Leaked memos Amazon warn 'be vigilant' due to threats to blow up data centers
businessinsider.com
businessinsider.com
https://twitter.com/JohnPaczkowski/status/134811382832466739...
"AYO ITS 2021. LETS KILL TRUMP" posted 8:04 PM · Jan 11, 2021·Twitter for Android just minutes ago
Another one. https://twitter.com/rowdon_ella/status/1348691401765224451
Just search "kill Trump" you only need to scroll a bit before you see threats to murder the president.
Maybe Twitter should be changed in some dramatic way if they can’t keep these calls for violence off the platform.
That being said, I can't think of any good solution if you'd like to keep that kind of stuff of your platform. I think it's basically impossible. The only solution that could deal with that amount of posts would be AI. But the problem with AI is that:
1. It's imperfect and generates false positives. There were stories about people being locked out of their Google accounts for no reason, without any possibility to appeal, because their support is also ran by AI.
2. It can get too perfect and thus extremely dangerous in wrong hands. (I think it's already in the wrong hands, but you get the point)
What's also not fine in my opinion is pulling down other services for their inability to deal with such content, when Twitter and Facebook themselves can't keep this stuff out.
People will repeatedly attempt to complain this is unfair by attempting to find 'double standards' - here, the idea is if anyone tweets 'kill trump' and we can find the tweet an hour later, clearly Twitter has a bias towards Duh Dems
Thinking people will find these concerns obviated by the platforms consistently kicking off groups when things turn violent. It's just that the only examples of this until now were international and off most people's radar.
There are alot more thought and there seem to be no moderation:
@xxx someone just fuckin kill trump already 6:11 PM · Jan 7, 2021
I removed the username to not expose her.
The problem is that you can close down any site with user generated content and point to your ToS while looking through the fingers concerning sites you like, like Twitter. Imagine having some full time employee generaring illegal content to get your competitors closed down where Twitter are immune. It is way too easy.
Parler was cited for not wanting to moderate at scale.
Nonetheless, you have a point, and despite both of us reporting it, it’s still up, and shouldn’t be. In any case, that doesn’t change my point about parler, however, AWS or whoever would be well within their rights to cite this as a mark against Twitter too.
It also isn't a credible threat of violence: It has no specific plans, locations, or times, it does not have the support of thousands of people and high-ranking politicians to gather and enact those plans, etc. If this manifested in a large movement that resulted in serious damage and harm (e.g. like the attacks on the Capitol) then this comparison would be a lot better supported.
The reason for the Amazon ban is not the simple fact that Parler had people making threats. Per Amazon's email to Parler, it's that they had seen an on-going escalation of credible (much more so than the two tweets you linked) incitements to violence, Parler had a history of dragging its feet about responding to specific complaints about infringing content, Parler failed to furnish a tenable plan for how to implement an effective content moderation policy, and public communication by executives indicated that they had no good-faith intent to do so.
Furthermore, while I agree that Twitter doesn't do as great a job at stuff like this as one would like, either, this is just a really ill-chosen place for Parler's supporters to try and make their stand. The cast iron pot does not help its case by calling the steel kettle black.
Is it any different from Twitter though?
Parler’s Complaint is replete with insinuations that AWS had equal grounds to suspend Twitter’s account and thus discriminated against Parler. . . But AWS does not host Twitter’s feed, so of course it could not have suspended access to Twitter’s content.
Just as an aside, did they really tried to argue in court that AWS should suspend Twitter too despite not having anything to do with them? Jesus Christ, the incompetence in some of these people.
And even suggesting otherwise furthers the whole, "I reject reality in order to substitute my own narrative that better supports the position I'm trying to take," thing.
And, in this particular context, it's also a non sequitur. Go read the court filings. The story of what actually happened is being told there. It's a public record.
The Secret Service investigates these types of things. What relevance is this to bomb threats at the data center of a company?
I get whataboutitism is retort de jure for anything that touches on politics these days. But at least try harder.
Report button is behind three dots on that tweet.
Are you saying the post was deliberately fabricated to provoke outrage? It's really not that far out of the realm of what really was happening on Parler. That doesn't mean it was a concrete threat, but assuming that Amazon is deliberately inventing threats against its own data centers seems a bit much.
I just... I mean I don't understand what's happening to discourse on the right of the US political spectrum right now. Our capitol was just sacked. The Oregon state house was invaded. The governor of Michigan was the target of a kidnapping plot... But everyone wants to pretend that none of these things should inform our priors about the potential for right wing violence?
