Sevoris

@Sevoris@mastodon.social

He/him

My general account for everything on the Fediverse - from programming to design to philosophy and social stuff to politics and making a better world. I want to network and learn stuff!

Also sci-fi, speculative fiction and writing, poking at worldbuilding with a bend towards coming from real science to inspire new scenarios not written about before.

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

treyhunner, to python
@treyhunner@mastodon.social avatar

What are your favorite one liners?

Sevoris,

@treyhunner "TypeError: cannot (operation) type 'Timestamp' with type 'datetime64[ns]'"

blogdiva, to ai
@blogdiva@mastodon.social avatar

"California could soon deploy generative artificial intelligence tools to help REDUCE TRAFFIC JAMS, make ROADS SAFER and PROVIDE TAX GUIDANCE, among other things"

yeah... that'll end well 🙄

Google & Microsoft will be paid $1 for the services but handed a fuckton of personal information linked to taxes and car ownership. that in itself is worth billions to the corrupt techbros running this new AI-is-labor-without-people scam.

https://www.independent.co.uk/news/california-ap-gavin-newsom-sacramento-google-b2542452.html?utm_source=press.coop

Sevoris,

@blogdiva I know what would reduce traffic jams. Less roads and thus less cars.

You don‘t need generative models to deal with induced demand jfc

cstross, to random
@cstross@wandering.shop avatar

5/8: When writing technical or scientific detail, how much is plausible vs imagined?

Correctly depicting the background details in an SF/F story makes the utterly imaginary conceit at the heart of the story more plausible—it gains believability by being presented in a realistic context. (The best present-setting SF should leave you unsure how much, if any of it is imaginary.)

Sevoris,

@Hcobb @khleedril @cstross ( @nyrath ) starfleet missions go off the rails predictably enough that some time ago, communications engineers installed a time series model that can reliably predict which officer needs communications routed where at what point.

cstross, to random
@cstross@wandering.shop avatar

This is a very possible outcome that the TESCREAL true believers are outraged by but can't refute. (We might get true AI … or we might be chasing an illusion caused by our false intuitions about how minds work.)
https://mastodon.social/@Lazarou/112389710943586505

Sevoris,

@trochee @Califury @cstross tbqh a lot of the contemporary models don't seem much like "building" as much as "playing around with a lot of computing power and massive copyright infringement".

Considering we know how much our brains are filled with specialist structures that reliably locate specific processes in specific volumes, you would think we would pay much more attention to the functionalities of specific neuronal structures and replicate them similarly in "artificial minds"...

nyrath, to random
@nyrath@spacey.space avatar

https://www.tumblr.com/sabotpetal/748045998640791552/f-117-nighthawk-in-production

Sevoris,

@nyrath sci-fi art ought to have a little bit more love for the skeletal of avionics…

grimalkina, to random
@grimalkina@mastodon.social avatar

Look, when computer science departments have some of the worst learning outcomes of any department on campus maybe they SHOULDN'T be elevated as the experts on how students should "learn with AI" with the only reason being "computers"??? Just saying. I will gladly listen to the absolute heroes in CS who HAVE centered teaching and ARE incredible teachers but I guarantee their colleagues aren't.

Sevoris,

@trenner @grimalkina To be fair, computer science != software developer

But that kind of feeds into the same problem: just because you have studied means to solve problems algorithmically, does not mean you have a good grasp of what the problem to be solved is, or how your solution interacts with people…

Sevoris,

@grimalkina @trenner reflecting on this for myself… how people teach and learn computers is an excellent kind of „desire path“ sort of inquiry into what patterns of design, capability and outcomes are desired by people. That’s a treasure trove of information about ontologies, epistemic processes, and generally worldviews and problems.

nyrath, (edited ) to random
@nyrath@spacey.space avatar

Scifi writers should know that an army marches on their stomach. Also amateurs talk about tactics while experts talk about logistics.

You gotta feed your armies and starfleets fuel, ammo, rations for the military humans or everything comes to a screeching halt.

But I had no idea the number of different types of naval logistic ships.

https://en.wikipedia.org/wiki/Military_Sealift_Command#Fleet_Oiler_Program_N031

Sevoris,

@nyrath one of my favorite vibes I derived out of „let’s do the mass effect kinda drives, but add more realism and the dimension of propellant fraction“ was that „main battle craft“ had the logistical endurance of „one solar system’s worth“ and could only project power over interstellar distances lugging the long tail of a bunch of tankers and supply vehicles…

Meanwhile „cruisers“ are often oversized but mainly storage and tanks, pushing a comparatively tiny mission module…

Sevoris,

@nyrath more generally there is something satisfying about a 400 meters murderhobo being able to glass continents while the kilometer-sized giant lozenge only has a few guns and mainly exists for awareness.

