@hrefna@hachyderm.io
@hrefna@hachyderm.io avatar

hrefna

@hrefna@hachyderm.io

SRE at Google. Queer. Poly :potion_polyamory: Trans :verified_trans: :nonbinary_potion: Engineer. Ace :flag_ace: Member of AWU-CWA. #ActuallyAutistic :rainbowinfinity: #UnionStrong

Opinions my own. Does not suffer fools gladly.

This profile is from a federated server and may be incomplete. Browse more on the original instance.

hrefna, to random
@hrefna@hachyderm.io avatar

Someone said I was a casual gamer because I was finding the core gameplay loop (the process of leveling, etc) of WW to not be something I was enjoying and both my girlfriend and I were like… whut.

There's nothing wrong with being a casual gamer, and also. What even. What definition of 'casual' are you using here, exactly, given that I play games a whole hell of a lot more difficult than WW.

hrefna, (edited ) to random
@hrefna@hachyderm.io avatar

The hardest part of building a quality filter for the fediverse has very little to do with the scoring system.

There are massive social problems and a host of tedious technical problems, but the actual scoring system for the filter itself? Not terribly difficult.

Like, sure, it's difficult to optimize and get right in the long run and there are gotchas along the way, but fundamentally all of that is gated on the rest of the system behaving and the social problems being solved.

hrefna, to random
@hrefna@hachyderm.io avatar

I do not recommend spraying yourself in the face with herbicide, just as a general rule.

hrefna, to random
@hrefna@hachyderm.io avatar

Any time someone says "there's no algorithm here" I enter into an internal war between two thoughts.

The first is that there absolutely is an "algorithm here." That's not pedantry, that's a form of technical correctness that is highly important and informative.

The second is that you most likely really want that algorithm to be more aggressive than it is today. No, seriously, you almost certainly do.

We seem to be stuck in this place where people refuse to learn from our predecessors.

hrefna, to random
@hrefna@hachyderm.io avatar

Evidently a few men have tried "would you rather talk to a woman or a tree about your feelings" as kind of a weird "clapback" to the man vs. bear thing and I'm like… you'll talk, voluntarily, out loud, about your feelings and not make it some poor random non-therapist's problem? Really?

Please do men! Go forth! Spill your secrets to trees. May you find it to be very fruitful.

hrefna,
@hrefna@hachyderm.io avatar

Also just.

As it pertains to the "man vs. bear" thing I can only go back to Margaret Atwood:

"Men are afraid that women will laugh at them. Women are afraid that men will kill them."

hrefna, to random
@hrefna@hachyderm.io avatar

I'm begging men to actually read articles/studies/whatever before posting them as replies in some sort of attempt at a gotcha.

Absolutely begging. Please. For the love of all that's holy. Don't just look for the first article with a headline that you think will work as a gotcha. ACTUALLY READ THE ARTICLE/STUDY.

hrefna, to random
@hrefna@hachyderm.io avatar

Stop sharing things that discourage people from donating to charity because you saw in a meme that a corporation is doing something that they aren't actually doing.

Thanks.

hrefna, to random
@hrefna@hachyderm.io avatar

What does UC think is going to happen here, exactly?

Go get 'em.

#UnitedAutoWorkers Local 4811

#unionStrong #solidarity

hrefna, to random
@hrefna@hachyderm.io avatar

An example of my problem:

Arkansas is ranked above Colorado. Alaska is one of the top 4.

Arkansas does not have discrimination protections for gender orientation or identity. It does not have a hate crime bill and the "watered down" hate crime bill it does have does not include gender orientation or identity "implicitly or explicitly." They ban trans people in athletics.

They passed a bill to allow healthcare practitioners to deny healthcare to LGBTQ patients.

1/
https://hachyderm.io/@hrefna/112576215889428781

hrefna,
@hrefna@hachyderm.io avatar

Alaska is not quite as extreme, but is also pretty bad.

Their board of education bans trans individuals from womens sports, they don't have a hate crime law that protects us, etc.

They don't have universal anti-discrimination laws in housing.

The list goes on.

Now let's compare and contrast with Colorado.

2/

hrefna,
@hrefna@hachyderm.io avatar

Colorado bans discrimination, it has a hate crime law, it doesn't restrict trans people from sports, it bans conversion therapy in minors, it doesn't have a "bathroom bill," it bans the "gay panic defense."

It includes gender affirming procedures in its benchmark health insurance plan for essential health benefits.

People can even come from other states without fear of prosecution here.

Our per capita reported hate crimes is higher, but emphasis on the word reported.

3/

hrefna,
@hrefna@hachyderm.io avatar

If you have a model and it puts Alaska and Arkansas above Colorado, you need to treat that as a signal that your model is so wrong as to be no longer usable and you need to seriously question whether to share it with others, at least without massive caveats.

It is unconscionable—data malpractice—for a group to release a report like this.

4/4

hrefna, to GenshinImpact
@hrefna@hachyderm.io avatar

Snerk. Someone I know came up with this and then this got put together as a result.

hrefna, to random
@hrefna@hachyderm.io avatar

There's a chart going around either from or citing that "ranks" LGBTQ "safety grades" by state.

It's an object lesson in why the assumptions behind your data matter significantly and in how not to communicate data.

First red flag: there are no dates anywhere on it.

Second red flag: A large chunk of its data comes from hate crime reporting rates.

What is a hate crime and how often do things get filed as such matters significantly for these things.

This misleads.

hrefna,
@hrefna@hachyderm.io avatar

There are a slew of reporting biases (and other biases) and an entire set of social constructs around whether something is considered a "hate crime" by state.

You can't just look at the numbers and get a full story and, even if you did, you would need to absolutely under every circumstance report the year the data was collected

Not doing so is data malpractice.

