| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
SPREAD THE INFORMATION
Any information or special reports about various countries may be published with photos/videos on the world blog with bold legit source. All languages are welcome. Mail to lucschrijvers@hotmail.com.
Autobiography Luc Schrijvers Ebook €5 - Amazon
Search for an article in this Worldwide information blog
zondag 31 juli 2022
#WORLD #WORLDWIDE #FRANCE #RUSSIA #Homosexualité #LGBTQ+ #Gay #News #Journal #Update - La « propagande » homosexuelle punie par la loi en Russie
#WORLD #WORLDWIDE #FRANCE #Elle #Magazine #News #Journal #Update - L'édito de ELLE : « Ces gens-là, c'est moi »
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
#WORLD #WORLDWIDE #USA #LGBTQNATION #LGBTQ+ #Gay #News #Journal #Update - The Digital Cover Story: Call Me By His Name
|
#WORLD #WORLDWIDE #UK #ANARCHISM #News #Journal #Update - (en) UK, AFED: WHAT THE CORONA VIRUS PANDEMIC CAN TEACH US ABOUT SECURITY CULTURE (ca, de, it, pt, tr)[machine translation]
Security culture is fundamentally about collective action to identify and
mitigate risks. Police can read SMS messages, so your crew uses an encrypted chatapp. Local fascists might learn your identity and show up at your home or place
of employment, so you use a fake name and pseudonymous online accounts.
Counter-protesters at a nazi march might get arrested, but the risk of rising
fascism is greater, so you accept the risk of arrest and put your body on the
line. ---- We've picked up many of these OpSec1 tips in a piecemeal fashion as we
joined activist circles. Maybe your first experience with security was being
asked to leave your mobile phone in a box outside the room where a meeting was
taking place. Or maybe it was being told to hide your face from right-wing press
at a demonstration. Many people feel that their involvement in liberatory
movements does not necessitate much security, so they've never developed holistic
security practices. Of the many who do recognize the need for security, these
practices have become so habitual that they happen without conscious thought.
It is rare that there are paradigm shifts in our security practices either
because of changing circumstances or new technologies. These things often happen
slowly and in ways we don't notice until we retrospect over long periods of time.
We don't have the opportunity to see a drastic change and compare how things were
before and after.
In our daily lives we have at least an inkling of a concern about our protecting
our health from disease. We wash our hands after we go to the loo or take our
shoes off in the entryway of our flats. Many such practices can be cultural, and
they are often habits we picked up one at a time but hardly think about.
The COVID-19 pandemic has been a shock to many people who previously were rather
lax about protecting themselves from diseases like the flu or common cold. In a
short period of time, the majority of us went from having no real threat model of
how we might get sick or what the consequences were, to being acutely aware of a
threat that was all around us. Most were unaware of how many died yearly from the
cold, if they even knew it was killing people at all. The new threat to our
health from the corona virus forced us to make massive changes to our perception
of the world and our behavior as we navigated it.
Security is often taught via analogy. Encryption is explained by describing two
parties sending letters to each other via some special kind of post. When we talk
about the security of the individuals in a group who live in disparate locations,
we use metaphors about the physical security of a building. The aphorism "all
models are wrong, some are useful" comes from the statistician George Box who
nearly a decade before more eloquently said:
For such a model there is no need to ask the question "Is the model true?". If
"truth" is to be the "whole truth" the answer must be "No". The only question of
interest is "Is the model illuminating and useful?"
Using the pandemic as an analogy, this text explores how protecting oneself and
others from disease is remarkably similar to developing a security culture to
protect against repression. The descriptions of the pandemic focus on how it was
experienced in Europe. Some of the experiences that will be described are rather
generalizable. Others may be less so as they are the experiences of my particular
radical circles.
There are many definitions for security culture and OpSec, so to provide clarity
I'm using the two definitions here as this is generally how they are used by
activists.
Security culture is the set of norms with a social group or movement designed to
counter surveillance and disruption either from the State or private entities
such as militias, far-right gangs, or corporations.
Operational security is the set of specific practices an individual can take that
reduce surveillance and disruption.
Leaving your phone at home for an action is OpSec. Your affinity group
normalizing the practice that everyone leaves their phone at home is security
culture.
Pandemic Pods and Affinity Groups
Early in the pandemic, we were encouraged to form pandemic pods, or rather closed
social groups. The idea behind these closed groups rather than the naturally
forming webs of human connection was to limit the spread of the disease. If one
person in a pod got infected, the whole pod might get infected too, but it would
be limited to just the few people within that pod. While this practice
complicated by living situations (those with roommates) and working situations
(those who couldn't work form home), it still provided one way of limiting the
spread.
Goals of security culture are to reduce information leaks, prosecutions, and
violence faced by its practitioners. Affinity groups (AGs) are closed groups of
trusted individuals that form in part to carry out political actions together.2
The closed nature is in part due to the threat of informants or infiltrators. AGs
can plan political actions without worrying that their plans will be overheard,
and if they agree to silence, they can carry out these actions knowing that no
one but the members of the AG will know who did it.
Affinity groups are robust against infiltration and State-led disruption in much
the way pandemic pods-when employed correctly-are robust against viral
transmission. Loosely networked AGs exist in opposition to classic organization
structures like political parties or NGOs. Leadership in classical structures
often has an overview of all members including what they are doing, and members
may shuffle between working groups or task forces within these structures.
Infiltrators and informants can rise in the ranks to see everything or float
through the working groups collecting information on everyone because they are
granted implicit trust by merely being a member of the organization. Vanguardist
political groups and so-called "big tent" organizations are the equivalent of
superspreader events in terms of infiltration and information leakage. If the
"disease" is having your details gathered by the police, "infection" spreads
rapidly in these groups. Organizing via AGs helps prevent infiltration and
information leaks.
Defense in Depth
Health measures that have been emphasized over the pandemic include avoiding
in-person social contact, maintaining distance when around others, ventilating
indoor spaces, wearing masks, frequent hand-washing, and the sanitization of
surfaces. There was not just one panacea, but many steps we were told to take.
Each of these on their own contributes to decreased viral transmission, though
some more than others. If an individual enters a shared space with only a
surgical mask, they may still be significantly protected by everyone else wearing
FFP2 masks. If everyone wears only surgical masks, they are once again protected
by ventilation and air filters in the building.
Defense in depth means using more than just one control against a single threat.
Sometimes these controls are additive such as when two individuals wear masks,
the first has a lower chance of being infected by the second than if only the
first or second alone wore a mask. Sometimes these controls are simply redundant
such as disinfecting surfaces, frequent hand-washing, and avoiding touching one's
face while out in public. These controls protect against the same vector, namely
infection via transmitting the virus to mucous membranes after touching surfaces
that are infectious. If someone doesn't wash their hands, keeping surfaces clean
still protects them. If someone breaks the habit of frequently touching their
face, the risk of infection from going long periods of time without hand-washing
is also reduced.
Likewise, good OpSec and security culture use defense in depth to prevent
information leaks or disruption. To implement this, one can ask themself: if this
control fails, is there another that prevents my adversary from achieving their
goals either partially or completely? Because your phone might be seized and
searched, you might have device encryption and a strong passphrase, but moreover
you might enable disappearing messages to reduce the amount of recoverable
material on your device if it is compromised. Maybe all these layers of defense
will eventually fail, but often it's better if they fail after 1 year than 1 day.
If your AG agrees to a code of silence about secrets, you can further prevent
accidental information leaks by not telling them about your past hijinks. One
layer of protection is their silence; the second layer is yours. When making a
plan for your and your AGs' security, plan for many layers of defense.
Overreacting
Another part of the early pandemic was the attempts to counter the skepticism
people showed toward the effectiveness of countermeasures like masking and
avoiding in-person contact. There were trends on social media and statements from
public health officials saying things like "if you feel like you're overreacting,
you're doing it right." Others said things like "if there ends up being no
pandemic, you might feel like you made sacrifices for no reason, but that's proof
it worked." Many people pointed at the Y2K bug3 as an example where disaster was
averted due to what felt like an overreaction.
