If you work with a database and are asked to alter the table structure to comply in advance for citizenship or gender categorizations it's really important to NOT do it.

"The governor is concerned about all this stuff they want us to update our record keeping so we store both gender AND biological sex."

"We need fields to store the country of origin of people's parents."

If you don't have the power to rebuff this yourself, ask for help. At minimum ask for help online anonymously.

in reply to myrmepropagandist

Depending on your job you probably have in the past made compromises. Maybe to keep your job. Maybe to survive. This is a bright line. If you are asked to be the one to update the table don't let it be your fingers typing those changes.

If you can't just say "No I won't do that." Stall, run away, feign incompetence. Just don't let it happen.

I suspect this might be where the rubber hits the road first for us around here.

Nothing has changed. You do not have to do it. It is not even ordained.

in reply to myrmepropagandist

in reply to myrmepropagandist

I know someone who rebuffed such a request. Boss was apologetic "it's what the higher ups want, oh *I* think it's a lot of nonsense, but I don't want us to be out of step ... blah blah"

It was proposed to them in sheepish way. They said it would be a lot of work, not add anything of value, and most important they would not do it. It didn't come up again.

Fascism can be the work of zealots, but there are also many sheepish middle management helping hands who "don't even believe in this really"

in reply to myrmepropagandist

With the sheepish ones just letting them know you won't do it, (that it crosses the bright line) can make them back off some of the time.

This can be very scary and if you are thinking "but I could be fired" I understand that. Ask for help. Talk about it outside of work or with people you trust. Don't go looking for an excuse to comply and not feel bad about it. If you do you should feel bad.

Find a way to NOT do it instead.

This entry was edited (10 months ago)
in reply to myrmepropagandist

I'm honestly just psyching myself up for if I need to do this kind of thing again. I might.

It's the whole "we're just doing it to go along with what's happening now" ethos that I think might hold the most little victories for us.

So many terrible things happen because of people just going along with a bunch of little bad things that come together into a much more ugly and unstoppable bad thing.

in reply to myrmepropagandist

It’s so easy to think, “all I did was place one little cog over here in a quiet place, I sure it doesn’t matter” when you aren’t close to where the machine is grinding people up in a horror of blood and violence. That’s what they count on to get the machine built. If you really have no choice but to put a cog there, *make sure* that first you bend it, or break off some teeth, or weld it in place so that it will never ever actually function as intended.

Aaron reshared this.

in reply to myrmepropagandist

There is a legendary problem in US government graph analytics/entity alignment circles called the "Maria Gomez Problem."

When you have a bunch of records (e.g. financial transactions, phone calls, etc.) It's important to keep track people across them, even when. The records store slightly different information (that's entity alignment). So one DHS analyst, working on SA cartels, aligned everyone named "Maria Gomez." The system Palintir built for them had no way to undo it.

@futurebird

in reply to myrmepropagandist

by way of inspiration:

In the 70s, a bunch of workers in East Kilbride 🏴󠁧󠁢󠁳󠁣󠁴󠁿 grounded half of Chile's air force after Pinochet's coup. For 4 years they refused to fix the engines for Chile's Hawker Hunter aircraft.

There's a documentary about it "Nae Pasaran":
scottishdocinstitute.com/films…

Kees AntiFA van Malssen reshared this.

in reply to myrmepropagandist

I want people to challenge these folks on every lie, every distortion, every act of malice, every obfuscation, never letting up, not for one second, not now, not in the future, not ever.

They do not get to be comfortable in the performance of their acts of evil and cruelty, no matter how minor their trangressive behavior may seem to others. Get in their faces, speak up in public, stand up, push back, speak out, without ceasing.

Ω 🌍 Gus Posey reshared this.

in reply to myrmepropagandist

Corporates now promote people who are risk-adverse: "yes men". You don't get fired if you don't take chances. That's why startup culture is thriving and Google, Facebook, Amazon etc. are struggling to recruit. They used to be exciting (if exploitative) but they're now too scared to say "NO" to an invitation to a fascist's party in case they don't get even richer.
in reply to myrmepropagandist

Also, learn lessons from these middle managers. If you aren't empowered to say no, then use their strategies for avoiding things you don't want to do, create processes to function as obstacles, create arbitrary requirements that work as obstacles.

E.g., "Oh sure, I can do that no problem, I just need you to fill out this project intake form."

"Oh definitely, that's a great, form, I can't help but see that you left this section blank, unfortunately I'm going to need you to fill in the section explaining the strategy for rollback in the case of emergencies."

"I'd love to help you with this, but unfortunately our technical standards require us to justify all changes using this scoring point system."

