Dana Fried

Staff engineer, Google Chrome. Roller derby athlete. Mom. Gamer.

Warning: crush hazard

Public Key

npub1syuuldp8kpkhpy3ugeatr66u2teht3ynl8h6ynvdv0z7ywmy7c5qctel2e

NIP-05 Address

tess_at_mastodon.social@mostr.pub

Profile Code

nprofile1qqsgzww0ksnmqmtsjg7yv743adw99um4cjflnmazfkxk830z8dj0v2q38xmc4

Author Public Key

npub1syuuldp8kpkhpy3ugeatr66u2teht3ynl8h6ynvdv0z7ywmy7c5qctel2e

Show more details

Last Notes

2024-10-20 08:05:53

If (like me) you use #ASMR videos as relaxation aids, you may have noticed that the audio quality of some videos has become inconsistent or even jarring lately.
YouTube has introduced a normalization setting that forces all volume into a single more listenable range, like most broadcast radio uses. It's great for many videos, but terrible for ASMR (and probably classical music as well).
To disable it, tap the gear icon, go to additional settings, and turn off "stable volume".

2024-10-05 16:39:49

- reply

@npub1rv4…ft5j @npub1znz…8l64 Tumblr is fascinating but I quit when I realized I'd never be good at the meta

2024-09-29 17:17:44

It never stops at just trans issues, or LGBTQ+ issues, or even abortion - Russia is now cracking down on people who talk openly about not wanting to have children:
https://www.nbcnews.com/news/world/russia-proposes-ban-child-free-lifestyle-rcna172616
This is the exact same movement and thought process that gives us Vance's "childless cat ladies". And believe me, if the natalist fascists in the US could ban the child-free movement, they absolutely would.

2024-09-04 22:45:10

I don't need the 14-year-old who shot 30 people to be "tried as an adult".
I need the system that caused a 14-year-old to have access to a gun and to want to shoot random people at his school to be fixed.
Locking up a child for the rest of his life (or allowing the state to murder him) won't bring anyone back, or heal the injuries of the living victims.
But fixing a toxic culture awash in guns and radicalizing messages aimed at young boys might save future lives.

2024-08-29 22:31:09

- reply

The existential risk is that the incredible repository of nearly all human knowledge that is the internet will be flooded with so much LLM-generated dreck that locating reliable information will become effectively impossible (alongside scientific journals, which are also suffering incredibly under the weight of ML spam).
The existential risk is that nobody will be able to trust a photo or video of anything because the vast majority of media will be fabricated.

2024-08-18 00:18:37

- reply

Kill the tiktok cop in your head.

2024-08-18 00:16:41

- reply

And gods help you if you do it here on fediverse, because I will block you.
Euphemisms are filter-evasion. They're *worse* than noting on non-algorithmic sites like this.

2024-08-18 00:15:14

Fuck everyone who uses "unalive" with zero sense of irony. Fucking socmed algorithm brain poisoning.
Say "kill". Say "murder". Say "suicide". Say "pedophile". Say what you mean. If people need to not see or hear those words they'll use a filter.
(Or if you're on a platform and you feel like you're forced to elide a word, make it obvious you know you're being censored and don't use stupid euphemisms like it's a totally normal thing to do.)

2024-08-16 17:32:43

We recently "upgraded" from polymer3 to lit and ".html.js" is an abomination that should never have existed.

2024-08-13 01:13:17

Found out that in order to post to subreddits for topics I'm interested in I first need to go on a rep farming expedition, and while I understand, I am done with social media platforms that require me to take up a second job in order to get any engagement.

2024-07-31 18:16:02

Hear me out:
VICE PREISDENT GRITTY
https://files.mastodon.social/media_attachments/files/112/882/467/067/510/560/original/1152383e46c2a3ec.webp

2024-07-28 20:26:40

- reply

@npub1x76…2qpx there needs to be a cooling-off period for the *chan-to-programmer-socks pipeline where you're not allowed to post online until you figure out how to be normal

2024-07-28 05:42:31

- reply

@npub1lrv…tth9 I want to believe he knowingly spread the obviously fake video because the alternative is much worse

2024-07-17 19:01:29

"Solidarity is a conspiracy by woke billionaire CEOs" is kind of a wild take to hear from a union leader, huh.
Someone needs to figure out what is melting the brains of the rich and powerful in the US. (Honestly, it's probably Twitter.)

