Mastodon Feed
Boosted by jsonstein@masto.deoan.org ("Jeff Sonstein"):
overholt@glammr.us ("John Overholt") wrote:
RE: https://mastodon.social/@arstechnica/117321581979243460
JUST BE COLD THATS ALL I EVER WANTED FROM YOU
Mastodon FeedMastodon FeedBoosted by jsonstein@masto.deoan.org ("Jeff Sonstein"):
CStamp ("Carolyn") wrote:
8 suicide attempts.
It can’t help that it’s not a justifiable war & there is so little respect for the US these days.
And those jokers show no respect.
https://www.newsweek.com/trump-accused-disrespect-navy-admits-uss-lincoln-suicide-attempts-12480936
Mastodon FeedBoosted by jsonstein@masto.deoan.org ("Jeff Sonstein"):
jeffjarvis ("Jeff Jarvis") wrote:
G20 should boycott and turn it into the G2, the fascist bromance.
Trump Invites Putin to Attend Group of 20 Summit in Miami https://www.nytimes.com/2026/09/23/us/politics/trump-putin-g20-summit.html?smid=tw-share
Mastodon FeedBoosted by chris@fedi.freebooters.uk ("Chris Were ⁂🐧🌱☕️"):
hamishtpb@mewblog.thepolarbear.co.uk ("Hamish The PolarBear") wrote:
Which vessel would you want to be the captain of?
Mastodon Feedchris@fedi.freebooters.uk ("Chris Were ⁂🐧🌱☕️") wrote:
My latest TV and movie reviews.
Mastodon FeedBoosted by cstanhope@social.coop ("Your weary 'net denizen"):
wim_v12e@scholar.social ("Wim🧮") wrote:
Have you wondered how bad CO₂ emissions, and therefore global heating, could really get if the AI industry gets its way?
I modelled the emissions based on their projections:
"The growth the AI industry wants is disastrous for climate change"
https://limited.systems/articles/ai-datacentre-emissions/
(this is based on my peer-reviewed paper published in the proceedings of 2nd International Workshop on Low Carbon Computing, linked in the article)
Mastodon FeedtimnitGebru@dair-community.social ("Timnit Gebru (she/her).") wrote:
I'm gonna do a recap of this week when I get a breather because the hype DDOS strikes daily 🙄 As for the whole CRISPR Dario "discovery" announcement:
See below
Mastodon Feeddysfun@treehouse.systems ("gaytabase") wrote:
of course not listening to your users while pretending to be a community project and then mischaracterising the behaviour of people who choose to involve themselves in your community to strengthen your point might be considered a massive dick move.
Mastodon FeedBoosted by krig@goto.liten.app:
kano@social.sunet.se ("Micke Nordin :sunet:") wrote:
I morgon ska jag träffa kollegorna på Försäkringskassan, Skatteverket, Trafikverket och Digg, bland annat, för att prata om hur man kan självhosta Forgejo. På fredag bär det sedan av till Finansinspektionen för att berätta hur man kan självhosta Kubernetes. Digitalt självbestämmande är äntligen på agendan verkar det som.
Mastodon FeedEmilyEnough@hachyderm.io ("Emily 🏳️🌈🏳️⚧️") wrote:
Transitioning in your 40s is like Dr. Who. Sure there are occasional references to previous iterations, sometimes an old companion drops by, and some of the rules don’t make sense because they were established in the 80s, but you’re really just focused on the current plot.
…. oh, and obsessed “superfans” care a hell of a lot more of about canon than you do.
Mastodon Feeddysfun@treehouse.systems ("gaytabase") wrote:
of course it's not mandatory that software considers the opinion of its users, but if it doesn't, it might find it no longer has any.
Mastodon FeedBoosted by soatok@furry.engineer ("Soatok Dreamseeker"):
afranke@mamot.fr ("Alexandre Franke") wrote:
> If you’re not following this sector, you may not know that we call something different AI every five to ten years and have done since the mid 1950s. The thing we’re calling AI this year…
— @pluralistic in “This is how the AI bubble pops”
Mastodon Feeddysfun@treehouse.systems ("gaytabase") wrote:
i've got the beatles' "blackbird" stuck in my head except it's not "blackbird" now, it's "batgirls"
baaaaaat giiiiiiiiirls flyyyyyyy right into the night...
send help
Mastodon FeedBoosted by dysfun@treehouse.systems ("gaytabase"):
danlew@androiddev.social ("Dan Lew") wrote:
Regarding Samsung bricking their smart fridges...
I once told a contractor that I explicitly did NOT want a smart appliance. His response was “you must be a software developer, they’re usually the ones who don’t want them.”
Mastodon FeedBoosted by baldur@toot.cafe ("Baldur Bjarnason"):
davidgerard@circumstances.run ("David Gerard") wrote:
i know people use chatbots for search because search has been made deliberately bad, but i can't see how they use it. the AI Overview is a reliable source of wrongness, and remarkable claims that are absolutely not in the sources.
does anyone have informed ideas on what they get out of this wrong answer machine?
(this is not a call for evangelism on how to search, STFU. it's asking about thought processes in the people who do use it)
Mastodon Feedandrewnez ("Andrew Nesbitt") wrote:
Package Manager Sandboxing
https://nesbitt.io/2026/09/24/package-manager-sandboxing.html
Mastodon FeedBoosted by dysfun@treehouse.systems ("gaytabase"):
endrift@treehouse.systems ("endrift 🏳️⚧️") wrote:
@pinchartl This post has a quick summary, and it's really just scratching the surface. I feel like these tools are waterboarding me, and I don't get to opt out. https://social.treehouse.systems/@endrift/117154802758493056
Mastodon FeedBoosted by dysfun@treehouse.systems ("gaytabase"):
pinchartl@floss.social ("Laurent Pinchart") wrote:
Over the past few months, I have found myself under the spotlight on kernel mailing lists. What started as targeted discussions quickly turned into advocating for the kernel community to respect the mostly silent crowd of developers and maintainers who do not want to fully embrace AI usage. It appears this didn't go unnoticed, as shown by Jon Corbet's talk at Kernel Recipes. (1/5)
Mastodon Feedjonny@neuromatch.social ("jonny (nonvenomous)") wrote:
THERE IS NO FUCKING WAY IT JUST RUNS AS ROOT. That can't be right. It just can't. It must be a VM within a VM.
Mastodon Feedjonny@neuromatch.social ("jonny (nonvenomous)") wrote:
in case i left it in suspense, i will post the tarball once i can spend some daylight hours cleaning it of PII
Mastodon FeedBoosted by jonny@neuromatch.social ("jonny (nonvenomous)"):
elseweather wrote:
@jonny Fuck yes. We're disrupting Big Personality Quiz. Dump another $1T into this industry immediately
Mastodon Feedpzmyers@freethought.online ("pzmyers 🕷") wrote:
I've got an easy choice for Senate.
https://freethoughtblogs.com/pharyngula/2026/09/24/how-not-to-campaign-in-minnesota/
Mastodon Feedjonny@neuromatch.social ("jonny (nonvenomous)") wrote:
that is 100% claude authored by the way, unmistakable.
Mastodon Feedjonny@neuromatch.social ("jonny (nonvenomous)") wrote:
ah yes, here we are. the largest companies in the world releasing a flagship product they are betting their entire competitiveness on that is built on bash scripts that have hardcoded string interpolation python scripts and in fact hardcoded string interpolated bash scripts that seems to be the machinery that manages runaway agent spawning and concurrency.
[![# A function so callers can RE-snapshot after in-pass ledger writers (the # cutover conversion appends its converted entry mid-script; replay must # fold that entry in the same pass, not next boot). snapshot_os_intent_ledger() { rm -f "$ledger_snapshot" "$ledger_snapshot.sha256" "$ledger_snapshot.skipped-bytes" # Existence pre-check is NOISE control only (quiet fresh-VM boots); the # python open owns enforcement, so a race here changes nothing. -e follows # symlinks, so a DANGLING guest-planted symlink must also pass the gate via # -L and reach the python open — that is what turns the plant into the loud # tamper message instead of a silent healthy-looking skip. if [ -e "$ledger_source" ] || [ -L "$ledger_source" ]; then snap_rc=0 python3 - "$ledger_source" "$ledger_snapshot" <<'PYSNAP' 2>/dev/null || snap_rc=$? import hashlib, os, stat, sys src, dst = sys.argv1, sys.argv2 try: fd = os.open(src, os.O_RDONLY | os.O_NOFOLLOW | os.O_NONBLOCK) except OSError: sys.exit(3) try: if not stat.S_ISREG(os.fstat(fd).st_mode): sys.exit(4) hasher = hashlib.sha256() retain = 16 * 1024 * 1024 ceiling = 64 * 1024 * 1024 hashed = 0 skipped = 0 out = os.open(dst, os.O_WRONLY | os.O_CREAT | os.O_EXCL, 0o600) try: while True: chunk = os.read(fd, 1 << 20) if not chunk: break hashed += len(chunk) if hashed > ceiling: sys.exit(5) hasher.update(chunk) take = chunk[:retain] if retain > 0 else b"" if take: os.write(out, take) retain -= len(take) skipped += len(chunk) - len(take) finally: os.close(out) if skipped: with open(dst + ".skipped-bytes", "w") as sidecar: sidecar.write("%d\n" % skipped) with open(dst + ".sha256", "w") as sidecar: sidecar.write(hasher.hexdigest() + "\n") finally: os.close(fd) PYSNAP](https://files.mastodon.social/cache/media_attachments/files/117/325/680/632/773/289/original/dd651c0ecb40c9d6.png)
]3