Now maybe SafeHome did exactly that, or maybe this is just some random person's interpretation based on their data. That isn't clear.

hrefna,
@hrefna@hachyderm.io avatar

Oh cool, it actually is from their official sources: https://www.safehome.org/data-lgbtq-state-safety-rankings/

This is a very, very poor methodology and it heavily misleads. Their presentation of data in this way is basic data malpractice.

In the flavor text Alaska gets brownie points for considering a bill that would add protections, but that doesn't seem to weigh for sanctuary states.

They used the 2022 FBI's Hate Crime Statistics report uncritically and, uh, no. Bad company, no cookie.

hrefna,
@hrefna@hachyderm.io avatar

See also:

https://www.npr.org/2023/01/01/1145973412/researchers-say-the-fbis-statistics-on-hate-crimes-across-the-country-are-flawed

hrefna, to random
@hrefna@hachyderm.io avatar

Thanks, I hate it.
https://hachyderm.io/@nedbat/112569940156009404

hrefna, to random
@hrefna@hachyderm.io avatar

It'd be very convenient if I could negotiate with ants.

There are a group of western harvester ants in a rather inconvenient spot and I'd love if they could move like… 20' over. Or a few hundred feet in any of three different directions.

But right where they are is not ideal.

hrefna, to random
@hrefna@hachyderm.io avatar

One idea I've toyed with before that I think is really interesting is the idea of having a persistent inbox but an offline processor

So think of it like this:

You have a queue that sits online with a webserver in front. The queue has some logic (the kind you can configure with RabbitMQ) for deduping and routing, but is still just a queue.

That's the only persistently online piece.

(@jenniferplusplus mentioned this earlier as well, this is not unique nor original to me as a thought)

1/

hrefna,
@hrefna@hachyderm.io avatar

To start with, the idea that your inbox and outbox are readable is an assumption that is no longer valid.

We probably also need some method of signing objects with a client signature.

You will likely lose a lot of flexibility in what clients connect to it and there's not a good way forward here.

We would need some norms around backpressure, how to handle it, and what to do when you can't reach the queue or the queue tells you to go fish.

Oh, and we won't be able to cache media files.

5/

hrefna, (edited )
@hrefna@hachyderm.io avatar

Then there's the matter of backups.

Maybe all of this is okay!

That's why we talk about tradeoffs, and not every system needs the same tradeoffs.

But it's an example of an interesting technical problem and a hard social problem to solve. It's also a good thought experiment to think through what assumptions we would need to challenge.

Is not caching media files okay? Could we use a camo proxy instead? Would a CDN be more appropriate? Is it okay if ids aren't public? etc.

6/6

hrefna, to trans
@hrefna@hachyderm.io avatar

What the hell about this woman's background makes her suited to be an "independent investigator" of issues?

https://en.wikipedia.org/wiki/Hilary_Cass

There's literally nothing here that ties into any of the fields that are relevant. Her specialization is all in developmental and mental disabilities in high-support-need younger children.

"Independent" evidently means "so far out of her lane that she can't see her lane on this side of the horizon."

hrefna,
@hrefna@hachyderm.io avatar

Overall, this section discussing possible factors which have influenced the patient profile shows several issues in providing evidence to substantiate its claims. For several key assertion, single studies are cited when a wealth of contradictory studies are available, showing a lack of balanced consideration of the literature. This calls into question the robustness of the listed conclusions of this section (8.52-8.62), and any recommendations the Review makes based on these.
All in all, this commentary raises numerous concerns regarding the biological and psychosocial evidence in the Cass Review. These concerns include inferring trends and causality about demographic trends and comorbidities within statistical substantiation, misrepresenting results from the literature, varying thresholds for the inclusion of studies and using unbalanced evidence or references to make one-sided claims. Together, these concerns call into question whether the Review is able to provide sufficient evidence to substantiate its recommendations to deviate from WPATH’s international standard of care for trans children

hrefna,
@hrefna@hachyderm.io avatar

Grijseels, D. M. (2024). Biological and psychosocial evidence in the Cass Review: a critical commentary. International Journal of Transgender Health, 1–11. https://doi.org/10.1080/26895269.2024.2362304

Overall, this section discussing possible factors which have influenced the patient profile shows several issues in providing evidence to substantiate its claims. For several key assertion, single studies are cited when a wealth of contradictory studies are available, showing a lack of balanced consideration of the literature. This calls into question the robustness of the listed conclusions of this section (8.52-8.62), and any recommendations the Review makes based on these.
All in all, this commentary raises numerous concerns regarding the biological and psychosocial evidence in the Cass Review. These concerns include inferring trends and causality about demographic trends and comorbidities within statistical substantiation, misrepresenting results from the literature, varying thresholds for the inclusion of studies and using unbalanced evidence or references to make one-sided claims. Together, these concerns call into question whether the Review is able to provide sufficient evidence to substantiate its recommendations to deviate from WPATH’s international standard of care for trans children

hrefna,
@hrefna@hachyderm.io avatar

WPATH. WPATH and USPATH Comment on the Cass Review. 17 May 2024.

https://www.wpath.org/media/cms/Documents/Public%20Policies/2024/17.05.24%20Response%20Cass%20Review%20FINAL%20with%20ed%20note.pdf

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • JUstTest
  • mdbf
  • ngwrru68w68
  • tester
  • magazineikmin
  • thenastyranch
  • rosin
  • khanakhh
  • InstantRegret
  • Youngstown
  • slotface
  • Durango
  • kavyap
  • DreamBathrooms
  • megavids
  • tacticalgear
  • osvaldo12
  • normalnudes
  • cubers
  • cisconetworking
  • everett
  • GTA5RPClips
  • ethstaker
  • Leos
  • provamag3
  • anitta
  • modclub
  • lostlight
  • All magazines
    •