Generally it gets lost in the "who is worse than who" and some unmentioned desire for a regression to the mean. The problem is that a regression to the mean is either prompted by an event or time.
Left wing violence is at an all time high? Really?
The beautiful thing about your comment is that you assume I have some indefensible, reductionist view of the world. I don't. I've provided examples on this website of that violence and the divisive rhetoric.
Here's some highlights from this weekend:
https://www.reddit.com/r/sandiego/comments/kuzeha/recap_of_p...
https://www.reddit.com/r/sandiego/comments/ku1di1/terrorist_...
https://www.reddit.com/r/sandiego/comments/ku277j/protrump_r...
What do you make of this? https://csis-website-prod.s3.amazonaws.com/s3fs-public/TNT_G... (from https://www.csis.org/analysis/escalating-terrorism-problem-u...)
edit: It was 2019, my bad.
This is bias. I can't think of a shooter or murderer who probably doesn't have mental health issues. It does not change whether it's left or right wing violence.
It looks like it pretty well aligns with my understanding as it can be understood from the data we have today.
Two notes:
- not all violence is terrorism
- the definition of terrorism stays within an upper bound and lower bound of tolerance, but nonetheless is constantly in flux
Generally speaking, left wing violence (even in history) is coordinated by loosely connected cells. This makes things quite difficult to track. I always hark back to my days in the military and how hard of a time between 2009 and 2012 they had understanding disjointed cells of terror with respect to the Taliban.
They're better understood now, but it took a long time. The only benefit that I had in understanding the organization of these cells was that I was on IRC as a young person and got to watch organizations like Anonymous flourish. I can see left wing violence but I can't point to a specific organization or person leading it. I won't speculate on why left-wing groups do this, but it is of strategic advantage even in history. The bombings of the 1970's are a great example of this, if we choose not to pay attention. [0]
This is in contrast to right wing violence that has big names like Gavin McInnes, Richard Spencer, at times Donald Trump or go by names like The Proud Boys, The Alt-Right, the KKK, etc... When you have a name and a domain you're a whole lot easier to track and it's easier to talk about substantively and definitively.
The above explains why left-wing violence is just plain missing from this graph. It misses my overall point though.
My overall point is that any violent group deters discourse. Any group that deters discourse is a threat to democracy. I would suspect even if we tracked them equally that right-wing violence would outpace left-wing violence. That's a neat comparison for a graph but has very little substance at the end of the day. We are not single task machines and we can certainly focus on shutting down two groups that are a threat to democracy.
Any violence which is used to deter discourse deserves to be shut down and it is most effectively shut down by the people who belong to the groups most closely related to it. That's to say, if you're a Republican you can more easily shut down the Alt-Right because you have better access to understanding and tools within the RNC to get rid of them. The same would go for Democrats who could easily shut down left-wing violence and divisive rhetoric, but like Republicans have let it create a home.
My last point is rhetoric, which of course is not captured by CSIS. Both of these groups (violence aside) have toxified public and online spaces with hate speech and divisive rhetoric. I cannot count how many left-wing people have entered into these spaces and accused anyone and everyone of some moral crime to the point that people are starting to posit that it sounds like a religion. Having grown up in a repressive religion, I have very little to argue against this point with. On the other hand, right wing groups have saturated online spaces with hate that has driven down diversity in communities and has created this constitution looking rule systems. They both use similar methods: brigading, harassment, stalking, doxxing, etc...
I suspect all of the above are interconnected. All of them are worth stopping. It just takes people who don't mind saying, "enough is enough".
We are too divided and it’s because of the media, both social and mainstream.
An analogy might be the phenomenon of Complexity in software.
Which to me seems like slinging jargon rather than registering any kind of context. To me, it sounds like a misreading of McLuhan more than anything. In fact, McLuhan's point was exactly that kind of misunderstanding and misreading should be avoided by not taking the context for granted.
This is pure whataboutism, basically. Even if you're right (for the record, I think you're right), you're applying it to deflect an argument against "your" side. Isn't that making the problem you are explaining WORSE and not better?
Surely you'd agree that step one to being less divided is to stop reflexively defending "our" side when bad things happen, right?
It could also be the fact that Parler isn't really good at that whole database thing:
Or some dumb bs like that
> They are banned from social media because they were inciting violence. > > Their platform is being shutdown because they were inciting violence. > > And now they are inciting violence because of that?