![# Reconcile guest packages via hatch-manifest, run INSIDE the live cell. # # hatch-manifest maybe-apply compares the manifest fingerprint against a state # file: identical -> exits ~1ms; changed -> applies the package delta via apt # (through the cell's Sentinel egress). # # hatch-manifest and runtime-cell.kdl are image-vendored under # /opt/hatch-image/bin (the cell binds it read-only; see htch-runtime.nspawn). # Older bundle-era layouts shipped them under /opt/hatch/bin, kept here as a # fallback. Resolve each from the new home first. # # Prefer the /opt/hatch-image/manifests dir layout when present (bind it into the # cell once that diff lands); otherwise fall back to the single-file # runtime-cell.kdl resolved from /opt/hatch-image/bin (or legacy /opt/hatch/bin). # The manifests branch is kept so the cutover to the dir layout is a no-op here. in_cell_reconcile=' set -eu # Memory containment (see the maintenance-leaf note above: cgroup bounds # cannot reach this subtree). RLIMIT_AS, inherited by every child this # shell spawns — hatch-manifest, apt, dpkg, their methods. 3GiB of address # space never troubles a legitimate reconcile (normal apt peaks a few # hundred MB) but kills the degenerate multi-GB runs (observed 6.5G RSS, # SEV S698078) with a clean allocation failure — a nonzero exit and a # retry, never a guest-wide OOM hunt. Fail-open where unsupported. ulimit -v 3145728 2>/dev/null || true # Heal an interrupted dpkg BEFORE the reconcile, not only before the replay: # the reconcile runs first in every pass, so without this arm a pass that # died mid-dpkg would fail its NEXT manifest apply too (one wasted pass and # one wasted strike) before the replay-side heal repaired the db. Fast no-op # on a clean db; a failure here just lets the apply fail loudly as before. nice -n 10 ionice -c3 env DEBIAN_FRONTEND=noninteractive dpkg --configure -a >/dev/null 2>&1 || true manifest_bin= for cand in /opt/hatch-image/bin/hatch-manifest /opt/hatch/bin/hatch-manifest; do if [ -x "$cand" ]; then manifest_bin="$cand" break fi done if [ -z "$manifest_bin" ]; then echo "preflight-opportunistic: hatch-manifest missing in cell" >&2 exit 1 fi # Deliberately NO lock-file removal here: apt/dpkg fcntl locks die with # their process, so a stale lock cannot persist — while unlinking a LIVE # lock detaches it from its path and lets a second dpkg start against the # same database (status-db corruption). A lock-blocked run below just # fails this opportunistic pass and retries on the next cell restart. manifest_kdl= for cand in /opt/hatch-image/bin/runtime-cell.kdl /opt/hatch/bin/runtime-cell.kdl; do if [ -f "$cand" ]; then manifest_kdl="$cand" break fi done](https://files.mastodon.social/cache/media%5Fattachments/files/117/325/680/719/600/627/original/6741709f4df8bab7.png)
Mastodon Feedjonny@neuromatch.social ("jonny (nonvenomous)") wrote:
oh for heaven's sake the entire thing is quite literally cron tasks and bash scripts
[![[file tree under /opt/hatch. there entire "runtime" is just a bunch of bash scripts" opt hatch assets. » I blockiist » I ideas » 8 bin runtime-cell hosts resolv.conf wgetrc bin-scopes.conf /+ build-cell-trust-store.sh /+ control-daemon.sh /+ control-execd.sh /+ ensure-rootfs.sh /+ guest-runtime-env.sh [3 guestenv [9 hatch-preflight-opportunistic /+ is-loopback-rv.sh /+ launch-daemon.sh /+ post-start.sh /+ post-stop.sh /+ pre-start.sh /+ require-modules.sh /+ resolve-rootfs-path.sh /% run-daemon.sh /+ runtime-cell-entry.sh 3 skill-scopes.conf /+ stop.sh]3
Mastodon Feeddysfun@treehouse.systems ("gaytabase") wrote:
one particularly sweet point in the design space is used by java depending on how you configure it - you can have up to 32gb heaps with 32-bit pointers, because pointers are always aligned to 8 bytes so there are 3 bits to steal (or 8x the address space)
Mastodon Feedjonny@neuromatch.social ("jonny (nonvenomous)") wrote:
weird,
/opt/hatch-imageseems to have openai whisper, qdrant minilm, and jina ai reranker models in it right next to a bun and codex binary... did meta do anything at all?
Mastodon Feedbaldur@toot.cafe ("Baldur Bjarnason") wrote:
A few more observations:
- Short videos really are a separate medium (as was my conclusion after the second video) and posting excerpts as shorts doesn't have any traction
- Videos and podcasts do really seem to be separate audiences and I have this unsubstantiated sense that the podcast audience is likely to be more patient and reliable than the video one
- Nothing beats the existing newsletter audience, so continuing to do right by them is vital.Anyway, all of this is genuinely kind of fun.
Mastodon Feedpzmyers@freethought.online ("pzmyers 🕷") wrote:
Don't fear the wasp spider, Englanders!
Mastodon Feedbaldur@toot.cafe ("Baldur Bjarnason") wrote:
So, as useful as "going meta" can be, I think it makes sense to shift topics a bit after the fourth video
There's still a place for "look at the bear trap we're in" videos, especially if they come with useful and actionable ideas
But I think there might also be something to videos themed along the lines of "follow me on my journey to make things" where people might get ideas for approaches to apply themselves.
Maybe alternating the two? Not decided yet.