2024-07-14 16:28:31

In male dominated fields, the responsibility for creating spaces that are safe for women to work in falls on men.
If you don't understand why that is, you're exactly the kind of person I don't want to work with.

2024-07-08 19:58:05

- reply

@npub17vj…g088 I used to be pretty cringe, honestly.

2024-07-02 19:31:49

- reply

@npub1xzm…eykm this is good and you should feel good

2024-06-24 21:26:18

Auto industry: we cannot make a car that will brake for a stopped vehicle or pedestrian in the road.
Also auto industry: please allow us to put autonomous cars on the road; trust us, they're perfectly safe!
https://www.theverge.com/2024/6/24/24185107/aeb-nhtsa-reconsider-auto-industry

2024-06-08 15:04:44

The four Israeli hostages freed today will have to live with the fact that tens to hundreds of civilians - many of them children - were slaughtered in the operation that freed them.
And this was after their government backed out of a cease-fire agreement that would have had them come home peacefully.
That's a lot of weight to carry for people who have already been traumatized.

2024-06-03 17:29:57

- reply

@npub1hpc…hjn3 you win the internet

2024-06-03 17:26:59

tfw u walk past a guy and you're like "wow, he smells really nice" and you realize that it must be *that* part of your cycle 🙃
(Does it count as bisexual if it's only for like three days a month?)

2024-05-31 22:48:57

2018: Deeply skeptical of companies plastering rainbow flags all over everything for one month a year
2024: Deeply suspicious of companies that don't plaster rainbow flags all over everything for one month a year

2024-05-30 21:37:12

LB: I know everyone is talking about the *other* news right now, but being able to log onto a government website and just file your stupid taxes without having to buy software or hire an expensive accountant is a *huge* boon for most working Americans.
I'm not currently one of them, but I was for most of my life and this would have been really nice.

2024-05-29 16:37:35

- reply

@npub1su4…a0dw WAT

2024-05-28 20:09:51

After spending the weekend in Portland (a bunch of it on transit) I am thinking more about how we talk about the scourge of homelessness in most American cities.
It's insidious because the people who suffer the most are also the most visible symptom of the problem and therefore end up the targets of policy instead of the actual underlying causes.
How do we (and how do politicians) talk about homelessness in a way that makes people see the homeless as victims rather than the problem?

2024-05-23 04:59:36

Amused at how Altman helped himself to a woman who denied him multiple times, because he was fond of her*, and despite the fact that she was literally the only person in recent history to sue Disney and win - and that no one else in his circle tried to dissuade him (or had enough pull or made enough effort to be successful).
Says a lot about the people at the helm of the "AI revolution".
* or perhaps I should say, "Her"

2024-05-22 20:05:22

A streamer I like got her account hacked, so if anyone knows someone who works over at YouTube well enough to ask for a favor, send me a DM.
The hackers deleted a decade of videos and turned her channel into scam Tesla content (unsurprisingly).
(responses which are not specifically to help me connect with someone on the inside who will help will be muted/blocked)

2024-05-14 17:10:05

- reply

the clusivity is kind of odd; it's "I'm addressing everyone *except* the person I'm talking to" which kind of straddles the line between second and third person

2024-05-14 17:08:02

In which I passionately argue that "chat" is indeed a valid second-person pronoun:
https://www.tumblr.com/leftoblique/750471889313497088/well-now-thats-needlessly-confrontational-were

2024-05-13 20:04:42

manic pixie torment nexus
https://www.theverge.com/2024/5/13/24155652/chatgpt-voice-mode-gpt4o-upgrades

2024-05-04 20:46:42

- reply

@npub17vj…g088 Simon and Garfunkel then helpfully offer... thyme?

2024-05-03 22:36:06

- reply

@npub1tpv…zwsk also really shows how poor cameras at the time were at capturing dark skin.

2024-04-13 22:16:05

- reply

@npub1swy…snpj it feels like Netanyahu has decided the only way to stay in power is endless war and he's just... going for it?

2024-04-03 16:31:04

- reply

@npub1yas…andl okay but like spatial computing is cool (if largely useless and impractical and also making half the population vomit).

2024-03-26 21:09:59

If someone is claiming that any bad outcome is the result of DEI they are saying that they think Black and brown people and women are inferior to white men.
There is no other reasonable way to interpret those statements.

2024-03-21 22:29:57

- reply

@npub1su4…a0dw
"[When messages are presented in those telltale green bubbles,] the conversation is not encrypted, videos are pixelated and grainy, and users cannot edit messages or see typing indicators. As a result, iPhone users perceive rival smartphones as being lower quality because the experience of messaging friends and family who do not own iPhones is worse—even though Apple is the one responsible for breaking cross-platform messaging."
FINALLY SOMEONE FUCKING SAID IT

2024-03-15 07:46:13

- reply

@npub14uk…2erc tesla's business plan makes way more sense when you realize it was just doing arbitrage on government subsidies and a little bit of greenwashing, and was only incidentally in the "sell cars to people" business

2024-03-13 17:12:00

- reply

As far as I can tell, the pronunciation of the development is something like /sʌˈnˀɔ.qʷʼ/, which is delightful because they've managed to include at least two sounds and three phonological elements not found in [General American] English.

2024-03-13 17:03:06

This slaps.
Indigenous people really just said, "our people need housing; fuck your zoning laws".
https://macleans.ca/society/sen%cc%93a%e1%b8%b5w-vancouver/

2024-03-12 00:28:05

Mathematically, a first-past-the-post system will inevitably result in a two-party system (perhaps with the occasional success of a regional party).
There is nothing you can do within the system to change that - you can only vote for the party that better reflects your views (and pressure them to continue to do that).
The thing you do to dismantle two-party politics is advocating for other electoral models: proportional representation; single-transferrable vote or approval voting; etc.

2024-03-08 18:19:04

- reply

@npub1upk…tcwy you're doing great work with these analyses.
Honestly, I kind of wish someone would put out a version of git that just used consistent terminology and command-line flags; it would still be overly complex for source management but at least it wouldn't also be confusing.
(Like, why does git add -A add everything including unversioned files, but git commit -a only adds modified files? Why is one -A and the other -a, and there's no git add -a or git commit -A?)

2024-03-05 03:55:33

- reply

Libs and Leftists Stop Letting the Right Play You Challenge 2024 (difficulty: impossible)

2024-03-05 03:53:52

If there are "AI"-generated images designed to mislead people about the upcoming election, then posting them all over social media as examples of how awful the other guys are is actually doing their work for them.
At the very least mark the images as sensitive.

2024-02-26 10:34:11

- reply

@npub15x4…3p3l "I sat down with Chaya Raichik--"
Lemme stop ya right there, chief.

2024-02-24 23:31:32

- reply

@npub1q9m…zgdu it's interesting because LLMs self-poison if they ingest too much of their output; also more people are using image poisoning tools.
So, there may come a point where we can no longer safely train models on new data.

2024-02-24 03:05:56

AI is biased because reality is biased.
ML gives you a picture of the world as it is - or rather the subset of the world in its training data, the selection and labeling of which also introduces bias.
All ML can ever do, therefore, is reproduce existing biases; it does this automatically and by default.
ML can never be made unbiased because you would have to account for every structural inequality specifically. At that point you might as well just have a person do it.

2024-02-22 21:44:11

I want to address some of the differences between AT and AP listed here:
https://bsky.social/about/blog/02-22-2024-open-social-web
(a thread)

2024-02-22 00:14:01

- reply

Also, like, a lot of people who want EVs live in cities and denser environments where it's expensive to live and a pain in the ass to park and maybe your $100K electric monster truck doesn't fit with their lifestyle.
Stop trying to make EVs that appeal to the kind of people who would rather be rolling coal in a modified F450.

2024-02-22 00:12:11

If your EVs aren't selling maybe that's because you're making giant fucking trucks and SUVs and hyper-expensive luxury sedans and the kind of people who want EVs mostly just want a compact car for getting to work and running errands.
Just sayin'.

2024-02-15 05:25:33

- reply

Is it surprising that the CCP intervened to censor works they didn't like? No.
It *is*, however, surprising how eager the North American administrators were to take the initiative in removing works with what seems like very little prompting - not only reading through the works in question, but also trolling through the authors' social media to see if they had ever said anything the host country might take umbrage at.