Some of these angry users obviously don't appreciate being silenced, no matter how awful their message is. I totally get silencing their message if it's threatening someone, but how else can they voice their opposition to Parler being shut down? They become deplatformed, and no one is willing to listen to them or publish their message.
At some point, they'll just reach a breaking point where they feel the only way to be listened is to use extreme measures. I frankly don't know what we can do to defuse this situation other than punishing those directly associated with hateful messages instead of deplatforming everyone on it.
Bingo.
edit: I should clarify - hateful messages might not reach the bar, but calls to violence and other illegal activity do.
I mean, 'we' (as in 'the west') try to deplatform Jihad based terrorism. I think it's a pretty effective strategy, just not complete. Why should we _not_ deplatform far right white supremacy violence and terror?
Of course you also need a deradicalization approach, but that doesn't scale to the current levels of extremism. We need to stop this massive rise of fascist violence rooted in conspiracy thinking from spreading, which was amplified by social media in the first place, only then can a targeted deradicalization approach be effective.
in fact, i would say by now Trump has lost the benefit of being aired live and uncut. you can't spout untruths constantly and expect people to just keep mainlining your BS to every TV in the country.
it sounds to me that you think that if Trump doesn't get full, unrestricted, on-demand use of american companies' broadcast facilities, he's being censored and oppressed. Sadly, he's President not King. Just because he's President doesn't mean anything that he says has value to anyone any more than anything i have to say.
The transition from a situation in which public officials can speak to the public directly to one in which they can only do so at the whim of corporate titans is not a safe precedent.
and his proper press conferences still get aired. this is despite the nonsense spewed during them.
what isn't being aired live is more a rally than a government briefing.
the President doesn't get to stump on national TV whenever he wants. if he were a King he could ofc.
Public officials have not lost any ability they ever had to speak directly to the public. They have at the same time always had an additional ability to do so mitigated by the need to get other private parties (including, but not limited to, “corporate titans”) to relay their speech, whether it was newspapers which existed before the founding and expanded over time, broadcast media starting in the early 20th century, or online media starting in the late 20th century.
And it's not a new phenomenon that some subset of current incumbent politicians might lose some degree of the favor they have had with one of those mechanisms that relies on the exercise of First Amendment rights by some intermediary.
I understand where you are coming from but that's pretty much textbook "negotiating with terrorists". It's coddling a group of violent (or at least violent-speech) people out of fear that if we don't give them a space to be horrible then they might act out.
*a handful or users. More numerous than hate groups on twitter.
Pure speculation, but I wonder if they're also trying to protect themselves against insider threats as well with this move.
You don't want engineers breaking things when you've got something else distracting you.
Simpler to just stop making big changes that aren't 100% required to happen just in case. If it can't wait, approvals from someone senior who agrees it can't wait suffices.
(Source and bias note: AMZN for 8 years).
Even routine code pushes and changes to production infrastructure typically need a second-person review, among other things.
It's worth pointing out that a lot of other countries on Earth have laws that prohibit hate speech. Quite a few countries have libel laws, too. The US does not have either of those at the federal level, making it much more difficult to use the law to stop people from inciting violence and spreading blatant lies.
It's true--I think--that there is no federal libel statute (though perhaps someone can correct me on this), but that assertion is misleading without a little more context. Whether it is a state or federal law doesn't really mean much in the American legal system. In the U.S., most such laws are state, not federal laws. That doesn't make them less important or enforceable.
Every state that I know of recognizes libel as a tort (i.e., you can sue someone for it), and you can do so in federal court if the required ents for federal jurisdiction are met. (In this case, probably that the plaintiff and defendant are from different states.) And, in any case, what's wrong with state courts? A judgement in a state court is just as good as one in federal court.
Well that in itself is a crime, so prosecute it.
Where did I indicate getting banned from a website was too far?
I liken it to having a toxic person in your life who can't admit that their actions are ever wrong - what do you do in a situation like that? At some point, you just have to cut them off.
Unlike Twitter and Facebook, Amazon has a much larger attack surface.
I think it was 1983, just after I started there, that an edict came down to "truck-proof" the building. If I remember, 1983 was when Hammas blew up the US embassy in Beirut with a truck bomb.
They put up those "anti-truck" columns in front of the doors.
Like a bollard? Or something much more substantive.
I had to look it up.
Like "double-wide" bollards. Made of concrete, in the front, and planters in the back (much bigger).
