Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

If I was a junior dev who got such a ticket, I can see myself implementing it all the while laughing at the utter ridiculous pointlessness of it.

I can also see the ticket being written with examples of “disallow a dictionary of forbidden words”, with a sample dictionary of [“motherfucker”, “slut”, <list of racial epithets>] and then the dictionary function being used later to include “union” etc.

If I got a ticket to block the dictionary containing MF and racial epithets, I’d also think it was pointless, would point out it was easily able to be evaded, but would implement it and implementing it would not be obviously evil or showing a lack of solidarity.



It’s not just jr devs that should be acting in solidarity with warehouse workers. Their team leads, IT, project managers, etc. I don’t care how much amazon stock you’re getting as part of your comp, until the capital gains on your stock outpaces your salary you’re closer to a warehouse worker than you are to an owner - and your allegiances should be with your coworkers, not jeff.

So yeah, this specific case might have flown under the radar - and got implemented unwittingly - but unless a top level exec got in there and edited the dictionary file themselves, there was opportunities in this being implemented to show solidarity and throw some clogs into the machinery


Ok. Preventing a coworker from calling another a “slut”, “motherfucker”, “fag”, or “n####r” seems like a pretty reasonable feature for a workplace chat system. Lacking such a system might even constitute creating or contributing to a hostile workplace environment.

What’s the level of dev who should show allegiance to fellow coworkers by refusing to implement that?

Then some non-dev employee realizes you can block these other words…


And then some dev can accidentally roll back the changes as part of a bug fix.

Actually we’re having problems with the whole filtering system, it turns out it relies on a library that has some security vulnerabilities and it’s no longer operational.

Or if you must, wait until it breaks on its own, and then refuse to fix it. “Suppressing labour organizing isn’t part of my job description”

I.e. Grow a spine and stand up for what’s right


Why assume this list is hardcoded? It's possible/likely it's controlled by an API (ultimately through a web frontend) and stored on a machine devs don't even have permission to access... that's pretty par for the course for Amazon. The person who coded that functionality, and the person who has credentials to edit the list are likely not one in the same.

(The exception would be, when the person who coded the API goes on-call. Still there's lots of checks and balances there and any such change would be quickly undone.)


You’re right - diffusion of responsibility and technological layering / obfuscation is a great way to avoid making anybody feel bad about doing something bad. It’s quite possible that this was done in a clever way that avoided anyone implementing anything they felt bad about. But ultimately, you can’t “just doing my job” or “i didn’t know it would be used that way” your way out of the harm your job does - it’s your responsibility to notice and act, preferably proactively, but there’s always retroactively.


It sounds reasonable and that is why it's dangerous to censor and everyone should fight against any censorship. If the guy working next to you is the type to send you slurs via messenger on company time wouldn't you want to know that so you could take the messages to management and have them removed? Or you would you want him to have a new imagined grievance on top of his already hateful viewpoint?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: