this post was submitted on 07 Feb 2025
86 points (100.0% liked)

Indigenous

740 readers
26 users here now

Welcome to c/indigenous, a socialist decolonial community for news and discussion concerning Indigenous peoples.

Please read the Hexbear Code of Conduct and remember...we're all comrades here.

Post memes, art, articles, questions, anything you'd like as long as it's about Indigenous peoples.

Chunka Luta Network, CLN linktree

Library

founded 2 years ago
MODERATORS
 

example of the lands lost due to the dawes act

The Dawes Act of 1887 was a post-Indian Wars law that illegally dissolved 90 million acres of Native lands from 1887 to 1934. Signed into law by President Grover Cleveland on February 8, 1887, the Dawes Act expedited the cultural genocide of Native Americans. The negative effects of the Dawes Act on Indigenous tribes would result in the enactment of the Indian Reorganization Act of 1934, the so-called “Indian New Deal.”

It authorized the U.S. to divide indigenous tribal land into allotments for heads of families and individuals, leading to a loss of 2/3rds of land (~100 million acres) over the next 50 years.

The law converted traditional systems of land tenure into a state-imposed system of private property by forcing Native Americans to "assume a capitalist and proprietary relationship with property" that did not previously exist in their cultures, according to historian Kent Blansett. The act declared remaining lands after allotment as "surplus" and available for sale, including to non-Natives.

Between 1887 and 1934, indigenous people lost control of about 100 million acres of land, or about two-thirds of the land base they held in 1887, as a result of the act.

The loss of land and the break-up of traditional leadership of tribes had such devastating consequences that many scholars refer to the Dawes Act as one of the most destructive U.S. policies for indigenous people in history.

Hexbear links

reminders:

  • 💚 You nerds can join specific comms to see posts about all sorts of topics
  • 💙 Hexbear’s algorithm prioritizes comments over upbears
  • 💜 Sorting by new you nerd
  • 🌈 If you ever want to make your own megathread, you can reserve a spot here nerd
  • 🐶 Join the unofficial Hexbear-adjacent Mastodon instance toots.matapacos.dog

Links To Resources (Aid and Theory):

Aid:

Theory:

you are viewing a single comment's thread
view the rest of the comments
[–] Cimbazarov@hexbear.net 13 points 4 months ago* (last edited 4 months ago) (1 children)

I had a bit of eureka moment today about the whole American company performance culture of chasing metrics for the sake of showing off good metrics and how its exactly why American LLM's were beaten by Deepseek. They dump money at the problem and try to squeeze the LLM's for results. But really the only results they need to show are that the metrics are increasing -- so you have things like more compute, more model parameters etc. that becomes a self-fulfilling prophecy in that its going to increase stock price value because investors think its good. You can't throw money at optimization, because optimization means scaling back and re-assessing "ok what are we trying to achieve with this LLM" and that doesn't provide return on investment in the short term.

The companies are treating LLM's just as they do workers -- squeeze them for productivity by assessing them on performative metrics, increase shareholder value by doing what shareholders think are good. Same thing for a layoff. Like there's no evidence that a layoff makes a company better off, other than the fact that the stock price goes up. But the stock prices goes up precisely because shareholders think a layoff will make it go up, and it becomes a self-fulfilling prophecy. So its completely divorced from the reality of "what is this useful for?"

I do like how capitalism contradicts itself in this way, where it can't even provide the optimal production of value without abandoning it. The squeezing of productivity from each individual worker is going to bring failure to a company, just like throwing compute power and metric-driven development at LLM's failed to produce something like deepseek.

[–] BigBoyKarlLiebknecht@hexbear.net 4 points 4 months ago (2 children)

I think this is very accurate.

I do like how capitalism contradicts itself in this way, where it can't even provide the optimal production of value without abandoning it. The squeezing of productivity from each individual worker is going to bring failure to a company, just like throwing compute power and metric-driven development at LLM's failed to produce something like deepseek.

The obsession with metrics has been a massive trend in my experience of software engineering over the last decade, with a corresponding deep trend toward Taylorism (despite that being shown as bullshit in the mid 20th century, if not sooner).

What’s funny to me is that a lot of this is justified by the work in Accelerate/DORA. If you’re unfamiliar, the research (which still happens annually) looks at the correlation between organizational performance, ways of working, and indicator metrics. The data is gathered via surveys that are vibes-based, and analyzed rigorously to identify patterns. The Accelerate book essentially argues: if you want to perform well, focus on adopting these specific practices and ways of working (like CI/CD, trunk-based development, and a culture of learning, etc), and then iterate and pay attention to whether they’re helping you or not. Oh, and here are some indicators - like deployment frequency, lead time for changes, and mean time to recover - that’ll help you understand if you’re making progress.

So of course - execs focus on the indicators like flies to shit, and set targets against them. They also measure them totally differently - not surveying the vibes of people, but measuring data like timestamps on Jira tickets or pull requests. They also add a bunch of pet favorite metrics too despite the book saying "don't measure too many metrics" (teams in my company have to track, measure and report on no less than 19 metrics every 2 weeks). Engineers naturally game them all, as they are being evaluated by them - while the software they're building is continually gets worse. Execs can then simply use the metrics as a cudgel to blame engineers for things being terrible if they're bad, or accuse engineers of gaming them if they're good.

It's funny and sad, because the stuff in Accelerate is really solid - and the book was explicitly written to avoid the behavior it's inspired/been used to justify. The good stuff is sadly doomed in the vast majority of companies as it's not what execs really want. They just want the metrics to play the game.

If you've not seen Ed Zitron's talk on how product (rather than engineering) metrics impact software products, leading to how shit everything now is, it's fantastic: https://www.youtube.com/watch?v=7Slib2bbMs4

Sorry for the boring software TED Talk comrades, this stuff cuts close to home for me kitty-birthday-sad

[–] Cimbazarov@hexbear.net 3 points 4 months ago* (last edited 4 months ago)

Great links, read the substack and watched the video.

I knew about Taylorism before, and I remember when I first read about it I had a "aha" moment as I realized that is exactly how every company operates internally. Even before learning about it, I felt that obsession with metrics had alot of limits. The only thing is I don't have a good answer other than just "increase the value" of whatever it is you are working on, which is subjective, but not necessarily wrong. People's reactions to your product are always going to be subjective.

Another component of Taylorism is dividing the planning and executing among the manager and the worker respectively. This I think is straightforward for manufacturing industries, but in jobs like in tech maybe a bit more muddled, since I do think engineers do quite a bit of planning. Execs just "set the vision" and everyone else is supposed to rally around that trying to understand what the vision is. I think probably the metrics are important to, since these execs are so far removed from the product that they need a way to understand how things are going with as little investigation as possible (these people also get paid a ridiculous amount more than everyone under them for being a glorified metric nazi). Also a funny contradiction I've seen at the workplace countless times, is that the exec is allowed to make vibes-based decision, but workers have to justify everything with metrics. When the exec's decisions are so batshit that even the most obedient workers begin to question it, management says "Oh they are very visionary". Workers are not allowed to be visionary.

There is always this question I had, which is what is the purpose of a tech manager? Sure they can make decisions, lead the team etc. but that can also be done by a worker on the team. All that is needed is to pass the authority down to them. The only thing I could come up with is that, in order to enforce the metrics driven approach, there needs to be consequences for not meeting them and thats what the manager is for -- basically to fire people.

Unfortunately I think we are stuck in this metric-driven culture because people are way to deep into it to think of anything else. There are always new products that come out (like DeepSeek) that "disrupt" these companies, but when they try to answer the question of why did it happen "there" and not "here" they would never question the metric culture. One is that the people making these decisions are the execs who only know to follow the metrics, thats their entire job and they don't want to lose it. The other is that it kind of makes intuitive sense that if you want to know whether something is successful you need something to measure. Maybe a good alternative will come up one day?

Sorry for the boring software TED Talk comrades, this stuff cuts close to home for me

Nah, its great, keep on TED talking comrade. rat-salute

[–] HexReplyBot@hexbear.net 1 points 4 months ago

I found a YouTube link in your comment. Here are links to the same video on alternative frontends that protect your privacy: