[HN Gopher] Documenting what you're willing to support (and not)
       ___________________________________________________________________
        
       Documenting what you're willing to support (and not)
        
       Author : zdw
       Score  : 83 points
       Date   : 2025-07-12 17:08 UTC (4 days ago)
        
 (HTM) web link (rachelbythebay.com)
 (TXT) w3m dump (rachelbythebay.com)
        
       | bravesoul2 wrote:
       | > At some point, I realized that if I wrote a wiki page and
       | documented the things that we were willing to support, I could
       | wait about six months and then it would be like it had always
       | been there. Enough people went through the revolving doors of
       | that place such that six months' worth of employee turnover was
       | sufficient to make it look like a whole other company. All I had
       | to do was write it, wait a bit, then start citing it when needed.
       | 
       | Like!
        
       | paol wrote:
       | > a giant social network. You know, the one with all of the cat
       | pictures
       | 
       | This really doesn't narrow it down.
       | 
       | > and later the whole genocide thing and enabling fascism.
       | 
       | Still not helping.
        
         | leosanchez wrote:
         | > genocide thing
         | 
         | I can only think of one social network when I hear that word.
         | Are there others?
        
           | tehjoker wrote:
           | As of 2023, it's all of them promoting a pro-genocide
           | narrative, particularly Twitter these days is promoting
           | outright pro-genocide accounts in my feed from people I don't
           | follow, but I think the author is referring to the Rohingya
           | genocide case in Myanmar.
        
             | stdbrouw wrote:
             | See https://www.amnesty.org/en/latest/news/2022/09/myanmar-
             | faceb...
        
         | zdc1 wrote:
         | The funny thing with social networks is they all have both of
         | these cohorts, so long as you go down the right rabbit-holes
         | and engage with the right content creators
        
         | jraph wrote:
         | Rachel worked at Facebook for some time a while ago. She's been
         | consistently mentioning the cat pictures to refer to it in her
         | blog.
         | 
         | I suppose this way of referring to it is also meant to belittle
         | it.
         | 
         | I assume the "the whole genocide thing and enabling fascism"
         | part is more a dig than something to narrow down.
        
         | grishka wrote:
         | Yeah, might be Twitter, but might as well be Facebook. Though
         | I'm leaning towards Twitter
        
           | stingraycharles wrote:
           | I interpreted it as the Rohingya / Myanmar genocide and
           | Facebook's role in it, and the CambridgeAnalytica scandal
           | that (allegedly) enabled Trump to get elected.
        
         | loloquwowndueo wrote:
         | Maybe she didn't really want to tell you which one it is :)
        
         | drjasonharrison wrote:
         | Rachel Kroll worked at Facebook. Her work history is not
         | explicitly listed on her blog, but you can find it:
         | 
         | https://news.ycombinator.com/item?id=13400687
         | 
         | https://medium.com/wogrammer/rachel-kroll-7944eeb8c692
         | 
         | https://www.usenix.org/conference/srecon16/speaker-or-organi...
        
       | znpy wrote:
       | > I used to be on a team that was responsible for the care and
       | feeding of a great many Linux boxes which together constituted
       | the "web tier" for a giant social network. You know, the one with
       | all of the cat pictures... and later the whole genocide thing and
       | enabling fascism. Yeah, them.
       | 
       | I love how people are so willing to criticise companies that paid
       | their salary but not their willingness to ignore the issues in
       | the name of the fat paycheque.
       | 
       | This is clearly a reference to Meta, and in that sense the
       | writing has been on the wall for years.
       | 
       | I wonder if the author feels or takes any responsibility in
       | directly enabling that genocide and that fascism with their
       | direct work.
       | 
       | But hey, a fat paycheque is a fat paycheque.
        
         | philipwhiuk wrote:
         | Maybe she left when it transitioned from cat pictures to
         | fascism.
        
         | ivape wrote:
         | I would say you were less of a douchebag for taking those gigs
         | once upon a time than now days.
        
       | RhysU wrote:
       | Also, document what "support" means.
       | 
       | "That's supported" has no universal interpretation beyond
       | physically describing a tabletop atop legs.
       | 
       | Absent a clear/consistent definition, people interpret "support"
       | in the most favorable way possible from the seats in which they
       | sit. Then, all around sadness ensues.
        
         | bombcar wrote:
         | This is really important - because "not supported" can range
         | from "eh, we don't have to help you if it explodes" (which they
         | often DO help with, but it's not official) to "this is
         | literally against the license, the law, and the basics of human
         | decency to even attempt".
        
       | simonw wrote:
       | Related, the concept of the "golden path" advocated by Charity
       | Majors: https://charity.wtf/2018/12/02/software-sprawl-the-
       | golden-pa...
       | 
       | > 1. Assemble a small council of trusted senior engineers
       | 
       | > 2. Task them with creating a recommended list of default
       | components for developers to use when building out new services.
       | This will be your Golden Path, the path of convergence (and the
       | path of least resistance).
       | 
       | > 3. Tell all your engineers that going forward, the Golden Path
       | will be fully supported by the org. Upgrades, patches, security
       | fixes; backups, monitoring, build pipeline; deploy tooling,
       | artifact versioning, development environment, even tier 1 on call
       | support. Pave the path with gold. Nobody HAS to use these
       | components ... but if they don't, they're on their own. They will
       | have to support it themselves.
        
         | hnthrow90348765 wrote:
         | >or maybe it's their executives who are afraid?
         | 
         | I'd say yes. It seems like an unforced error to them to
         | broadcast publicly how bad things might be, but that's because
         | they want software to be an efficient factory that turns
         | computer cycles into money.
        
         | roryirvine wrote:
         | You do need to be careful that you don't end up with a "build
         | it and they will come" mindset when creating that Golden Path,
         | though - it needs to have early and continuous input from
         | actual users.
         | 
         | There's a tension between a theoretical Golden Path that leads
         | someplace no-one actually wants to go, and simply paving every
         | possible "desire line". Managing that is one of the trickiest
         | parts of platform engineering.
        
         | cogman10 wrote:
         | Step 3 is where I've seen my org completely fall on its face.
         | 
         | Yes, the golden path is "fully supported" yet after a year or
         | two the company executive swoop in and say "why are we spending
         | so much money on the golden path" and slowly, but surely, the
         | support for is whittled away into nothing until the golden path
         | is out of date and actively punishing anyone that chose to use
         | it.
         | 
         | For example, one of our golden paths was a UX framework built
         | on top of standard web tech for the time. The team maintaining
         | that framework is no more and it's now very far out of date.
         | Adopting it means you are pulling in Angular circa 2016 and
         | that you'll be dealing with incompatibilities between that and
         | any new web component you want to start using.
        