Messaging that tried to normalize the idea of overreacting was not just about
motivating people to put in the effort themselves, but also to help them overcome
the feeling of foolishness for taking action others might not. Maybe you felt
ridiculous for wearing a mask before anyone else at your local market was, and
maybe your friends told you that you were being paranoid by avoiding cafes and
restaurants when there were only a handful of reported cases. Maybe you told off
your friends for stocking up on non-perishable goods in case they got quarantined
or there were neighborhood lockdowns like in Bergamo4. Avoiding action is often
done to avoid being shamed by others, and these criticisms are often
justifications to oneself. Telling someone off for taking action makes you feel
better about not taking action you suspect you should.
With activism, you don't notice the arrests you prevented or the doxxes that
didn't happen because of good security culture. Maintaining a sustained
high-level of security can feel overly paranoid or like wearing a tin-foil hat,
and of course there are many genuine cases of overreaction ("No one should ever
use mobile phones!"), but often in radical circles, security practices are
dismissed as going too far. A virus can only infect you at the time of your
exposure to it, but data lives forever. An email you sent or security footage can
come back to haunt you years later, and governments can retroactively criminalize
previously legal activity as part of a campaign of repression. The pandemic
aside, masking all the time may be unnecessary, but practicing security culture
is always prudent.
Misinformation
When we think of misinformation in the context of the COVID-19 pandemic, we often
first think of the strong misinformation machine that gave rise to people eating
horse dewormer or claiming that the various vaccines were part of an NWO plot to
put microchips in our brains. Many other beliefs based on various "alternative
facts" also waxed as the pandemic went along like naturopaths pushing urine
therapy. More subtle were the lies that were pushed early on like the claim that
novel corona virus was just "the common cold." Late in the pandemic, even more
liberal sources and radicals themselves were claiming that the omicron variant
"wasn't that bad." Misinformation based on uncertainty is no stranger either.
Early in the pandemic, people decried masking as they weren't sure it was
effective despite corroborating evidence from other regions for other disease
outbreaks. When the vaccines rolled out, there were claims that they were rushed
through safety checks and that they weren't safe because we didn't yet know the
long-term effects, something most prominently seen in the panic following a
miniscule number of cases of blood clots. It was true we didn't know the
long-term effect because of the linear nature of time, but extrapolating from the
billions of other vaccinated humans for dozens of diseases, it was safe to say
that the long-term effects were negligible to non-existent.
While we were able to point at many of these things coming from the
establishment, there was significant misinformation that came from radical
spaces. Like many conspiracy theories, they stood to benefit the person spreading
them and weren't accidental misinformation. Tankies made claims of China's
flawless response forgetting the initial reaction was to arrest the doctors who
reported on what appeared to be a SAR5 outbreak. Pundits and influencers spread
anti-vaxx conspiracies because being vaguely contrarian is part of their brand,
and stirring controversy gains them attention and the associated clout and
donations. Some BLM-adjacent6 cult-like groups and egotistical influencers said
that "white medicine" or "imperialist vaccines" couldn't be trusted and that the
vaccine was just Tuskegee 2.0, thus implying that said groups or individuals were
the only ones uniquely able to spot these abuses and therefore protect their
members or followers.
These corona virus and vaxx conspiracies are difficult to debunk because they
contain grains of truth. The CDC and WHO has made contradictory statements, so
there is reason to doubt what they say. The Tuskegee Syphilis experiment was
real, and medical racism has not gone away. Healthcare is run for profit, and
there are dangerous medical products that haven't received enough scrutiny.
Starting from a position of skepticism or even uncertainty is valid, but failing
to analyze further from this position is harmful.