2024-02-15 05:22:16

A new report shows that American and Canadian administrators for the Hugo Awards preemptively researched and disqualified works and authors they deemed as potentially upsetting to the Chinese government from the list of nominees.
It also seems as if the Chinese government may have directly removed a number of other works.
https://file770.com/the-2023-hugo-awards-a-report-on-censorship-and-exclusion/

2024-02-14 08:10:05

The issue with the bsky bridge isn't that it's opt-out; as people have pointed out much of fedi is opt-out by default.
The problem is: who do you send a moderation report to? The problem is: it's a server which most fedi servers will have to immediately block for basic policy reasons. The problem is: it's a technical solution to a cultural divide.
And, unless things like blocking, muting, and notifications are done very carefully, it will be easy for bad actors to exploit.

2024-02-13 21:03:35

- reply

@npub1cx7…8tc2 @npub1cmg…rzd2 I mean, you can always just post a hard link to a public post, or screenshot.
The big thing is to make the easy path the one with the protections, which forces people to be actively malicious instead of just casually cruel, and also to make the easy path nudge people towards prosocial behavior.
This is also why I continue to advocate for bottom text on QBs, because it encourages "yes, and..." and discourages "hey look at this asshole".

2024-02-13 20:37:12

- reply

@npub1cmg…rzd2 @npub1cx7…8tc2 perhaps if the link was encrypted and went through the server? There has to be a way to do it; whether it's practical or efficient is another question.

2024-02-12 18:44:34

If you told me a year ago that right-wingers would be upset because a football team from red state middle America beat a football team from San Francisco and one of the players on the winning team was a straight white man who was dating a famous, conventionally attractive straight white woman country singer...
I would believe you, because the Right is irrational and conspiracy theories don't have to make sense.
But it's also a useful demonstration of how fascism eventually comes for its own.

2024-02-06 19:36:36

luv 2 b told that the "future of computing" is always wearing an uncomfortable headset that takes away my peripheral vision and makes me nauseous
honestly this will have me praying for elon to put a chip in my head so I can finally just die

2024-01-31 22:09:04

iRobot facing "death" because it only has 46% of the market is bonkers to me; capitalism has forgotten what to do with a mature company.
"We sell vacuums to half the population" is no longer a valid business model, evidently.
https://www.theverge.com/2024/1/31/24057077/irobot-amazon-deal-collapse-roomba-uncertain-future?utm_content=buffer04ff0&utm_medium=social&utm_source=bufferapp.com&utm_campaign=buffer

2024-01-31 19:54:26

Normally I see people being ridiculous on social media over israel-palestine by equating zionism with judaism (which, as a Jew, sucks).
But a new thing I'm seeing is people defending israel sending special forces dressed up as doctors to assassinate militants in a West Bank hospital (because the guys they killed were "bad")...
How do I explain to people that you don't get an exception from the Geneva Convention because the dudes you war-crimed were naughty?

2024-01-26 22:27:44

The state should not be in the business of murdering its own residents (or preferably, anyone).
However, I'm seeing a lot of really stupid arguments over why N2 anoxia is *specifically* a bad or inhumane choice, and, they are at best deeply misinformed.
The reason inert-gas anoxia is bad is because it allows people to believe that executing a human being could ever be a clean, tidy, "humane" act. Guillotine is equally humane but messy; you can't look on pretending you're not watching a murder.

2024-01-24 18:16:21

It was not just one mistake to hold the Hugo Awards in Chengdu this year.
The first mistake was holding a con in a place where acceptance speeches could be punished for illegal speech.
The second mistake was holding a con where the government could disqualify candidates.
The final mistake was holding a con where - when the government inevitably put its thumb on the scale - the organization couldn't even acknowledge what had happened for the safety of its members.
https://www.theguardian.com/books/2024/jan/24/science-fiction-awards-held-in-china-under-fire-for-excluding-authors

