GNU social JP
  • FAQ
  • Login
GNU social JPは日本のGNU socialサーバーです。
Usage/ToS/admin/test/Pleroma FE
  • Public

    • Public
    • Network
    • Groups
    • Featured
    • Popular
    • People

Notices by Dawn Ahukanna (dahukanna@mastodon.social)

  1. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Monday, 26-May-2025 00:12:51 JST Dawn Ahukanna Dawn Ahukanna
    in reply to
    • Adrianna Tan
    • Ludwig Vielfrass

    @skinnylatte @lerxst
    Food is like music, composed of many notes (pun intended) into rich tapestry for the senses.
    Notes for flavor & singing
    - http://understandingflavor.com/flavor-notes/
    - https://foodnhotelasia.com/glossary/horeca/tasting-notes/
    > Tasting notes are fundamental aspect of the world of food & beverages. They serve as descriptive language that helps us communicate & understand the complex & nuanced flavors that we encounter. Bridge between sensory experience & words, capture & share the essence of what we _taste, smell, and savor_.

    In conversation about 5 hours ago from gnusocial.jp permalink

    Attachments

    1. Domain not in remote thumbnail source whitelist: understandingflavor.com
      Flavor Notes from Understanding Flavor
      Cultivating Flavor
    2. Domain not in remote thumbnail source whitelist: foodnhotelasia.com
      What are Tasting Notes: A Guide to Flavor Description - Food & Hospitality Asia
      from FHA Editor
      Tasting notes are a fundamental aspect of the world of […]
  2. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Sunday, 23-Feb-2025 06:14:06 JST Dawn Ahukanna Dawn Ahukanna
    in reply to
    • Jason Gorman

    @jasongorman
    It’s incredible that the decision making people responsible for the delays and bottlenecks (prevaricating deciders) only ever focus on efficiency of the means of production (coding, factory production line, etc) that only happens after prevaricated decision making.

    In conversation about 3 months ago from mastodon.social permalink
  3. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Sunday, 23-Feb-2025 06:14:05 JST Dawn Ahukanna Dawn Ahukanna
    in reply to

    Related - https://hachyderm.io/@anderseknert/114049370871795480

    “Why is it that all the success stories of AI increasing ‘developer productivity’ to extreme degrees are always reported from a corporate context? Where are the open source projects that used to have 10 features commited per month, and now have 50? Or even just 20? These claims should be easy as hell to prove given how few industries do so much work in the open as we do, no?”

    In conversation about 3 months ago from mastodon.social permalink

    Attachments

    1. No result found on File_thumbnail lookup.
      Anders Eknert (@anderseknert@hachyderm.io)
      from Anders Eknert
      Why is it that all the success stories of AI increasing developer productivity to extreme degrees are always reported from a corporate context? Where are the open source projects that used to have 10 features commited per month, and now have 50? Or even just 20? These claims should be easy as hell to prove given how few industries do so much work in the open as we do, no?
  4. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Monday, 27-Jan-2025 17:40:33 JST Dawn Ahukanna Dawn Ahukanna

    “ The problem with #AI and ethics is training, the source. A good writer takes training, and that includes willing teachers, reading, experience writing, and feedback. A good AI requires the same thing, they call the feedback "curation." That's "educating" it what is crap and what is gold. ” -
    #EthicsInAI #techConsequencesInRealWorld
    https://eldritch.cafe/@sfwrtr/113771969864497878