If we ignore the atrocious security practices of conservative and fascist groups,
we can still see that security is a discipline within activist milieus where
misinformation runs rampant. Some poor security practices are furthered because
of what individuals or orgs are trying to peddle. Live-streamers are defended as
being necessary for the cause ("Good propaganda!") despite only dubious claims of
the benefit of riot-porn and the obvious harm to those arrested because of the
evidence the streamers generate. Large organizations downplay State infiltration
efforts because they rely on the strongman illusion of infallibility or the
belief that absolute numbers is the primary goal of movement building. People who
call themselves organizers will elevate themselves as doing radical praxis the
"right" way by chiding and shaming those who hide their names and faces, and they
take this stance because their "importance" to the movement is dependent solely
on them being a loud voice with a name and a face. Some of the misinformation
is-like medical misinformation-many people simultaneously reaching the same wrong
conclusions without malice as we often see in the smart phone vs. dumb phone debate.
Like how we looked to virologists and epidemiologists to inform our response to
the pandemic, we need to look to security experts to inform our security culture.
Experts may be individuals who work in cybersecurity to teach us about encrypted
chat apps, but just as well this can be seasoned members of the movement who have
lived through occupations, police raids, and repressive legal cases. We need to
agree on the basic facts of the world if we are to analyze it. Misunderstanding
the internet and encryption will lead to poor IT security. Inaccurate models of
how the police and legal system enact State repression will lead to inadequate
countermeasures. To avoid repression, one needs an accurate threat model, and
accurate threat model necessarily requires an accurate model of the world. Do not
let yourself become clouded by dogmatic adherence to your ideology or the words
of your heroes. Beware people who claim "I do this, therefore it is correct."
Seek facts and strongly supported theories.
At-Risk Groups
The corona virus did not affect us all equally. Some individuals were at greater
risk because of factors like their age or medical history. Others were placed at
increased risk due to their working conditions. Those who could work from home
had particularly low rates of infection compared to those forced to work service
jobs. Affluence afforded additional layers of protection like better access to
preventative measures, testing, and treatment.
The threat individuals and orgs face from the State or other malicious parties is
not equally distributed. Some of us have traits we are born with that massively
change how we are surveilled and treated by the State such as our skin color or
what passports we hold. Others develop traits over time-like blossoming queerness
or a radical political position-that attract scrutiny from the State or
conservatives at large. Like with preventing infection, money plays a roll in
preventing repression. More expensive electronics often provide better security,
and being able to throw clothes in the bin after a risky action is a luxury.
The early pandemic placed a strong emphasis on doing one's part to protect
at-risk groups. It leaned in to the sense of altruism many of us have to help a
neighbor. We all masked up and stay home to save our nans and the
immunocompromised. It worked, and almost too well as many people who weren't in
at-risk groups believed they had no risk.
Security functions in much the same way. If only the most active radicals hid
their activity, they would stand out from the crowd. Good security also functions
by obfuscation. Poor insight into groups means surveillance needs to be deployed
against more targets and infiltrators need to target more groups. You may feel
that you're not personally at-risk, and you may think-wrongly-that no one in your
circles is at-risk, but adopting stronger security practices helps provide cover
for those who most critically need to avoid disruption, and furthermore it helps
protect you.
Crisis Fatigue
Many people who took the pandemic seriously from the onset found themselves
unable to maintain their precautions as time went on. The routine of hand-washing
upon returning home may have become hand-washing "only if I think I touched
something gross." Consciously avoiding indoor gatherings may have slipped into
justifying going out to a friend's no-mask birthday party. This phenomenon has
been called "pandemic fatigue." Sometimes people stop following preventative
measures because they simply don't want to or find them fucking annoying. Other
times it's the belief that they earned a cheat day by being so good for so long.
Some people see the pandemic as never-ending and can no longer rationalize giving
up so much for so little perceived benefit.
I'm not claiming that our FOMO7 is baseless. We've all given up on attending
weddings or funerals, hosting birthday parties, or traveling to see friends and
family. How many regular social activities like hanging out at home, going to a
favorite pub, or spontaneously catching a movie have we missed in the last two
years? It's not even just the voluntary measures we've taken, but seeing that
others aren't taking them, so why should we suffer while they go out and party?
Maybe it's even seeing the intentionally poor response of the ruling class that
makes so many of our individual efforts seem worthless.
But just because we're tired and burned out doesn't mean the pandemic is over.