Amazon, unlike say, Google, does not disclose their carbon emissions. Like at all. It is important to have this information public, for the sake of our planet. Amazon gets an F for disclosure, while Google is getting some A's:
https://www.cdp.net/en/responses/658?queries%255Bname%255D%3...
https://www.cdp.net/en/responses?utf8=%E2%9C%93&queries%5Bna...
I assume the reason is that doing a GHG assessment gives a lot of information about the operations of your company, so Amazon would rather be secretive about its operations. Knowing the location of datacentres allows you to deduce what power grid they're attached to. Scope 2 emissions depend a lot on where the electricity is coming from, and they're likely the vast majority of Amazon's emissions due to the size of AWS. Their Scope 3 emissions due to deliveries might also be significant, but it's harder to be secretive about that.
I can think of another purpose which is valid to some people: sowing chaos.
If you know the general area you can probably use public satellite imagery to narrow it down a bit. I suspect there some other clever information you could use to find the locations if you really wanted to.
Even with all the information I'm not sure how feasible it is for any small group to orchestrate a real disruption given how distributed data AWS centers are. Lots of failovers exist and probably a lot people outside AWS know little about.
This is one from 2016 and is red-blue binary. It is perhaps the most illustrative: https://blueshift.io/election-2016-county-map.html
This makes peaceful resolution very difficult, as we're dealing with two groups who largely live apart from one another, with different geography, culture, and social mores. I think it's important to emphasize, there are a lot of people who genuinely believe that systemic voting irregularities influenced our last election, to the point of having changed the outcome, and don't view Joe Biden as having been legitimately elected. Furthermore, they believe that the courts, especially the Supreme Court, refused to even hear their case, and that Congress would in no way give their grievances with the process an open debate, before accepting results they viewed as fraudulent. This puts them into a position of desperation. The videos of people being arrested at the airport do not help: https://www.youtube.com/watch?v=lxwJ5HXSzuA
Given the geography. I expect to see red-hats (Trump people) going into the cities where the blue-hats (centrists) and the black-hats (Antifa/BLM) live, and then retreating into the rural areas, where supportive sheriffs are more likely to run interference against other law-enforcement agencies. To some extent, this is what we already saw at the Capitol Building, as well in places like Kenosha where people like Kyle Rittenhouse came from out-of town (albeit within reasonable driving distance) to the city where violence was occurring.
EDIT: better link
EDIT 2: The phrase "I've thought a lot about this" suggests that one has thought more of themselves thinking about the issue, than about the issue itself. As such It has been removed.
Randall Munroe did a twitter thread[0] on the election map he did for xkcd 2399[1] and he actually discusses about the bias in common election maps: in fact there is no such geographic split.
Or, as he puts it:
> There are more Trump voters in California than Texas, more Biden voters in Texas than NY, more Trump voters in NY than Ohio, more Biden voters in Ohio than Massachusetts, more Trump voters in Massachusetts than Mississippi, and more Biden voters in Mississippi than Vermont.
Don't take the map literally, the key ideas are in the Twitter thread. The main point of his thread, is that maps aren't good at saying where voters are and are fundamentally biased (“You can be a Biden voter in a Trump household in a Biden precinct in a Trump county in a Biden district in a Trump state in a Biden country. Then what color is your land? Dirt-colored, I guess! Not only does land not vote, it doesn’t even have a preference.”). He made his map as a counterpoint to nationwide maps, but he could have made the same map at the state, or even the district level: there are trumps voter in the middle of San Francisco and Bidden voters in the middle of nowhere in rural California. In fact, there are more Trump voters in LA alone than in 80% of the red areas in nationwide maps.
They also do not reflect movements that are fundamentally leninist in organization, where a small cohesive nucleus of activists defines the agenda and directs action. Without a Fox News stoking and directing anger, there is no "red rage", and Fox people are urban people. Without someone running 4chan, there is no Anon/QAnon, and the likes of 4chan are run by urban geeks. Trump always defined himself as a proud NewYorker. And so on and so forth.
What is true is that nowadays people can live in socially-isolated bubbles that do not communicate with other bubbles. This is not a function of geographical location, though, but hyper-personalized media consumption. It happened already in the 70s on the left, for example, where people who only read "alternative" media could not understand how their points of view where not as hegemonic as they thought. Now it's just spread to everyone thanks to the web: at some point you'll gravitate to a media bubble and probably stay there forever.