2024-01-23 20:13:43

Jaywalking is a made-up crime created by car company lobbyists to make it more appealing to drive than to walk.
In WA, all intersections are already crosswalks, but using an unmarked crosswalk is dangerous because drivers either don't know or don't care. 90% of what is perceived as "jaywalking" actually happens at legal crossings; the rest is on intentionally hostile roads, often in poor neighborhoods
It should be drivers' responsibility not to hit pedestrians, period.
https://komonews.com/news/local/washington-jaywalking-laws-crosswalk-citations-free-to-walk-sb-5383-decriminalization-disproportionally-impact-minorities-homeless-traffic-sponsors-saldaa-liias-nguyen-valdez-wilson-politics

2024-01-23 17:27:34

Broke: Ryan Gosling getting an Oscar nomination for Barbie but not Margot Robbie is proof of Hollywood's sexism.
Woke: The inability of the industry to see men in anything other than a leading role, which created a vacuum that allowed Gosling to sneak into the nomination list, is proof of Hollywood's sexism.
Bespoke: The stark difference in screen time and plot relevance between the roles nominated for Supporting Actor and Supporting Actress is proof of Hollywood's sexism.

2024-01-22 19:07:49

I just had what I think is an important realization:
A long time ago, when the term "anchor babies" was being wheeled out to demonize the children of immigrants, conservatives were also complaining that immigrants were not assimilating.
(Strap in, this might be a long one.)

2024-01-20 07:14:52

Want to give a shout-out to @npub1u99…xqru for having an amazing handle; 10/10, would use it as a derby name if I didn't have one already

2024-01-19 19:05:12

- reply

If you're worried that your images have already been scraped for training, be aware:
Training sets tend to be metadata-and-URL-only; nobody actually packages and distributes terabytes of (often copyrighted!) image data,
So, if you replace existing images with nightshade+glaze, any new models will receive the altered payload.

2024-01-19 18:55:21

PSA for visual artists:
Nightshade is out!
https://nightshade.cs.uchicago.edu/
Nightshade is an ML-poisoning algorithm that causes models trained with treated images to malfunction. Use in conjunction with Glaze to protect your work and give a big "fuck you" to the capitalists stealing your shit for their own profit.
(Feel free to throw this on *any* image you put on the internet; it doesn't just work on art.)
#mastoart

2024-01-19 17:02:31

- reply

@npub1alq…mnj6 @npub1fx5…kq6u @npub10cr…q72n
I want to be clear: this response is fairly damning. And I don't say that to be mean.
Treating accessibility as an afterthought or assuming it's easy ("just make sure the fonts are big enough and the contrast is high enough!") is why there's such an incredible problem with it in the industry.

2024-01-19 16:55:36

- reply

@npub1fx5…kq6u @npub1alq…mnj6 @npub10cr…q72n I worry about how accessible the pages that are created by these tools are.
Because if they're not, they're hurting low-vision users and putting businesses who use these tools at risk of significant ADA liability.

2024-01-16 20:47:09

Thing I just wrote to a co-worker who asked me if something was technically possible:
"This is a greased rabbit hole with spikes at the bottom. Feel free to slide down and see what happens, but don't say I didn't warn you."

2024-01-10 17:43:45

- reply

@npub1alq…mnj6 @npub1lla…2zhe @npub10cr…q72n I find #2 hilarious because tests already constitute 80% of the code I write, so having an LLM write the interesting 20%* would make my job completely miserable to do.
* and then spending just as much time debugging the errors

2024-01-10 16:52:39

Found on Tumblr (via homoluigi):
https://files.mastodon.social/media_attachments/files/111/732/686/792/565/999/original/bb6d6c4c3631828a.jpg

2024-01-08 21:54:53

OpenAI complaining that NYT unfairly "tricked" ChatGPT into regurgitating an entire NYT article misses the point.
Yes, it's bad that they've copied a bunch of copyrighted material to train their models. No, "Maybe we got it from a third-party site that also pirated the article?" isn't a defense.
But most concerningly, GPT's ability to regurgitate entire articles means that the GPT model being distributed by OpenAI contains and distributes copyrighted work its creators don't have rights to.

2023-12-31 02:48:14

- reply

@npub1ye6…6a2z seeing the time-lapse at something like 2,000,000x makes you realize how slow light really is on even interstellar scales.
Spacetime seems empty and flat and smooth from our perspective, but on a galactic scale it's basically molasses.