Etc., etc.,

We've spent decades watching this bullshit be used to obstruct, delay and defer progress, we can use the same tactics to obstruct, delay, and defer anti-progress. It won't work as well, because fascists don't care about following the rules, but there's enough people who don't want to think of themselves as fascists around that it will still be effective.

reshared this

in reply to myrmepropagandist

Insist on doing everything through "channels." Never permit short-cuts to be taken in order to expedite decisions.
Make "speeches," Talk as frequently as possible and at great length. Illustrate your points by long anecdotes and accounts of personal experiences. Never hesitate to make a few appropriate "patriotic" comments.
When possible, refer all matters to committees, for "further study and consideration." Attempt to make the committees as large as possible - never less than five.
Bring up irrelevant issues as frequently as possible.
Haggle over precise wordings of communications, minutes, resolutions.


web.archive.org/web/2020022517…
The CIA removed their WWII workplace sabotage manual from their website but you can still find it on archive dot org

reshared this

in reply to myrmepropagandist

I once told my manager I was not prepared to do something because it both defied the company values and my personal professional standards. He told me that as a professional my duty was to do as I was told. A little while later, my name appeared on a list of techies to be made redundant. It MAY have been a coincidence. The others were good people. My union fought to get me moved to another job. I liked it much more and how would I have felt if I hadn't stood up to him? #JoinAUnion
This entry was edited (10 months ago)

reshared this

in reply to myrmepropagandist

this!

Also, if you are being forced to do this, ask for this request/order *on paper* (or at least in writing).

I have been asked, at the eve of my IT career, to do a slightly unethical thing. I refused, but was told I have to do this anyway. I said: fine, but I would like this on paper, including the acknowledgement of my concerns.

It was never spoken of again, and never implemented. I kept my job, never had any grief from the higher ups about it either.

Paper trail is power.

in reply to Michał "rysiek" Woźniak · 🇺🇦

@rysiek To build on what Raven said, you can often present a focus on clarity rather than ethics, and get people to see the ethical dilemmas and go do something else. This is less satisfying than a good table flip, but comes with a paycheck and a chance to do it again.
in reply to I Thought I Saw A 2

@ithoughtisawa2 sure. But they can do that regardless of the paper trail.

The paper trail provides you with a CYA policy. If they issue such an unethical directive and then try to blame it on you, you have something to prove you were pushing back against it.

"The more paper, the cleaner the arse", as they used to say in the Soviet Union.

in reply to Michał "rysiek" Woźniak · 🇺🇦

@rysiek This is super important, I learned it from one of my managers early in my career. If someone has a verbal conversation with you at work where they tell you to do something problematic, either tell them to put it in writing or send them an email following up on/recapping the conversation to cover yourself.
in reply to Misuse Case

Best advice I had from the best boss I ever had was: Save everything. Every email, every letter, every post-it note. Never volunteer for anything. Make sure your job is spelled out and nothing gets quietly added (got monumentally screwed on the last part when they refused to fill back necessary full time positions).

A post-it note signed by the person who was trying to get me into trouble for doing what they specifically asked me to do worked wonders.

This entry was edited (10 months ago)
in reply to Misuse Case

@MisuseCase YES. Absolutely and the way I used to do this without sounding accusatory or aggressive as I would just make a joke about ADHD and I might forget before I even get back to my desk so please send an email.

And if they don’t I just forget before I get back to my desk. And for real, sometimes it wasn’t even on purpose.

in reply to Michał "rysiek" Woźniak · 🇺🇦

@rysiek this is a good idea because when I was suspicious of something I was told to do I actually forwarded that email to my home email, because if I get fired tomorrow I don’t have access to my work email and if I get arrested in six months I’m not going to be able to prove someone else told me to do it (it wasn’t a crime, just sayin’) and IT called my supervisor to tell her I sent the email to my home.

I worked in the mortgage department of a credit union so I thought it would be much more suspicious and problematic if I was printing papers and then squirreling them away into my bag, but I was very very wrong. Lesson learned lol

in reply to myrmepropagandist

I was asked by my boss, a rabbi no less, to photocopy illegally an entire book that was written for a very small audience and therefore over $100 to buy, and though he had an expense account which would cover it, wanted me to do the legally and morally wrong thing for him. I refused and told him it was copyright infringement. I might have got him hating me for it, but he already hated me by even asking me to do the evil on his behalf.

myrmepropagandist reshared this.

in reply to Jo - pièce de résistance

@JoBlakely I lost a job for refusing to do something similar (refusing to use copyrighted for-purchase music on a church website w/o paying for it). They stole EVERYTHING—every image, every video, every song. I just flat out refused, saying that I’d always been taught stealing was a sin. @futurebird
in reply to myrmepropagandist

[Edit] How to slow down software dev.