Littlestymear linked a map in another reply to my comment that seems to break it down purely by population in clusters of 250,000: https://xkcd.com/2399/
That map still suggests that clusters of blue in highly urban areas are surrounded by clusters of red in "less urban" areas, and never the opposite. Rural, or less urban environments have different cultures than highly urban environments, and this is readily observable.
>They also do not reflect movements that are fundamentally leninist in organization, where a small cohesive nucleus of activists defines the agenda and directs action.
I'm afraid I'm not sure what you mean by this. Please be more explicit.
>Without a Fox News stoking and directing anger, there is no "red rage", and Fox people are urban people.
This is a misconception. One blames the media for the information purveyed rather than the sorts of people amenable to Fox's viewpoint in their news. If there was prima facie no distinction between those who adore Pres. Trump, and those who despise him, prior to their exposure to Fox News, then this might be a tenable position, but given that there are two different demographics, both of which have access to Fox News and yet react distinctly differently to it, tells us that there are other factors at play.
>What is true is that nowadays people can live in socially-isolated bubbles that do not communicate with other bubbles. This is not a function of geographical location, though, but hyper-personalized media consumption.
And yet geography seems to be at play in which media bubble you end up in, as do a number of other factors. Perhaps the media bubbles are as much a function of existing divisions as the exacerbated divisions are of the bubbles?
But that is not something that changed of recent. New York has always been different from St. Louis, but both "universes" could coexist and now they seemingly cannot. That doesn't look to me like a function of geography.
What has changed is that it's now easier to be an "undercover radical stlouite" in NY and viceversa.
> I'm afraid I'm not sure what you mean by this.
Trump's organization is fundamentally bolshevik in nature, with a small inner circle of "avantgarde" that decides everything and then pushes decisions down to the masses. It's why significant chunks of his own party don't like him, because there is little or no policy elaboration even slightly down the pyramid. Both Clinton and Obama did it to an extent in their first few years, but nobody has done it as much as Trump. Trump could make it stick because the social web reduces the need to cozy-up to the larger press ecosystem for his message to spread. It's the sort of qualitative jump that early-XX century parties did with radio and films, which removed a lot of influence from local bigwigs in favours of national leaders.
> One blames the media for the information purveyed rather than the sorts of people amenable to Fox's viewpoint in their news.
Amenable people still need to articulate and support their views to be effective, and for the message to be efficiently evangelized and legitimized. Dedicated media does that work for them. Without that work, they wouldn't be as effective.
> there are other factors at play.
Of course there are, but the likes of Fox are important factors. Their tone has been extremely divisive since the '90s, this is a long wave coming.
> And yet geography seems to be in play in which media bubble you end up in
Yes, but this was always the case and it's arguably much less so now that local papers are dying off. The point is what is changing, and geography has largely not changed; but the media landscape has, and pretty dramatically too.
It really is becoming country vs. anything even remotely close to a city. The only exceptions seem to be rural college towns and hipster/artist enclaves. There are some good examples of the latter in Appalachia.
Agrarian: traditional, religious, rigid social roles, rigid hierarchies determined by gender/race/etc., suspicion of change, and fiscally conservative with a suspicion of debt or loose monetary policy.
Industrial/cosmopolitan: fluid, either non-religious or liberal religion, fluid social roles, hierarchies only as-needed (e.g. corporate hierarchies that are opt-out), welcoming to rapid change, and fiscally liberal with a pragmatic attitude toward things like monetary policy and debt.
The wrinkle in this model is that the top of the hierarchy of "red America" seems to be located in urban centers. Trump is from New York. Most 4chan /pol users are in cities. The people operating Fox News and other further-right news sources almost all live in cities. So it's sort of like you have this minority of urban America either sympathizing with or puppeteering a rural revolt against modernity.
Of course I wonder if the civil war was different. Banks, governments, and major property owners in urban areas profited a lot more from slavery than poor and middle class farmers and skilled laborers in the South. Quite a lot of the profit from slavery ended up in New York, Boston, DC, and even London.
Maybe part of the answer to this is for rural Americans to turn off Fox News and OAN and the /pol orbit and all the rest of that "ruralist propaganda from urbanites" and figure out what they really think and articulate it themselves. What emerges may well be something toward which urban "elites" could be more sympathetic.
Need you be reminded that a group of armed radical right wing extremists forcibly invaded a session of congress with the expressed intent of executing politicians?
Honestly, it is difficult to believe that anyone saying further right-wing violence is unlikely is arguing in good faith.
Please elaborate on which part of my characterization you disagree with.
I've not seen any photos of them being armed either.