2023-12-26 20:25:41

It's cool to shame people for antisocial behavior; that's one of the ways we establish community norms.
But that works both ways - if you try to shame someone for perfectly acceptable behavior because you're a judgemental, intolerant asshole, you're gonna get those same fingers pointing right back at you. That's *also* how we establish community norms.

2023-12-24 01:04:47

- reply

If you feel bad/awkward writing triggery words out, think about how the threat of shadowbanning or downranking on corporate platforms has made you afraid to speak openly about difficult topics, and how nice it is to be on a platform where we can choose how we protect others and ourselves from potentially triggering or upsetting posts.
It's time to unlearn the things other platforms have forced us to internalize.

2023-12-24 01:01:09

On fedi:
- we have CWs
- we have the ability to mute words
- there is no algorithm
Therefore, it is bad form to censor words (e.g. "r*pe", "k*ll") because that evades people's mute lists. These censoring conventions were created to evade algorithmic moderation on other platforms, which, again, we don't have here.
Instead, just use the words and apply an appropriate CW if you think it's needed.

2023-12-22 02:56:21

Does substack allow nazi porn?

2023-12-19 23:28:14

- reply

I'm not gonna dump on the OP because for all I know he's right, but also, man, "I'm gonna use the CentOS of social media and I don't care how it looks or feels" is such a bleak attitude.
Like, y'all hate Papa Johns, fine, but then y'all just sit around eating dry crust and feeling sad. Throw some toppings on there.

2023-12-19 23:20:22

Just saw a post that contained the line: "most of the people on Mastodon are here for ideological or privacy reasons"
I want to be clear: I'm here to talk to cool people and shitpost.
I do care about features, because they make it easier to talk to cool people and shitpost.

2023-12-14 16:17:48

It's been fascinating to see CNN's online coverage of the war in Israel swing completely against the Netanyahu regime in, like, the past week.
I usually see them as a bellwether of where center(-left) sentiment is going because they seem to try not to challenge people's opinions too much.
This (plus Biden's increasingly vocal criticism of the regime and the war) gives me hope that our official policies might actually start to move on Israel for the first time in decades.

2023-12-09 21:07:14

Time to stop sharing images on *checks notes* Instagram?
(and Facebook)
https://arstechnica.com/information-technology/2023/12/metas-new-ai-image-generator-was-trained-on-1-1-billion-instagram-and-facebook-photos/

2023-12-05 02:33:36

One thing I do like about fedi (and maybe it's just m.s?) is that while moderation effort is low it's fairly consistent; I don't see a lot of abhorrent behavior despite having been the target of multiple hate brigades on other platforms that need not be named.
This is not the experience of some folks on smaller instances or who are more, shall we say, visibly minoritized?
But like, there's some consensus that certain behaviors *should* be banned here; these incidents are seen as failures.

2023-12-04 15:35:48

- reply

I am so annoyed by this because there are a TON of clever, space-saving devices which we CANNOT USE because they are made for men and which even I - a serious athlete who has spent literally decades powerlifting and who carries 40-50 lbs of extra muscle on her frame - cannot operate.
You should not need a boyfriend/husband to be able to use the stuff around your house.

2023-12-04 15:34:26

Pet peeve:
Devices with friction-based twist-locking that are too hard to actually twist for even a fairly strong woman.
Product designers, please note: Grip strength is one of the most sexual dimorphic components of human anatomy.
Top female athletes have the grip strength of a mediocre, untrained man.
If you cannot hand the device you are designing to your girlfriend and have her easily open/extend/unlock it (and then close it again securely), you have failed.

2023-12-02 23:37:04

Americans' sentiment about the state of the economy is informed not by purchasing power or wages, but by precarity.
This is because humans are, rightly, risk- and loss-averse.
Also: economists need to start looking at populations and medians and stop bothering with means. (This should be obvious.)

2023-12-01 18:54:13

I see a lot of takes like "Zionism is X" where X is "settler colonialism" or something more unsavory.
Yes, as it's practiced by the Israeli government and promoted by right-wingers in the West it is.
But there's a difference between thinking Jews should be able to return to their ancestral homeland (a common refrain in the religion is "Next Year in Jerusalem") and the idea that historical Judea should be an apartheid Jewish ethnostate.