    In conversation about 4 months ago from mastodon.social permalink

    Attachments

    1. No result found on File_thumbnail lookup.
      RS, Author, Novelist, Prosaist (@sfwrtr@eldritch.cafe)
      from RS, Author, Novelist, Prosaist
      I turned off #Apple #Intelligence because not only did it write things resembling "My grandfather fell out of one of those things and died when he was a baby," they made it so that if you press the space bar twice (which used to insert a period) now will insert that frequently used phrase or sentence ending instead. If you're typing and not reading what you're typing, chances are you're gonna say some *very strange* things. I could probably go up on a rant on this, but when did people decide that #writing ought to be dumbed down to what everybody else is saying, rather than using your brain to come up with something eloquent to move your audience? Are we really that lazy? Yeah, I see the root causes: Greedy corporations wanting to jump onto the next big thing, Gullible #programmers racing to #code the killer feature and bonuses rather than judging the worthiness of the feature. It's a little bit like radium and watch dials, which I brought back for one of my stories with a twist. Madam Curie died because science didn't understand something essential about radioactivity, *yet,* and who knows how many cases of lung cancer resulted from the glow in the dark watches. The problem with #AI and ethics is training, the source. A good writer takes training, and that includes willing teachers, reading, experience writing, and feedback. A good AI requires the same thing, they call the feedback "curation." That's "educating" it what is crap and what is gold. Moreover, the training ought be done on *the writing of the person using the AI,* or by willing and well paid teachers, and I really don't get why Apple didn't take this tack and create something truly useful to that writer. My best guess why not? The psychology of their perceived user: The need for instant gratification. If we train each AI the way we train each human student, where's the ability to cut out the labor and mass produce writers? I want the personally trained #AI that can fill in the missing word that's at the tip of my tongue, that can reliably flag my recurring grammar errors, that can complete my sentences with the accuracy of my spouse of decades. As an SF writer or as an essayist, it should recognize my idiosyncratic vocabulary and not autocorrect it, to the point of accepting new words or phrases. It astounds me that AI contribution isn't color-coded in a ADA safe way during composition to show where the writer might want to verify the generated contributions. That all auto correct auto complete *doesn't* do this is a serious set back to the technology. Okay, I ranted. A big thank you to @vextaur@blog.taursnd.haus for the discussion I sourced this from. #BoostingIsSharing #writer #writing #author #fiction #SF #sciencefiction #WritingCommunity #writersOfMaston #greed #economics #genai #chatgpt #openai
  5. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Friday, 03-Jan-2025 01:40:55 JST Dawn Ahukanna Dawn Ahukanna
    in reply to
    • Scott Jenson
    • Eaton

    @eaton @scottjenson

    I’m hacking on regular forum software, that solves my conversation focus with reply threading, branching & digressions, to figure out if it could do orchestration work as a POSSE approach: https://indieweb.org/POSSE.
    I post on the modified forum software & it logs in to the target specified server(s), posts, boosts & collects any comments, likes, etc with the relevant sequencing etc.

    Also allows me to take time to create & review posts before publishing, plus scheduling.

    In conversation about 5 months ago from mastodon.social permalink

    Attachments


  6. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Friday, 03-Jan-2025 00:45:57 JST Dawn Ahukanna Dawn Ahukanna
    in reply to
    • Troed Sångberg
    • Scott Jenson
    • Jer Warren

    @troed @scottjenson @nyquildotorg
    Rather than a “Fediverse ID” have an account/profile/ whatever that allows more then 1 ID to be supported, a concept tweetdeck et al introduced. You could manage, schedule and post to 1 or more IDs. You the person decides what the purpose of those IDs serve and take action accordingly.

    In conversation about 5 months ago from mastodon.social permalink
  7. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Wednesday, 01-Jan-2025 23:48:13 JST Dawn Ahukanna Dawn Ahukanna
    in reply to

    3/7
    Now, after the lecture, really understanding & creating your individual, applied reference mental model for the topic is up to you, ignorant student, to figure out on your own. Formal tool used are assignments & homework that provide evidence you can “confidently” repeat (parrot) the material, at will, not that you understood or can apply the material. Basically a Human LLM.
    So I’m not surprised that formal students are resorting in droves to LLM to complete assignments and homework.

    In conversation about 5 months ago from gnusocial.jp permalink
  8. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Wednesday, 01-Jan-2025 23:48:13 JST Dawn Ahukanna Dawn Ahukanna
    in reply to

    2/7
    Current “formal teaching” system:
    1. You attend a lecture where syllabus, maybe lecture material and hardly ever the exam questions are shared.
    2. You’re expected to sit, listen & take notes for 30-45 minutes, maybe answer a few questions & rarely ask any.
    3. Content delivery completed.

    In conversation about 5 months ago from mastodon.social permalink
  9. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Wednesday, 01-Jan-2025 23:48:13 JST Dawn Ahukanna Dawn Ahukanna
    in reply to

    1/7
    Interesting thread on formal teaching, specifically mathematics & students who, by design, are ignorant of the subject/topic. Students are resorting to commercial Large Language Model(LLM) for explanation, self-study & producing content for completing assignments - https://mathstodon.xyz/@jonmsterling/113555911887640230
    (insert post URL here)