This winter has been particularly harsh for many of us: a second winter into the
pandemic after the summer that wasn't, all with the backdrop of pandemic-related
scandals, soaring infections, and sustained fatalities.
OpSec fatigue is just as real. Nearly every security measure we take has some
cost. Abandoning WhatsApp after they changed their privacy policy meant losing
out on communication with those who hadn't switched to Signal. Leaving phones at
home for actions complicates coordination. Using multiple devices for multiple
aliases means more shit to constantly lug around with you. Refusing to organize
direct actions with people who have poor security practices leads to conflict in
meetings or solitudinous actions.
Because of these costs-either material or "merely" mental-people often relax
their security standards, but the threat of surveillance and disruption do not
disappear because we lose interest. The longer one is in radical movements, the
more they tend feel the pressure of external threats. This stress can make it
hard to maintain the desired level of security. There's also the mistaken belief
that actions taken during periods of relaxed security not leading to arrest is
proof that the reduce security is still secure enough. With the corona virus,
many people have been burned out and resigned saying "we're all gonna get it"
before giving up many of their precautions. Many activists develop a view known
as security nihilism which amounts to believing that no amount of security can
prevent repression, so why bother with any of its encumbrance?
Crisis fatigue is a hard problem both for the pandemic and for security, and I
cannot pretend to have a clean solution. My own experiences and those of others
suggest that at least OpSec fatigue is countered by a stronger security culture.
If all your friends wear masks and only make plans outside, it's easy to go along
with them. Likewise, if we have each others' backs with our security, the small
slips we make are more easily corrected. It's also easier to be secure when
everyone around you is too instead of constantly fighting to just barely attain a
low baseline of security.
Prevention, Not Cures
Even where there is widespread deployment of the corona virus vaccine including
3rd booster shots, there are no specific and effective treatments for COVID-19.
Care is supportive. Individuals can keep themselves fed, hydrated, and rested to
help improve their ability to fight the disease and recover. Even when receiving
intensive care, much of the treatments patients receive is not to eradicate the
virus but to fight its effects on the body. Recovery from "mild" COVID-19 can
take weeks, and the individual may develop temporary or permanent disabilities as
a result. Moderate and severe cases require costly treatment at specialized
facilities with limited capacity.
On the other hand, prevention is simple, cheap, and does not require specialists
to be effective. Even the cheapest surgical masks have a marked effect on
reducing viral transmission, and skipping the pub to take a walk through the city
is free.
There is always some cost to prevention. Masks, even reusable cloth ones that are
washed daily, cost money. Bulk hand sanitizer also has some cost. Much of the
cost is psychological. Avoiding gatherings takes a mental toll, and isolation can
lead to depression. One can get fatigued by always asking "is this safe?" or
foregoing desired activities. Even if the chances of getting the disease is low,
people may weigh these preventative measures as being too costly and accept what
they see as tiny a risk of life-altering outcomes.
Security is also a case where an ounce of prevention is worth a pound of cure.
The bother of getting all your contacts to switch to an end-to-end encrypted chat
app might have some upfront and upkeep costs, but this effort is drastically
lower than the response necessary after being prosecuted. Dressing in Black Bloc
at actions and having clothing to change in to when traveling to and from the
action can be bothersome-especially in summer's heat-but this is a small price to
pay compared to the damages of being arrested or doxxed with possible subsequent
stalking and harassment. Every arrest that is prevented means more time legal aid
can spend on other cases. Every comrade who doesn't have to move flats because of
fascist harassment is funds that can be redirect to the community. Every
imprisonment that is avoided means a healthier community that isn't mourning the
abduction of a comrade and exerting effort for appeals or prison support.
Like we saw early in the pandemic, one of the goals was to "flatten the curve."
If it was assumed that some fixed cumulative number of people would require
hospitalization, it would be better for that number to be stretched out over 1
year rather than 1 month. If there was even some Universal Truth that all
activists had some fixed chance of imprisonment, it would be best to spread this
over a longer time thus giving them more time to act before being taken off the
board and more time to gather resources to support them. This certainly isn't the
case, but the goal should be to prevent and delay consequences of repression as
long as possible.