baldur, to random
@baldur@toot.cafe avatar

React, Electron, and LLMs have a common purpose: the labour arbitrage theory of dev tool popularity: https://www.baldurbjarnason.com/2024/react-electron-llms-labour-arbitrage/

Sevoris,

@baldur another post that is really setting off thoughts. Great writing to place impulses for thinking.

grimalkina, to random
@grimalkina@mastodon.social avatar

Opening my personal laptop to remember the debris of many tabs open to interesting papers from two weeks ago on a totally different topic than what I'm doing now 😭

The intellectual whiplash you experience as a knowledge worker is so much sometimes. "Interruption" is way too weak a word. You have these entangled threads and building little architectures in your mind for weeks or months sometimes. Long-form novel drafting is the closest thing to some of it in my experience

Sevoris,

@grimalkina definitely have experienced that. And when I come back a lot of the stuff doesn‘t feel relevant anymore, but it was relevant at some point, so maybe I should keep it around?…

I really want to build some way at some point to bottle up all these web pages and open (pdf) papers together with one’s notes about them etc and be able to treat them as these still-connected but "this is not relevant right now and I want to store it elsewhere" things - sometime, at some point.

Sevoris,

@grimalkina Kinda same. Bunch of zotero folders, notes in Obsidian and Logseq. I‘ve gotten better at it. But that‘s so much more intentionality than just Pressing A Button to make a fresh canvas, but being able to draw up the taps and notes back up Later.

Maybe that makes it too easy and causes worse information overload, but I‘ve certainly experienced interruptions and the assumption that "I will just come back later and pick right up where I left" doesn‘t really work always.

Sevoris,

@grimalkina Maybe, maybe not? Information overload is certainly real but I have never really bought into the whole "no FOMO" mindset. And having digital breadcrumbs has helped me a lot, so I do want to figure out how far it can be pushed before it stops working well.

grimalkina, to random
@grimalkina@mastodon.social avatar

deleted_by_author

    •
    Sevoris,

    @grimalkina learning about adultism and its first patterns sometimes gives me the irrational idea that how we abuse kids is a latchkey to many, many things broken in our society.

    trochee, to random
    @trochee@dair-community.social avatar

    The (apocryphal?) stories about how indigenous people complained that "cameras will steal your soul"

    are looking more and more like

    the (apocryphal?) stories of ignoring "ancient Indian wisdom" and getting murked by a curse or a wendigo

    https://restof.social/@restofworld/112366041705469261

    Sevoris,

    @trochee I do not understand how the term really came to be or how true it is, but:

    Well, the observation of how the method to capture people’s appearance in some format would later or sooner be turned to a means of power, abuse and exploitation is a pretty direct extrapolation if

    1. you assume the tech will keep on becoming more capable
    2. you have observed the patterns of capture and exploitation of colonialism

    so really… just an astute observation at basis, no?

    grimalkina, to random
    @grimalkina@mastodon.social avatar

    Reading a 2016 paper on inquiry-based learning where "cognitive load" is once again used as a reason to stop any learner from doing anything FUN and HARD and EXPLORATORY my God

    Being alive increases your cognitive load. We are not in some kind of robotic fry-out state every time we have to consume novel information. Working in a repetitive assembly line is low cognitive load and also destroys people

    This one little tiny concept taken out of context is used for so many bad arguments.

    Sevoris,

    @grimalkina hypothesis: it should not be about "reducing cognitive load" but "making cognitive load manageable" - as in, giving the person the ability to make the most effective choice about the load they experience.

    No undue burden, but also the explicit expansion of the action space available to you. If you want more load... you should be able to do that, and at that point the tool should not be in your way either. (?)

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

    Man you know what I haven’t heard about in AGES

    that weird libertarian text editing app whose name escapes me and had ugly design

    it really seemed to have a Moment, then Obsidian took over all its niches in my cultural view

    edit: @jennwrites reminds me it was Roam

    Sevoris,

    @danilo Roam was libertarian?

    I feel like I missed a whole slice of cultural lore there…

    grimalkina, to random
    @grimalkina@mastodon.social avatar

    I had this morning coffee thought that in many places in tech our odd distorted centering-of-product language betrays our implicit beliefs about agency and decision-making

    Like "users of the x product" instead of "people deciding to use x product"

    Overall the passivity and product centricity in how UX gets talked about has always bothered me mightily. Like this shared delusion we maintain

    Sevoris,

    @grimalkina Oh wow.

    I just read "the aura of care" and came away from this article with this feeling of "a lot of modern UX talk is about stripping people of agency and turning their minds against them instead of extending them" and in comes this.

    Major neuron activation.

    baldur, to random
    @baldur@toot.cafe avatar

    “Spotify CEO Daniel Ek surprised at negative impact of laying off 1,500 Spotify employees | Fortune”

    I’d say tech execs are clowns who don’t take their job seriously and are just spouting nonsense for the stock market while their orgs burn…

    But genuine clowns would actually do a better job https://archive.md/wdyDS

    Sevoris,

    @baldur I don’t know where I have read this before, but these kind of statements sound like an incredibly loud admittance of incompetence to me.

    grimalkina, to random
    @grimalkina@mastodon.social avatar

    This post from Corinne Canavan on LI is the best possible response to a paper I've ever experienced and what a testament to @CSLee's work & impact

    "It’s impossible to overstate just how impactful Carol Lee’s research study on Code Review Anxiety has been for me."

    "In the few days since [the paper was shared], my copy has been lovingly highlighted, marked up, brought to therapy, blasted in an email to coworkers, and talked about incessantly at home."

    https://www.linkedin.com/posts/corinnecanavan_osf-activity-7187208947903160321-R1GB

    Sevoris,

    @grimalkina @CSLee „lovingly highlighted“ !

    „Marked up“ !!

    „Brought to therapy“ :BlobhajPrideHeart:

    „Blasted in an email to coworkers“ :patcat: :patcat: :bongoCat:

    „And talked about incessantly at home“ :BlobhajBlanketSlate:

    quinta, to random Italian
    @quinta@mastodon.uno avatar

    Oxford shuts down institute run by Elon Musk-backed philosopher https://blog.quintarelli.it/2024/04/oxford-shuts-down-institute-run-by-elon-musk-backed-philosopher/

    Sevoris,

    @SparkIT @quinta considering that to my understanding SBF was himself recruited by Longtermist philosophers at university, all working as planned in the eyes of the instigators.

    baldur, to random
    @baldur@toot.cafe avatar

    “GitHub comments abused to push malware via Microsoft repo URLs”

    JFC. https://www.bleepingcomputer.com/news/security/github-comments-abused-to-push-malware-via-microsoft-repo-urls/

    Sevoris,

    @baldur

    "When leaving a comment, a GitHub user can attach a file, which will be uploaded to GitHub's CDN and associated with the related project using a unique URL in this format: 'https: //www.github.com/{project_user}/{repo_name}/files/{file_id}/{file_name}.'
    Instead of generating the URL after a comment is posted, GitHub automatically generates the download link after you add the file to an unsaved comment"
    This, uh, generally does not feel like a good design decision?

    quixoticgeek, to random
    @quixoticgeek@v.st avatar

    We're gonna start to see this a whole lot more in the coming years. Right to repair is great, but when software support ends, we're kinda screwed.

    Avoid buying smart TVs. Consider something like a Chromecast attached to the hdmi port, at least then it's a lot less e waste when the Chromecast goes obsolete, and needs replacing.

    Would also help if the likes of the EU required realistic support life's for the software on devices like TVs.

    https://mstdn.social/@goldstein/112305383553682995

    Sevoris,

    @quixoticgeek Right to repair and cryptographic overbearing schemes is going to be an interesting (read explosive, and Big Encryption Chain can get shafted) interaction.

    Intel deprecating the whole secure enclave tech broke an already exclusive ecosystem to decrypt blue ray media (and 4K streams?) on newer machines.

    baldur, to random
    @baldur@toot.cafe avatar

    For today’s , a few pictures from late 2022 when Kolka had just recently arrived at my sister’s and was still in the process of getting acclimated to people and investigating her surroundings

    Kolka, paws up on the furniture, discovering the thing people call “sofa”
    Kolka sitting and taking in the living room.
    A picture of the cat as she is just starting to relax around the photographer (my sister).

    Sevoris,

    @baldur What a dapper cat! She looks gorgeous!

    grimalkina, to random
    @grimalkina@mastodon.social avatar

    Many people have advised me to run a newsletter or something. I do not have the heart to tell them that I don't even know why I have a blog (where I upload approximately a full book chapter length piece of writing twice a year). My brain is simply not made for this Content Churn Era

    Sevoris,

    @grimalkina I‘m just here vibing along to whenever you post good stuff

    This is what microblogging was made for, change my mind.

    (Also you can RSS mastodon clients, AFAIK.)

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • Leos
  • kavyap
  • cisconetworking
  • thenastyranch
  • magazineikmin
  • hgfsjryuu7
  • DreamBathrooms
  • InstantRegret
  • Youngstown
  • slotface
  • PowerRangers
  • Durango
  • everett
  • rosin
  • anitta
  • vwfavf
  • modclub
  • ethstaker
  • khanakhh
  • tacticalgear
  • ngwrru68w68
  • osvaldo12
  • mdbf
  • tester
  • cubers
  • normalnudes
  • GTA5RPClips
  • provamag3
  • All magazines
    •