- Ask for complete and clear requirements before you can do any work.
- Take the time to do exhaustive estimates of how long it will take.
- Come up with multiple different implementations.
- Ask the architecture team to review the implementations.
- Request input from the platform team.
- Make sure to get buy-in from all the stakeholders.
- Schedule meetings at inconvenient times, if the team is in multiple time zones make sure that someone will suffer.
- Schedule meeting times without checking calendars first forcing people to ask for (later) reschedules.
- Create a comprehensive QA plan with a full regression test-suite.
- Setup meetings with the UI team to define an API to use the new functionality.
- Use the full iterative UX design process, sketches with feedback, mockups with feedback, UX testing on prototypes.
- Do the first implementation based on incomplete or misunderstood requirements so that rework is required.

This entry was edited (10 months ago)
in reply to myrmepropagandist

This entry was edited (10 months ago)
in reply to myrmepropagandist

yes opposition may not always be easy but you can just fail.

Not even via incompetence. You can take your time looking at the requirements and make a highly optimized implementation (for the sake of performance) that is difficult to get right and even more difficult to fix your data once you got it wrong.

Alternatively you can build a highly expressive solution where the extensibility causes all relevant queries impossible to answer (because different teams used different names).

in reply to myrmepropagandist

As this would require in most cases data to populate these fields this would seriously compromise the stability of the database.

We need a study on how this data is to be collected and verified and all of the possible cases such as unknown data, refusal to provide data and insufficient verification.

We will also need policies and procedures covering this and any liability we may incur in inadvertent incorrect data and its application in other systems.

in reply to Rich Puchalsky ⩜⃝

@richpuchalsky Thanks, I was scrolling through the replies and hoping someone had already mentioned IBM and the Holocaust. Such an important book for everyone to read.

There's a particular aspect of the book that this thread reminded me of: France was much better at noncompliance than the Netherlands, which is why a larger % of French Jews than Dutch Jews survived. Being sticklers for rules is normally an admirable national trait, but not when it's compliance with evil.

in reply to myrmepropagandist

"but how do I get away with that?"

openculture.com/2024/11/the-ci…

in reply to myrmepropagandist

First, read en.wikipedia.org/wiki/Weapons_…

My strategy is usually some variation of the following (parenthesis is your excuse if caught):

1) Silently don't do it. say yes, and then just don't do it (forgot).
2) Do it, but don't update the front end to comply (oops, oversight)
3) Do it, update the front end, but then don't maintain it until it silently breaks (oversight, forgot, update must have broken it)
4) Do it, but then after a month or so, remove it, or send the data to /dev/null (oh, you still needed that?)
5) Store the data as a hash or some obfuscated string (for privacy), and then "lose" the index (I don't know how that happened!)

Never say no. Just demure. Yes, yes, tomorrow, I'll get to it, don't worry... and then just go what we call in Puerto Rico, "brazos caídos" fallen arms or limp arms.

in reply to myrmepropagandist

They will never come back to you. This stuff is so complicated and difficult to track. In an ideal world, management has PERFECT vigilance over all aspects of a project, but this has never been the case in my own experiences.

It also works at the country/nation level too! Weapons of the Weak was introduced to me by my anthropologist partner and it's changed how I deal with the world.

in reply to myrmepropagandist

Or, you know...

"We can implement that, yeah, but it's more complicated than it first appears."

"The field lengths aren't really clear, and we'll need to get those right before invest in migrating this entire database."

"I've scheduled a meeting with a group of experts to outline how we can study the problem and get stakeholder input on the correct model to use in our database."

Let it snow....

in reply to PKs Powerfromspace1

Read the responses to my post from everyone here, there are many exciting ideas for people with all levels of power in these situations.

And really if you say "oh OK when I can get to that" then do a bunch of other things and make a big deal about that and never get to it can also work.

They may just forget about it.

This entry was edited (10 months ago)
in reply to myrmepropagandist

Also it’s a real good time for software engineers and DBAs to require:

- full design documents
- approval from the product owner
- prioritisation at the next quarterly product meeting
- (accidentally forget about the ticket)
- user stories
- test evidence
- sign off from correct product owner
- PR approval before merge
- unit tests
- full test coverage
- approved change paperwork
- (sorry missed the change window)
- downstream sign off

It’ll take at least 5 years to add that.

@futurebird

in reply to myrmepropagandist

Changes to data gathering practices are an early warning sign...

en.m.wikipedia.org/wiki/IBM_an…

theguardian.com/world/2002/mar…

cepr.org/voxeu/columns/fiscal-…

en.m.wikipedia.org/wiki/Judenv…

in reply to myrmepropagandist

You can practice malicious incompetence. All while superficially complying.

I *absolutely* affirm refusing to do it, but if you are pretty sure that they’ll just get someone else to do it, or you’re really trapped and can’t afford to refuse (and quit/be fired), then become incompetent.

Slow.

