lists for so long that I have not even thought about how a
>>>> system's perception of my identity could be used to influence the
>>>> appearance of people in ads. I rarely see targeted ads at all.
>>>>
>>>> What concerns me is how this process is driven; what online behaviours
>>>> do we exhibit that were noticed by the system and taken as indicators of
>>>> gender, ethnicity, etc.?
>>>>
>>>> Essentially, these recommender / target marketing systems model our
>>>> behaviour in order to predict and facilitate behaviours that are
>>>> favourable to the context of deployment. These systems may (likely do)
>>>> capture unconscious aspects of our behaviour that are useful in
>>>> predicting our behaviour.
>>>>
>>>> I wonder if this bypasses the very notion of identity as being an aspect
>>>> of us in some way. We continue on to act as we do, both intentionally
>>>> and unintentionally, and the systems we interact with increasingly
>>>> construct a model of our identity. This model may even be at odds with
>>>> what we think of ourselves, and yet be functional in the sense of
>>>> predicting our behaviour. Add to that the notion of the model being the
>>>> property of the organization collecting the data, and there is a
>>>> potential for the organization to not only know us better than we know
>>>> ourselves, but also own (and be able to trade) that knowledge.
>>>>
>>>> Does identity as a model become a caricature? In order to be socially
>>>> engaged online, do we unconsciously internalize these caricatures and
>>>> enforce them in order to fit the technologies we use?
>>>>
>>>> Ben
>>>>
>>>> On 15-04-20 12:32 PM, Anna Everett wrote:
>>>>> So, what got me thinking,
>>>>> and I may need you and the other co-panelists to talk me down here,
>>>>> why are nearly all e-commerce ads being pushed to me featuring black
>>>>> women models? I began noticing that occurring in print catalogues from
>>>>> department stores. It is as though there is the black and white
>>>>> consumer versions of ads.
>>>> _______________________________________________
>>>> empyre forum
>>>> empyre at lists.artdesign.unsw.edu.au
>>>> http://empyre.library.cornell.edu
>>>
>>>
>>>
>> _______________________________________________
>> empyre forum
>> empyre at lists.artdesign.unsw.edu.au
>> http://empyre.library.cornell.edu
> 
> 
>