's Avatar

@geofflangdale.bsky.social

228 Followers  |  35 Following  |  363 Posts  |  Joined: 15.11.2024  |  2.4471

Latest posts by geofflangdale.bsky.social on Bluesky

Very trad. Next I expect someone to be doing that whole "sword and scabbard" thing we used to make from fallen Norfolk Island Pine foilage, or making one of those little paper fortune-tellers/chatterboxes. Maybe they could figure out the future of the Coalition on it!

04.11.2025 11:29 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

They have few enough sitting days as it is; are they taking the piss, sitting there playing on their phones? Particularly galling since they'll pontificate at length about teenagers in schools on phones.

04.11.2025 08:39 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

I've been loving the anecdotes about Hemingway as very, very, mediocre boxer (who also sparred too hard). Jack Dempsey refused to fight him since he figured Hemingway would come out way too hard and he'd have to hurt him for real.

02.11.2025 00:49 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

I'm tickled by the idea that death taxes are somehow worse in terms of reducing incentive to work than, erm, taxing people's earnings from work.

31.10.2025 00:12 β€” πŸ‘ 3    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

I had a couple Asian friends who metabolized alcohol that poorly (they'd go red from one drink, and practically asleep from 3). It didn't seem like a nice cheap scale-down of the Potato Europe liver since there wasn't an equivalent 'pleasantly buzzed' phase for them.

30.10.2025 02:44 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
"Nirvana" shirt

"Nirvana" shirt

I hope you got her this one

24.10.2025 01:41 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Good piece. These "centrists" don't notice how little respect the right wing has for their ludicrous notion of compromise and will happily move the goalposts. In 2007, Labor tries to compromise with Turnbull, but "compromise" on climate in 2025 means trying to compromise with "No Net Zero" Canavan.

23.10.2025 03:07 β€” πŸ‘ 3    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

I first read about hGH in Fred Hatfield's "Power" back in 1990 (which was the very first book I'd seen on any kind of lifting; insane beginner guidance book). He had chapters on various substances and his chapter "hGH: Breakfast of the Sasquatch" began:

"Sasquatch. Big feet. Lordy me, what next?"

18.10.2025 01:26 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

I'm still laughing at the "You shot my brother in the name of justice" sequence. Insanely good stuff.

17.10.2025 02:44 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

The *do* need a vote from an LNP senator to pass the kind of laws that they want to pass, though.

16.10.2025 21:25 β€” πŸ‘ 3    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

I think of this quote when I think about the idea of AGI meaning "human and artificial intelligence have reached functional parity". Many effects of AI will help achieve this; unfortunately, most function by lowering effective human intelligence (e.g. hollowing out education).

15.10.2025 22:25 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
Sam Altman:
We made ChatGPT pretty restrictive to make sure we were being careful
with mental health issues. We realize this made it less useful/enjoyable to
many users who had no mental health problems, but given the seriousness
of the issue we wanted to get this right.
Now that we have been able to mitigate the serious mental health issues
and have new tools, we are going to be able to safely relax the restrictions
in most cases.
In a few weeks, we plan to put out a new version of ChatGPT that allows
people to have a personality that behaves more like what people liked
about 40 (we hope it will be better!). If you want your ChatGPT to respond
in a very human-like way, or use a ton of emoji, or act like a friend, ChatGPT
should do it (but only if you want it, not because we are usage-maxxing).
In December, as we roll out age-gating more fully and as part of our β€œtreat
adult users like adults” principle, we will allow even more, like erotica for
verified adults.

Sam Altman: We made ChatGPT pretty restrictive to make sure we were being careful with mental health issues. We realize this made it less useful/enjoyable to many users who had no mental health problems, but given the seriousness of the issue we wanted to get this right. Now that we have been able to mitigate the serious mental health issues and have new tools, we are going to be able to safely relax the restrictions in most cases. In a few weeks, we plan to put out a new version of ChatGPT that allows people to have a personality that behaves more like what people liked about 40 (we hope it will be better!). If you want your ChatGPT to respond in a very human-like way, or use a ton of emoji, or act like a friend, ChatGPT should do it (but only if you want it, not because we are usage-maxxing). In December, as we roll out age-gating more fully and as part of our β€œtreat adult users like adults” principle, we will allow even more, like erotica for verified adults.

β€œThe junk merchant doesn't sell his product to the consumer, he sells the consumer to his product. He does not improve and simplify his merchandise. He degrades and simplifies the client.”

― William S. Burroughs, Naked Lunch: The Restored Text

15.10.2025 22:25 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

seem determined to move further and further to the center-right (and not even a principled center-right, but sort of an intellectually slovenly mix of authoritarianism, state intervention and corporatism).