    In conversation about 5 months ago from gnusocial.jp permalink

    Attachments

    1. No result found on File_thumbnail lookup.
      Jon Sterling (@jonmsterling@mathstodon.xyz)
      from Jon Sterling
      The era of ChatGPT is kind of horrifying for me as an instructor of mathematics... Not because I am worried students will use it to cheat (I don't care! All the worse for them!), but rather because many students may try to use it to *learn*. For example, imagine that I give a proof in lecture and it is just a bit too breezy for a student (or, similarly, they find such a proof in a textbook). They don't understand it, so they ask ChatGPT to reproduce it for them, and they ask followup questions to the LLM as they go. I experimented with this today, on a basic result in elementary number theory, and the results were disastrous... ChatGPT sent me on five different wild goose-chases with subtle and plausible-sounding intermediate claims that were just false. Every time I responded with "Hmm, but I don't think it is true that [XXX]", the LLM responded with something like "You are right to point out this error, thank you. It is indeed not true that [XXX], but nonetheless the overall proof strategy remains valid, because we can [...further gish-gallop containing subtle and plausible-sounding claims that happen to be false]." I know enough to be able to pinpoint these false claims relatively quickly, but my students will probably not. They'll instead see them as valid steps that they can perform in their own proofs.
  10. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Wednesday, 01-Jan-2025 23:48:13 JST Dawn Ahukanna Dawn Ahukanna

    Root-0:
    2025 is my year of communal, autodidactic learning & education. Been thinking, reading, listening, watching, musing & discussing material on this topic for a while. It’s time for #CarefulCactusHugging

    In conversation about 5 months ago from mastodon.social permalink
  11. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Wednesday, 01-Jan-2025 23:48:12 JST Dawn Ahukanna Dawn Ahukanna
    in reply to

    Related-“More I work in edu & read history, I’m less convinced it's possible to break free from LLMs.
    At least in USA, incentives to get an education are all wrong. Rarely anyone ever concerns themselves with actually learning. Rather they're just trying to get through process as quickly & efficiently as they can to move on to a job.
    Unless we sort this out & make people actually interested in the learning process, we definitely won't be able to reckon with this tech.
    https://hackers.town/@mav/113564900420941301

    In conversation about 5 months ago from mastodon.social permalink

    Attachments

    1. No result found on File_thumbnail lookup.
      mav :happy_blob: (@mav@hackers.town)
      from mav :happy_blob:
      @jonmsterling@mathstodon.xyz The more I work in edu, and the more I read history, the less I am convinced it's possible to break free from LLMs. At least here in the US, the incentives for getting an education are all wrong. Rarely does anyone ever concern themselves with actually learning, but rather they're just trying to get through the process as quickly and efficiently as they can so they can move on to a job. Unless we sort this out, and make people actually interested in the learning process, we definitely won't be able to reckon with this tech.
  12. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Tuesday, 31-Dec-2024 20:47:35 JST Dawn Ahukanna Dawn Ahukanna
    in reply to
    • myrmepropagandist

    @futurebird

    … extract links from within the post and links to the source post?

    In conversation about 5 months ago from mastodon.social permalink
  13. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Tuesday, 31-Dec-2024 20:46:17 JST Dawn Ahukanna Dawn Ahukanna
    in reply to
    • myrmepropagandist

    @futurebird

    To remove & externalise bookmark dependency from browsers, I’ve resorted to manually collecting & curating links as I find them, with personal notes+tags reminding me why they are of interest. They’re always 100% searchable & findable.

    Given the inconsiderate, effective DDOS behavior of AI scraper bots, adding to that melee with more robo-indexing may not produce a usable search index - https://mastodon.social/@dahukanna/113741237599333856

    In conversation about 5 months ago from mastodon.social permalink

    Attachments

    1. No result found on File_thumbnail lookup.
      Dawn Ahukanna (@dahukanna@mastodon.social)
      from Dawn Ahukanna
      @recursive@hachyderm.io how do they, the producers and indoctrinators of “Artificial Intelligence Large Language Models (AI-LLM)”, not “compute” that: content production rate ≠ content request rate? I’d like to have and host a website without it or the environment being “knackered” from constant demands.
  14. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Tuesday, 31-Dec-2024 05:46:50 JST Dawn Ahukanna Dawn Ahukanna
    in reply to
    • Dr. Cat Hicks
    • Bill Thompson

    @billt @grimalkina
    Code is story time with clues frozen in inaccessible “amber” (people’s situation, thinking and practices of an era gone by). The computer instructions are secondary side effects. 😉

    In conversation about 5 months ago from mastodon.social permalink
  15. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Friday, 27-Dec-2024 18:55:22 JST Dawn Ahukanna Dawn Ahukanna
    • Valerie Aurora

    @vaurora

    Human Performance in academic setting is more than making money for another organisation based on charging fees for publishing free content, <cough> research findings. With all that brain power they couldn’t come up with a system? That is the antonym of “thinking”.

    Note: Businesses may want to replace humans as they are usually the largest expense on the balance sheet, but they don’t outsource human performance reviews. It’s a core function of a manager.

    In conversation about 5 months ago from mastodon.social permalink
  16. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Friday, 27-Dec-2024 18:27:17 JST Dawn Ahukanna Dawn Ahukanna

    How did we end up with business middlemen gatekeeping academic careers via paid for academic journals that get free contributions from the same academics?

