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 Palaiologos (kspalaiologos@fedi.absturztau.be), page 2

  1. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Sunday, 05-May-2024 19:48:33 JST Palaiologos Palaiologos
    in reply to
    • Palaiologos
    Most people I chatted with on Matrix will not see it, but see you on XMPP, finally!
    In conversation about 7 months ago from fedi.absturztau.be permalink
  2. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Sunday, 05-May-2024 19:48:32 JST Palaiologos Palaiologos
    in reply to
    • Palaiologos
    hahaha WOW Jabber is so bad. I spent like 45 minutes setting it up, and voice calls still don't work, the config syntax is poorly documented - the STUN/TURN tutorial I have been following linked by the docs uses outdated syntax, then after I fixed it it wouldn't bind ports, removing some entries from the default config mysteriously fixed it, I couldn't receive or send media, couldn't start video/voice calls, had no E2EE and kept getting certificate errors. STUN and TURN are soo googleable too. smelly nerds.
    In conversation about 7 months ago from fedi.absturztau.be permalink
  3. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Sunday, 05-May-2024 19:48:28 JST Palaiologos Palaiologos
    in reply to
    • Maciej "XGQT" Barć
    @xgqt yeah, i am staying on discord i think. i hate it and it sucks, but there's nothing better.

    - centralised corpo services: your data gets sold, you have no rights
    - decentralised services: whoever hosts your account will spy on you
    - selfhosted decentralised services: pain in the ass to maintain and host, need to be a smelly nerd to do that
    In conversation about 7 months ago from fedi.absturztau.be permalink
  4. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Friday, 12-Apr-2024 07:55:38 JST Palaiologos Palaiologos
    today's discovery: microsoft "plagiarised" haruhiko okumura's binary tree lzss implementation for DOS6.0 installer compression. it even uses the same sliding and search window size.
    In conversation about 7 months ago from fedi.absturztau.be permalink
  5. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Sunday, 03-Mar-2024 00:17:51 JST Palaiologos Palaiologos
    The reason why we patch most buffer overflow vulnerabilities is not because they're a potential RCE. You can't reliably exploit most of these bugs to get a RCE.Tthe real reason why they're fixed is that they're provide a surface for a DoS attack. There's a negligible difference between a heap buffer overflow leading into a segfault and a panic!("Out of bounds.").
    In conversation about 9 months ago from fedi.absturztau.be permalink
  6. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Saturday, 18-Nov-2023 07:42:44 JST Palaiologos Palaiologos
    Software engineering and academia make me want to quit and become a carrot farmer. Never in my life have I felt so helpless and trapped in a surreal and warped reality.
    In conversation about a year ago from fedi.absturztau.be permalink
  7. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Sunday, 30-Jul-2023 18:38:31 JST Palaiologos Palaiologos
    • Haelwenn /элвэн/ :triskell:
    • Ignas Kiela
    • niconiconi
    • Janne Moren
    @niconiconi @lanodan @ignaloidas @jannem this is largely a self-fulfilling prophecy: IBM PC with x86 and not m68k happened almost by _pure chance_. because people started using x86 we started getting more and more software to run on it, more manufacturers (Cyrix, AMD), then finally x86_64. at no point in its life x86 was "technologically superior", rather, it always has been lucky for history and everything that came after was a result of this.
    In conversation Sunday, 30-Jul-2023 18:38:31 JST from fedi.absturztau.be permalink
  8. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Sunday, 30-Jul-2023 18:37:37 JST Palaiologos Palaiologos
    • Haelwenn /элвэн/ :triskell:
    • Ignas Kiela
    • niconiconi
    • Janne Moren
    • Palaiologos
    @niconiconi @lanodan @ignaloidas @kspalaiologos @jannem have i just started a flame war :-)? obviously they are better when it comes to TDP than x86 machines, but that's because neither Intel nor AMD didn't care too much about low-TDP CPUs before. they are significantly cheaper than full x86 builds (esp. M1 Macs Mini), the CPU hardware is _just_ better (bigger OoO buffers, built with the 5nm process, less stupid legacy crust and bad design decisions like AVX512 taking die space, simpler instruction decoders, ...).