11.10.2025 08:58 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

I agree - he is the Thatcher/Reagan for Australia (delayed, since Labor's 1980s run was comparatively balanced given the neoliberalism sweeping the world at that point). Labor's complicity is somehow worse - they've been in power for a full term plus and are in with a huge majority yet ...

11.10.2025 08:57 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

I think this is a matter of where you start counting. If we stick to 20 years, it's pretty even. Of the 8 elections starting in 2004, Labor and the Coalition have each won 4. As the climate crisis worsens, it's notable that Labor has been in power since 2022 and done absolutely nothing for Scope 3.

10.10.2025 20:14 β€” πŸ‘ 3    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

Strangely selective list given that Labor has been just as keen to subsidize fossil fuels as well. After watching them wave through *every single* project in their first term I've come to the conclusion that there's no practical difference, particularly on scope 3 emissions.

10.10.2025 07:52 β€” πŸ‘ 5    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

Using AI for education is like thinking you're going to make champion powerlifters by introducing a forklift to the gym.

We *know* you can parrot basic stuff with an AI in writing/coding - that's cause the training corpora are full of samples. There's only so many 1st year assignments that work.

09.10.2025 05:15 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

I guess that's why ICE doesn't do anything in gang neighborhoods; they're afraid of all the non-violence and humor.

08.10.2025 02:34 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Back when we were a struggling startup (Sensory Networks, acquired by Intel) we had a server room that was about 60% loaner/ES boards from a big range of different companies. They'd often get us to run a few benchmarks and then never ask for it back.

07.10.2025 23:44 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Sitting it out until it stabilizes, but watching with interest. It's hard to take seriously the proliferation of different "profiles" (a recipe for bad software imo).

01.10.2025 21:58 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Yes! And given how rapidly AI coding companies are losing money on their product, you'd think that *any* model that could make money would be appealing, even as a stop-gap.

30.09.2025 23:53 β€” πŸ‘ 4    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image

Once again, this perfect @zeboydgames.bsky.social post comes to mind:

30.09.2025 21:12 β€” πŸ‘ 2721    πŸ” 843    πŸ’¬ 13    πŸ“Œ 12

Sort of the a "West Wingization" of Dem politics. Glib neo-liberals who regard it all as a game. They've taken over our nominally more left-wing party here in Australia and something very similar has happened in Britain (although at least our less-right-wing parties win elections!)

28.09.2025 23:23 β€” πŸ‘ 5    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Pluralistic: The real (economic) AI apocalypse is nigh (27 Sep 2025) – Pluralistic: Daily links from Cory Doctorow

Doctorow on the impending economic collapse when the AI bubble bursts. pluralistic.net/2025/09/27/e...

28.09.2025 17:59 β€” πŸ‘ 618    πŸ” 263    πŸ’¬ 4    πŸ“Œ 71

situation where "too much sophistication" (chasing every last optimization, building incredibly potent middleware layers that sprouted their own interpreted language, etc.) created a codebase that was difficult to understand and maintain.

28.09.2025 00:19 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

This seems like a simple fallacy. I'm saying if you jump off a cliff you will die. This does not mean that if you refrain from jumping off a cliff you will live forever.

It's possible to build a bad codebase with any level of sophistication. However, I'm specifically referring to a ...

28.09.2025 00:18 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

It's a vexed problem with Hyperscan. The circumstances under which it was built tended to encourage the "win at all costs" approach (i.e. 'let's go write another few thousand lines to get 5% more performance').

You tend to pay the cost much later - e.g. when there is "turnover" of staff.

26.09.2025 22:48 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

The danger is that persistently operating at a very high "skill level" in terms of sophistication/complexity, even if successful at the time, leaves you with a codebase that will be hard to operate on by anyone else - a "haunted graveyard".

26.09.2025 20:26 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

I still occasionally wear a (sadly very faded) version of your Internet Federation of Lifting Judges t-shirt. Goes to show that what they said was true: "in the future, everyone will be famous to 15 people".

25.09.2025 07:49 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
quote from Richard Russo's Straight Man: " β€œI know it runs deeper than that. You think we defend incompetence and promote mediocrity.” ...  β€œI wish you would promote mediocrity, Mediocrity is a reasonable goal for our institution.”

quote from Richard Russo's Straight Man: " β€œI know it runs deeper than that. You think we defend incompetence and promote mediocrity.” ... β€œI wish you would promote mediocrity, Mediocrity is a reasonable goal for our institution.”

Agreed, but mediocrity is a fuzzy target. A lot of Australia's problems lie from a pretentious fixation on excellence over competence (too many Nth tier unis, too little TAFE).

When Australians complain about mediocrity, I think of this quote from the excellent book Straight Man (Richard Russo).

25.09.2025 01:05 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

@geofflangdale is following 20 prominent accounts