    Alt text: Pooh bear, sitting with a pensive expression on his face, right arm folded under his left and his left paw tapping his temple, saying, “Think, think, think, think, think!”.

    https://archaeo.social/@mrundkvist/113723788751363475

    In conversation about 5 months ago from mastodon.social permalink

    Attachments


  17. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Sunday, 22-Dec-2024 22:39:31 JST Dawn Ahukanna Dawn Ahukanna
    in reply to
    • myrmepropagandist
    • Vayl Larkin (they/them)

    @futurebird @VaylLarkinPoet
    Even “people of various non-western cultures” with tertiary-level formal education want to spend “precious & limited” time in places & spaces where they’re welcome. Not treated to the ever common & very politely stated, “WTF are you doing here? Go where you belong”. Their current job title, education level, where they live, what they are capable of, etc. (that isn’t stated by the way, but assumed) is secondary.
    People are more than “how they earn money to pay bills”.

    In conversation about 5 months ago from mastodon.social permalink
  18. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Sunday, 22-Dec-2024 02:23:45 JST Dawn Ahukanna Dawn Ahukanna
    in reply to
    • Mastodon Migration
    • LAUREN :bacall_fire:
    • Chancerubbage

    @Chancerubbage @noondlyt @mastodonmigration

    Sampling bias
    Alt: SAMPLiNG BiAS
    Screen showing results of survey with 2 questions, without ”None of the above” or “Other” option,.
    Q1. “Yes, I love responding to surveys” = 99.8% (no sample size number indicated)
    Q2. “No, I toss them in the bin” - 0.2%

    Presenter states: “We received 500 responses & found that people love responding to surveys” - without counting the people who don’t 😱
    Audience:”HMM...”
    From sketchplanations

    In conversation about 5 months ago from mastodon.social permalink

    Attachments


    1. https://files.mastodon.social/media_attachments/files/113/690/204/098/303/987/original/b3a2105c8572f029.jpeg
  19. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Thursday, 19-Dec-2024 14:10:20 JST Dawn Ahukanna Dawn Ahukanna

    Watched “Downfall: The case against Boeing” https://www.netflix.com/title/81272421 & it’s frightening cos of avarice-induced:
    1. “gambling & betting” on Stock Market
    2. don’t bring me problems, “do it faster”-with no training
    3. “blame” pilots for no “skills” to fly plane.
    4. cavalier “we’ll have software ‘fix’ in 6 weeks” produced without feedback from pilots.
    It takes years to design & test a plane. It’s not like patching IT equipment in “immobile, closed system data center”.
    #techConsequencesInRealWorld

    In conversation about 5 months ago from mastodon.social permalink

    Attachments

    1. Domain not in remote thumbnail source whitelist: occ-0-993-988.1.nflxso.net
      Watch Downfall: The Case Against Boeing | Netflix Official Site
      Investigators reveal how Boeing’s alleged priority of profit over safety could have contributed to two catastrophic crashes within months of each other.
  20. Embed this notice
    Dawn Ahukanna (dahukanna@mastodon.social)'s status on Sunday, 15-Dec-2024 17:46:22 JST Dawn Ahukanna Dawn Ahukanna

    Perverse version of “you control what you measure” - https://mstdn.ca/@CAWguy/113654123593956447

    In conversation about 5 months ago from mastodon.social permalink

    Attachments

    1. Domain not in remote thumbnail source whitelist: cdn.mastdn.ca
      J.L.1285 (@CAWguy@mstdn.ca)
      from J.L.1285
      Attached: 1 image It’s a long article on these insanely bright car headlights, but it contains this paragraph ⬇️ https://www.theringer.com/2024/12/03/tech/headlight-brightness-cars-accidents?lid=dpe6tiyd81ev
  • Before

User actions

    Dawn Ahukanna

    Dawn Ahukanna

    Software Alchemist - Turning base code into precious applications. Devsigner == 'Dev'eloper + De'signer'. Married to @Anni ❤️ 🏳️🌈. Pronouns == she/her.

    Tags
    • (None)

    Following 0

      Followers 0

        Groups 0

          Statistics

          User ID
          101531
          Member since
          25 Feb 2023
          Notices
          103
          Daily average
          0

          Feeds

          • Atom
          • Help
          • About
          • FAQ
          • TOS
          • Privacy
          • Source
          • Version
          • Contact

          GNU social JP is a social network, courtesy of GNU social JP管理人. It runs on GNU social, version 2.0.2-dev, available under the GNU Affero General Public License.

          Creative Commons Attribution 3.0 All GNU social JP content and data are available under the Creative Commons Attribution 3.0 license.