    I understand the sentiment of disliking Apple - I purposefully don't support neither Intel or M1 Macs because of Apple's predatory policies, I think that their software sucks and further I would rather be trying to fit my cranium inside of my arse than try to port software to the weird proprietary BSD thingamabob they have brought upon us. But their hardware is just better. And it's the only Apple product worth the price.
    In conversation Sunday, 30-Jul-2023 18:37:37 JST from akkoma.0x68756773.moe permalink
  9. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Sunday, 30-Jul-2023 17:40:26 JST Palaiologos Palaiologos
    in reply to
    • Haelwenn /элвэн/ :triskell:
    • Ignas Kiela
    • niconiconi
    @ignaloidas @lanodan @niconiconi well, we had Symbolics machines with hardware stack lifetime escape detection, hardware gc barriers, etc... - imo computing has largely regressed since then.
    In conversation Sunday, 30-Jul-2023 17:40:26 JST from fedi.absturztau.be permalink
  10. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Sunday, 30-Jul-2023 17:40:23 JST Palaiologos Palaiologos
    in reply to
    • Haelwenn /элвэн/ :triskell:
    • Ignas Kiela
    • niconiconi
    • Janne Moren
    @jannem @niconiconi @ignaloidas @lanodan I don't think that people stopped using Symbolics machines because they had superior hardware. The reason it failed is very simple - it is effectively impossible to get rid of x86-oids on the market, even if you have better technology on paper (see M1 Macs) the product is probably not going to see much mainstream use. And you need to be one of the 5-6 richest tech companies to even attempt at doing something like this.
    In conversation Sunday, 30-Jul-2023 17:40:23 JST from fedi.absturztau.be permalink
  11. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Saturday, 29-Jul-2023 19:34:32 JST Palaiologos Palaiologos
    • Haelwenn /элвэн/ :triskell:
    • Ignas Kiela
    • niconiconi
    @niconiconi @lanodan @ignaloidas
    a similar situation: why is the current consensus such that garbage collection barriers should be done in software, while they could easily be done better in hardware instead?
    In conversation Saturday, 29-Jul-2023 19:34:32 JST from fedi.absturztau.be permalink
  12. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Thursday, 20-Jul-2023 17:17:10 JST Palaiologos Palaiologos
    in reply to
    • Puniko ?
    @puniko Yay!
    I wonder, how difficult is maintaining such a huge instance?
    In conversation Thursday, 20-Jul-2023 17:17:10 JST from fedi.absturztau.be permalink
  13. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Monday, 26-Jun-2023 00:41:40 JST Palaiologos Palaiologos
    • Puniko ?
    Hey @puniko - is it possible to export all posts and replies into something like CSV?
    In conversation Monday, 26-Jun-2023 00:41:40 JST from fedi.absturztau.be permalink
  14. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Thursday, 22-Jun-2023 22:54:56 JST Palaiologos Palaiologos
    I have 3 job offers piled up and they are all so interesting... I literally have an interview in 9 minutes!! The first is stuff that feels interesting to me, on-site, pretty ok salary; the second is stuff that is considerably less interesting but it's remote and pays better
    and the third is being a TA. I am certainly going to do the first over the summer break, and then after the summer break i will probably both TA and work the second job concurrently to uni.
    In conversation Thursday, 22-Jun-2023 22:54:56 JST from fedi.absturztau.be permalink
  15. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Friday, 16-Jun-2023 22:09:19 JST Palaiologos Palaiologos
    in reply to
    • Palaiologos
    SpiderSol, made on 22-12-2022 (over 6 months ago!), implementing the classic "Windows Spider Solitaire" from Windows XP. The other games come from Windows 3.x (Reversi) and Windows XP (Minesweeper):
    In conversation Friday, 16-Jun-2023 22:09:19 JST from fedi.absturztau.be permalink

    Attachments


    1. https://media.absturztau.be/media/5c7ffda063b1ac8b49adc78fbb6b2c7876dbf6082176d8adac81aceff3424ada.png?name=game.png
  16. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Friday, 16-Jun-2023 22:07:00 JST Palaiologos Palaiologos
    in reply to
    • Palaiologos
    Previously: SDLMine (under 1000 lines of code) made on 11-06-2023 - https://github.com/kspalaiologos/sdlmine.
    In conversation Friday, 16-Jun-2023 22:07:00 JST from fedi.absturztau.be permalink

    Attachments


    1. https://media.absturztau.be/media/19738bcc7f6c4646e8d5d020f41ddbcc7f36ddd3e5fa6b5c13ced48389445627.png?name=screenshot.png
    2. Domain not in remote thumbnail source whitelist: opengraph.githubassets.com
      GitHub - kspalaiologos/sdlmine: A SDL2 minesweeper in 1000 lines of code.
      A SDL2 minesweeper in 1000 lines of code. Contribute to kspalaiologos/sdlmine development by creating an account on GitHub.
  17. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Friday, 16-Jun-2023 22:05:29 JST Palaiologos Palaiologos
    Just finished writing my 3rd game from the "make a new SDL game every day" series - Othello, this time under 500 lines of code: https://github.com/kspalaiologos/sdlreversi
    An example Othello game against…
    In conversation Friday, 16-Jun-2023 22:05:29 JST from fedi.absturztau.be permalink

    Attachments


    1. https://media.absturztau.be/media/f2d5b3db66aa94cbfd35c9306ede78151d3d95913a59cf2ee0f16ef1fb8f8db7.gif?name=game.gif
    2. Domain not in remote thumbnail source whitelist: opengraph.githubassets.com
      GitHub - kspalaiologos/sdlreversi: A reversi/othello game under 500 lines of C++ code.
      A reversi/othello game under 500 lines of C++ code. - GitHub - kspalaiologos/sdlreversi: A reversi/othello game under 500 lines of C++ code.
  18. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Friday, 24-Feb-2023 15:18:28 JST Palaiologos Palaiologos
    Considering to move to my own Fedi instance.
    In conversation Friday, 24-Feb-2023 15:18:28 JST from fedi.absturztau.be permalink
  19. Embed this notice
    Palaiologos (kspalaiologos@fedi.absturztau.be)'s status on Saturday, 26-Nov-2022 03:42:39 JST Palaiologos Palaiologos
    How people manage to write 50 shitpost toots on here daily is beyond my comprehension.
    In conversation Saturday, 26-Nov-2022 03:42:39 JST from fedi.absturztau.be permalink
  • After

User actions

    Palaiologos

    Palaiologos

    Trying out decentralised social networks. Hobbyist programmer, "professional" mathematician.Unaligned leftist, too tired for political discourse.

    Tags
    • (None)

    Following 0

      Followers 0

        Groups 0

          Statistics

          User ID
          39053
          Member since
          25 Nov 2022
          Notices
          39
          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.