2023-11-28 00:40:58

"Food prices up 25% since the start of the pandemic" I don't wanna nitpick but it's been 3½ years; that's just 6% inflation; it's just how compound growth works.
6% inflation is a kind of normal pre-Reagan amount of inflation for the United States to experience, and it's also quite tolerable if you have a union negotiating guaranteed raises and cost-of-living increases and a functional minimum wage.
The problem is that Reagan dismantled both of those things.

2023-11-27 21:12:30

Y'all, I think I just figured out a scheme to get rid of all of the Mastodon reply guys.
It's kind of complex, but it involves a fake project to rewrite all of Linux in Rust, permissive software licenses, and possibly forking Edge.

2023-11-25 19:54:18

- reply

It's okay to be an evangelist for good technology.
It's okay to turn off a feature like motion smoothing on their TV where there's really no downside.
But going and changing the software they work with in a way that will affect their workflow materially is not cool. It's the kind of weird "I know better than you" consent violation we see all too often from people here on Mastodon.
(I've also heard of people switching out Windows for Linux and I hope those people enjoy their eternity in hell.)

2023-11-25 19:51:21

I'm seeing some people posting shit like "help your family this Thanksgiving by logging onto their computer switching their default browser from X to Y" and: PLEASE DO NOT FUCKING DO THIS.
Do not switch them from Edge to Chrome. Do not switch them from Chrome to Firefox.
Literally, these are different software environments with different features and access to different data, and if they suddenly cannot log into their bank because they saved their password in Edge, you have just fucked them.

2023-11-09 23:37:50

- reply

@npub178r…qekt @npub10xq…y98f @npub139w…wrrr I mean, visually, I can see significant differences with the two as rendered on my computer screen; they are in no way the same image?
Or is there data embedded in one of the images that causes it to be rendered differently when it's decompressed for display?

2023-11-09 02:20:11

- reply

@npub1mdk…aptk they're not winding down trust and safety (yet) so they might just be seeing if they can operate it without any dev resources.
To be honest, given the deep dark holes filled with scary shit on that platform, I'm not sure the fediverse would want to absorb them wholesale.
But I could see spinning it off into a nonprofit or something?

2023-11-09 02:10:59

Rumblings from inside Automattic that Tumblr may be being wound down.
Makes me really glad that Mastodon exists and that there are options for building communities online that don't depend on VC funding and number-go-up-forever.

2023-11-08 17:33:00

- reply

@npub1rl4…rg8n the host of a podcast I used to listen to thought "devil's advocate" sounded too pat and instead called it "playing kid-toucher lawyer".
Which is what I think of now, any time someone is playing devil's advocate.

2023-11-08 17:25:17

An actual recent exchange:
Them: *a bunch of speculative whataboutisn*
Me: 😒
Them: "I mean, I don't actually think that; I'm just playing devil's advocate."
Me: "The devil has enough lawyers."
Them: ...

2023-11-06 18:37:26

- reply

Also this is why you are better off writing your agreement on the back of a napkin than having ChatGPT draw up a contract.

2023-11-06 16:14:36

The difference between common law and civil code:
https://files.mastodon.social/media_attachments/files/111/364/485/357/645/897/original/4c52af3f97428d1c.jpg

2023-11-06 07:47:18

Attendees at an NFT event in Hong Kong were irradiated with UV-C, suffered skin and eye damage. Organizers apparently installed germicidal fluorescent bulbs instead of black lights.
https://cointelegraph.com/news/bored-ape-eye-vision-problems-apefest-hong-kong-lighting

2023-11-04 16:23:47

Just saw a 30-50 feral hogs joke in the Year of our Lord 2023.
Look, even if you absolutely need to post hog, this is how you should be doing it:
https://www.salon.com/2023/10/27/wild-pig-like-animals-are-tearing-up-an-arizona-golf-course-the-internet-is-delighted/

2023-11-02 20:02:54

I just found out that people I know - smart people - are using ChatGPT to draw up leases and contracts and...
Dear gods, please do not do this. Do not use an LLM to write legal documents.
It may be fine as long as nobody reads the contract and there's no dispute, but the second things go south, you are going to be so deeply fucked.