         | tialaramex wrote:
         | You can get screwed over by external (to your org, even more
         | globally) requirements
         | 
         | For example I work for a Research University, mostly our
         | software procurement is - while not always excellent - pretty
         | good stuff. Maybe the supplier isn't as responsive as we'd
         | like, maybe the software is buggier, maybe the documentation is
         | worse, probably not all three.
         | 
         | However, the Government, responding to the usual anti-immigrant
         | sentiment, decided it needs all Universities to check that
         | people who are here on a restricted visa to get a degree attend
         | classes.
         | 
         | The underlying sentiment is clearly racist, but OK in some
         | cases you could imagine that's a real issue, a cheap course
         | with foreign students who are registered but actually never
         | attending because they're out delivering pizza or whatever. For
         | a prestige research University though it doesn't make much
         | sense - maybe you graduated top of your class in Taiwan, your
         | parents pay an eyewatering sum so you can study here for an EE
         | Masters to get that job back home with a team designing CPUs -
         | then instead you skip classes to work as a taxi driver? No,
         | absurd. But the government legally _requires_ we solve this
         | imaginary problem, and the only bidding supplier is garbage. So
         | they 're basically requiring us to procure garbage.
         | 
         | Because the supplier knows we're obligated, why should they
         | support anything? Why care if it works, or is documented
         | properly, or integrates with all the things they've told the
         | government it can do? They know they're getting free money
         | because of anti-immigrant sentiment, and they can take
         | advantage of that until the winds change.
         | 
         | Their attendance tracking stuff _could_ be useful. You can
         | legitimately imagine having an early warning, OK, Sarah took a
         | week for her mum 's funeral, that's sad, but then she didn't
         | attend any lectures at all for the next three weeks, we should
         | have somebody go check Sarah is OK, or, well, she's clearly not
         | OK, have them figure out what Sarah should do next. Take a year
         | pause? Counselling on our dime? Right now, we aren't required
         | to track Sarah unless she's on a restricted visa and the
         | software is awful (for her, for us, for her teaching staff,
         | everything) so we don't track her. So chances are nobody
         | notices, especially if her class is big, and then in another
         | month's time we find out Sarah was struggling after losing a
         | parent, and we find that out because Police have to break her
         | door open as she's been dead for long enough that neighbours
         | noticed the smell.
        
       | Pxtl wrote:
       | Reading this I'm mostly shocked that groups had the power to
       | throw boxes onto their outward-facing infrastructure and she had
       | to handle that reactively. Like, it's not "requisition a public
       | network server" but rather "you just jam it out there and we'll
       | either baby it or boot it". That's crazy loose for a major corp.
        
         | NewJazz wrote:
         | She mentioned experiments. Facebook is indeed crazy loose with
         | how they handle experiments on the site's user base.
         | 
         | https://journals.sagepub.com/doi/full/10.1177/17470161155995...
        
           | Pxtl wrote:
           | Move fast and break people, eh?
        
             | busterarm wrote:
             | people, governments, political candidates, groups for old
             | ladies trading knitting patterns...all of the above!
        
         | yuliyp wrote:
         | Think "hardware groups with next generation experimental
         | hardware". Facebook runs their own data centers, including
         | designing their own servers. How do you verify that the servers
         | actually perform well? One natural choice was to put them to
         | work as web servers. Given its size, there was a lot of tooling
         | there to be able to measure very precisely what the throughput
         | of a web server was, so things like "How many of these new
         | servers should we order" could be answered, in addition to the
         | "does the rack catch fire" questions.
         | 
         | One example of a source of tension that such standards were
         | trying to deal with here was in a group trying to run web
         | servers on machines with SSDs that were way too small:
         | obviously for the bean counter saving a bunch of money on the
         | SSDs was nice, but for the team trying to make sure the disk
         | can fit all the code and logs on it, it was less nice.
        
           | Pxtl wrote:
           | On a certain level I'm kind of impressed how long FB has
           | celebrated that kind of hacker ethos, I just wish it wasn't
           | for... y'know, evil.
        
       | kazinator wrote:
       | You could write a _program_ that will check most of the system
       | requirements, and say you must run this program and have it pass.
        
         | fn-mote wrote:
         | It helps, but it's important to have a name and a green check
         | mark for each result so people understand WHY the magic doesn't
         | permit them to use the machine.
        
         | 12_throw_away wrote:
         | Yep, this is good. On one occasion, I saved myself a huge
         | amount of back and forth by writing a dumb "test suite" for a
         | software artifact to be shipped to a client; i.e. make sure it
         | was compressed as a zipfile; it included an INSTALL.md; it was
         | installable via [specific installer program]; it included
         | [specific runtime libraries versions]; etc. etc.
         | 
         | Turns out human software engineers are really bad at fulfilling
         | even well-specified requirements, so you will always save time
         | by automating as much of the conformance testing as possible.
        
       ___________________________________________________________________
       (page generated 2025-07-16 23:01 UTC)