Even when looking at the State level, many of the justifications for avoiding the
upfront costs were to avoid economic downturn from lockdowns and missed work. The
result of poor containment was that these consequences happened anyway. A lack of
security culture may have a short term benefit of alleged increased effectiveness
("Better recruitment and more reach!"), but the downside can be disruptors
ravaging a scene in ways that require great expenditures to rebuild. The benefit
of a strong security culture outweighs the near-term costs of developing it in
the first place.
Security Theater, Expectations, and Nonadherence
Some initial measures taken to slow the spread of the virus were ineffective
because they were based on poor or missing information. Individuals who cared
about managing the spread of the virus wanted to quick solutions, and often this
was in the form of superficial measures that gave feeling of effort rather than
tangible results. Policy makers wanted to show strength and that were were doing
something, so they did anything at all. Ineffective countermeasures that give
merely a feeling of security or action taken are called security theater. In the
case of the pandemic, we saw this with use of chin-strap sneeze guards rather
than masks, or the use of disposable nitrile gloves when out in public yet the
wearer still constantly touched their face. Other larger instances of security
theater were spraying disinfectants into the air from vehicles or requiring masks
while walking through sparsely populated public parks.
Security theater in its original sense is often present as a significant
influence on security culture. Often this is done by either intentional or
accidental conflation of "feeling unsafe" and "being in danger." Safe spaces-in
the sense that they feel safe regardless of risk-are often needed for healing,
and this is not decrying them, but inaccurately naming threats for many spaces is
security theater. Queer spaces that don't allow "straight" partners as a means to
keep out the abusive cis-hets is security theater (because queer people can be
abusers too, obviously). Radical spaces that shun people who "look like cops"
(i.e., middle-aged white men) likely have a very high false positive rate, and
the fact that there is enough "diversity" in local and federal police means that
infiltrators could be of nearly any demographic. Security theater also happens
through misunderstanding such as telling someone to use a VPN to avoid being
tracked on the internet8 or to use a dumb phone to prevent location tracking.
As corona measures became normalized, those who cared about minimizing their risk
would look for clear signals of the measures. This might be be presence of signs
requiring masks before entering a building, or conversely avoiding establishments
that forbid masks. In this sense, declarations of adherence were not just
invitations for other risk-avoidant individuals, but were deterrents for risky
individuals. However, as time went on, these posted signs and online
announcements became less meaningful as they were not always voluntarily
enforced. Events might loudly claim to require masks plus proof of vaccination
plus a test, but then totally fail to check those or even if one was showing up
at their booked times.
Security culture often suffers from a similar lack of enforcement. Posted bans on
photography in a radical space are meaningless unless people actually stop others
from recording. Actions that are declared as phoneless still occur if some
participants bring phones because "fuck it, they're already here." People
loudmouthing on social media are not ostracized when they endanger others. It's
not enough to declare an ideal. The ideals need to be enforced by some means or
another.
The social cost of nonadherence to corona measures can be rather high. It can
mean exclusion from a group activity such a requiring everyone to be tested
before attending. Many vaxx-"skeptics" or anti-vaxxers admitted to feeling shamed
about their position, so they started lying about it. When dating-or more
generally, when meeting new people-the threshold for expected adherence tends to
be higher. In much the way that partners lie about their condom use or STI
testing frequency, telling a new person "yeah, I'm safe with corona" and listing
a few key phrases has become a very noisy signal as many people know what they're
expected to say even if don't actually do these things. Moreover, what two people
consider safe can greatly vary. Making someone name what they do exactly
including specific activities is considered invasive and awkward. Clear
communication would allow two people to have accurate knowledge of each other's
measures which might mean that the person being lax about theirs has to face
immediate consequences for them. Lying or deflecting is manipulative and violates
consent, and it endangers the other party who is trying to minimize risk.
When groups practice security culture, there are too few conversations about the
exact nature of the security model they are operating on. There is not a shared
language, and simply saying things like "we practice anti-repression" or "we're
secure" can be meaningless unless the specifics are discussed. Individuals who
have poor security practices or have risks often don't declare them. Addiction is
one such case that can lead to sloppy practices or exploitation by the State9,
and its presence is often hidden for complex social reasons. Other behaviors that
have been security risks are relationships (sexual-romantic, platonic, etc.) with
people on the right, and while this is unconscionable for many reasons, I'm
flagging it as it specifically is a massive security risk to one's comrades that
is kept hidden for personal gain.
To avoid the harms-both accidental and manipulative-from poorly communicated
security practices, we need to normalize explicit discussions of security models.
Abusers and the selfish will always lie, but this at least gives a starting
point, and as it's said: trust but verify. As the need for security increases,
often as a result of the activities one's group carries out, these discussions
need to become more frequent. The level of detail needs to increase, and the
claims need to be more strongly verified. Being a loudmouth on social media with
a hot temper may be fine from a security perfective for a local Food Not Bombs
chapter, but it may be unacceptable for a tightly knit affinity group that
carries out direct actions. Groups that carry out high-risk activities may want
to recede from the scene at large and implement a ban on individuals taking other
actions because of the possibility of an arrest for tagging a building could draw
far too much attention to the other more secret actions. Clear and explicit
communication, and the trust it is built on, are necessary for effective
security. Like with the pandemic, we've had to (temporarily) cut people out of
lives for poor adherence, and our affinity groups may have to do the same for
those who lie about or fail to practice security culture.
Throughout the pandemic, we have seen that we have shifting models of what
constitutes risk, and we have seen the threat landscape rapidly change. First
there was the virus, then the vaccines, and next the viral variants (and there
will be further changes too). Some people may have had kids start school which
created a new vector for infection, and others have had life changes that caused
a shift in their individual threat model while others saw no change. These shifts
require us to re-model our risk to counter it.
By looking at how we reacted and created safety for ourselves and others-or at
the very least Reduced risk-we can see patterns in group behavior that mirror
group behavior with respect to security culture. Learnings from the pandemic are
more concrete despite the somewhat abstract nature of the virus (one can't see
it, and maybe one's friends all remained uninfected, but the threat is there).
These observations and lessons about modeling and reducing risk can be applied to
security culture. Many of the poor responses to the pandemic have analogues to
poor responses to repression.
Like how viruses spread through populations, so too spread the harms of
repression. The pandemic has rising and falling waves of infection that affect
the population unevenly, and we see the same with waves of repression. Zero-risk
of infection by the corona virus is near impossible as is zero-risk of disruption
by State and non-State actors. Using many tools, such as studies or analogies,
one builds a threat model for themself and their crew, and through this model
informs a security culture that counters surveillance and disruption. "We keep us
safe" applies to our health and our liberty. This mutual regard for one another's
well-being is fundamental to security culture and is the base for effective and
enduring radical movements. ?
Written by Håkan Geijer
Illustrations by zer0coil
Downloadable A4 Reader and A4 Imposed .PDFs:
What-The-Corona-Virus-Pandemic-Can-Teach-Us-About-Security-Culture-2022-A4-Reader-Hakan-Geijer-2Download
What-The-Corona-Virus-Pandemic-Can-Teach-Us-About-Security-Culture-2022-A4-Imposed-Hakan-Geijer-1Download
1 - Operational security.
2 - Sometimes these are called cells, but that term is used somewhat derogatorily
and has connotations of terrorism or militancy. Affinity groups can form for many
reasons.
3 - A problem where computers used two digits to represent the year so that 00
made 1900 and 2000 indistinguishable. There was minimal damage from this bug
because of the large amount of effort experts put in to correcting it.
https://organisemagazine.org.uk/2022/07/25/what-the-corona-virus-pandemic-can-teach-us-about-security-culture-theory-and-analysis/
_________________________________________
A - I N F O S N E W S S E R V I C E
By, For, and About Anarchists
Send news reports to A-infos-en mailing list
A-infos-en@ainfos.ca
Abonneren op:
Posts (Atom)