Make stupid errors. Especially ones that cause damage and/or are hard to undo. Build on top of fundamental errors. Sabotage. Delay. Obfuscate. Be a cunning little non-compliant weasel.

in reply to myrmepropagandist

Sensitive content

This entry was edited (10 months ago)

reshared this

in reply to myrmepropagandist

“I’m concerned about how this interacts with our CCPA/GDPR obligations, do we have sign off from legal/security/our chief compliance officer and a plan to update our policy and process documents?”

Realistically only relevant at a certain size of company, but a fine way to slow things down. And has the advantage of being a correct thing to ask!

in reply to myrmepropagandist

in reply to myrmepropagandist

Also, even if you find you can’t safely refuse doing such work, you can at least try to make it obvious.

Add it to the documentation, include it in the release notes, add a switch to turn it off that may be noticed, inform user forums of the existence of these features to give them the option to complain.

Depending on what precisely you're working on, there are small ways in which you might be able to give others in a more safe situation the information/tools they need to do more.

in reply to myrmepropagandist

If it’s too risky to say no outright, I bet most engineers instinctively already know: There are ways to effectively kill a project without breaking the rules (Because we spend so much time fighting *against* that very force ordinarily):

Insist on CAB approvals. Make sure all stakeholders are in agreement about the details. Did we get sign off from legal?

Before you know it, following the rules to a tee makes it impossible for anything to get done.

in reply to myrmepropagandist

This entry was edited (10 months ago)
in reply to myrmepropagandist

in reply to myrmepropagandist

@ansuz Agree enthusiastically to make those changes to prevent somebody else from actually doing it, then begin to implement the most complex version of the changes with many subtle flaws.

Document extensively, but use dueling systems with broken cross-links.

Don’t invite one different stakeholder to meetings on a rotating basis.

Update most but not all test systems to the latest beta of the OS.

Use intriguing exotic char encodings for some tables.

in reply to Mr. Lance E Sloan (IRL) 👤

@sloanlance

On these issues of helping to build human dislocation, segregation and discrimination machines there is a tendency for the actors to rationalize and externalize their responsibility for violence. A tendency to say “since we all benefit from these monstrous systems it’s not my fault.”

But to help build a new intake shaft for a factory that chops up humans is not excusable. To design it, to shape it? No. That is where we all need to say “no further I will not.”

reshared this

in reply to myrmepropagandist

What if the development is not intended for discrimination or segregation?

Obviously, if a project is specifically intended to HELP PREVENT bigotry and aid the people for whom the data is collected, that would be a good thing. What about projects that are clearly not inhumane, but their aid to people is ambiguous? What is the harm in those projects collecting this data?

myrmepropagandist reshared this.

in reply to Mr. Lance E Sloan (IRL) 👤

@sloanlance

My response, not expert, is that the road to hell is paved, consider with *what*. If you think your data collection could cause harm, even if you don’t know how can you talk to multiple people who are in the group who could be harmed (not just one!) if it’s non-citizens or trans people or POC can you get advice from someone with skin in the game who might notice what you could miss?

in reply to myrmepropagandist

Thank you for the kind, reasoned response. I'll keep your advice in mind.

In fact, thank you to almost everybody who responded to me. Everybody except the one person who compared me to #convictedFelon #DonaldTrump and when I called them out on it, escalated the hate with ad hominem attacks.

I hoped I wouldn't see that kind of behavior after I migrated away from Twitter/X.

in reply to myrmepropagandist

@sloanlance I think very often people suffer from narrowness of vision. They're asked to do a thing and given a justification. It's a stupid justification, but they want to believe it, lack the historical or theoretical perspective that would make it obvious how bad the justification is, and choose not to think too hard about it. Doing so would be uncomfortable, create extra work, and make them unpopular with their peers. So they do the practical thing and do as they're told.

reshared this

in reply to Amro has been

@amro My response was meant to be lighthearted, if not humorous, but sincere nonetheless. The absolute I read in the post was that one should ***never*** collect certain types of data on people. I thought there might be ***some*** good reason to collect that data, so my response was a subtle way to point that out. Fortunately, the poster understood my humor & responded kindly.
in reply to myrmepropagandist

And an associated reminder that the number of countries of birth of parents is INFINITE

Just as an example, both my parents were born in countries that don’t exist anymore if you go by their common names; but if you want formal state names, in one case you’d have to look up things like the Weimar Republic. And that’s an easy one to add to the drop-down list. I haven’t even gone into generations who have died yet, because it grows exponentially after that.

Endless validation …

in reply to myrmepropagandist

Months ago I gave the specific example of not using your database programming and management skills to help build systems that will be used to abuse people. But, that's just what I thought of first because it's what I know best in that space. There are many other points of contact and many other opportunities to be intransigent. Don't miss a single one.
This entry was edited (5 months ago)

reshared this