HNHacker News
TopNewBestAskShowJobs

gojomo

30,216 karma · joined February 24, 2007

Gordon Mohr, maker of software.

Twitter: http://twitter.com/gojomo

Project: http://thunkpedia.org

Idea blog: http://memesteading.com

Older blog: http://gojomo.blogspot.com

Homepage: http://xavvy.com (username @ here for email contact)

My HN peeve is formulaic downbeat comments, like: "How is this news, I already knew this!" "…Betteridge's Law…" "I stopped reading at…"

submissionscomments
gojomo··on It's Time to Stop Taking Sam Altman at His Word
I think your biographical testimony more confirms than refutes my suspicions.

You think "smart" implies a low chance of world-epochal AI breakthroughs that would ruin your "slow-and-steady path to retirement". You think OpenAI has "no moat". You think it's wise to satisfice – taking a "decent" paycheck – rather than shoot-for-the-moon (and beyond it, control of the 'lightcone').

That's closer to the "short-term exit" mindset – get a good-enough payout, in mundane dollars, while you can – that you're alleging in Altman than what I'm seeing in his statements & behaviors. Yes, someone with your beliefs would in fact be trying to pump OpenAI's short-term hype for a timely exit and an even-more-cushy retirement.

But Altman's already a billionaire with any creature comfort he'd want, for the rest of his life, plus at least one "doomsday bunker" (per a 2016 'New Yorker' profile).

For at least 10 years, he's been consistently expressing a desire to advance, and control, breakthroughs in AI, AGI, and ASI that could – in a true believers' worldview – wipe away as obsolete all his prior billions – as well as the "slow-and-steady path[s] to retirement" pursued by others.

He knows as an insider and contributor OpenAI's actual lead over others – who have tended to approximately-reach OpenAI's capabilities a year or more later.

That you rate as OpenAI's "big news this week" their "Artifacts" end-user feature, rather than everything else announced & rolled out, plus OpenAI's gigantic leads in funds raised and revenue run-rate, suggests a particular world-view.

I would suggest that you are over-focused on the kind of small-stakes products/markets/ambitions that are highly-legible to a "software engineer comfortably drawing a decent-but-not-FAANG paycheck at an established company" – but such considerations are barely afterthoughts in the game of power & legacy that Altman is now playing.

gojomo··on It's Time to Stop Taking Sam Altman at His Word
I think you're projecting a short-term cash-out mentality here.

Altman's actions are even more consistent with total confidence & dedication to a vision where OpenAI is the 1st and faraway leader in the production of the most valuable 'technology' ever. Plus, a desire to retain more personal control over that outcome – & not just conventional wealth! – than was typical with prior breakthroughs.

gojomo··on Popular gut probiotic craps out in randomized controlled trial
I know that's the conventional wisdom, but as an adult of European descent who usually handles lactose/dairy well, I sometimes have periods of reduced tolerance, perhaps due to changes in my gut biome.

And there's evidence even lactose-intolerant ("lactase nonpersistent") demographics can tolerate ever-larger amounts of lactose with a gradual ramp-up of consumption that shifts their gut biomes in a different direction:

https://pubmed.ncbi.nlm.nih.gov/38159728/

There's a chance the conventional view overweights human genetics and underweights biome, which could create a self-fulfilling diagnosis of "lactose intolerance".

That is: a person notices a bad reaction, and is told (influenced by demographic tendencies) "oh, as an adult you'll inevitably be lactose-intolerant". They avoid dairy, further thinning (or completely eliminating) the bacteria that could help from their gut, ensuring a more-noticeable negative reaction upon further consumption.

OTOH, if the right bacterial mix was maintained, or reintroduced then maintained if necessary, many considering themselves lactose-intolerant might not persistently be.

(Separate from the bacteria themselves assisting in lactose digestion, there's even some chance that the positive or negative shock of a wildly-changed biome might epigenetically affect the beyond-childhood activation of the human genome's own lactase genes – a possibility mentioned in the middle of this microbiologist's writeup: https://www.cnn.com/2019/04/17/health/lactose-intolerance-mi... )

gojomo··on Popular gut probiotic craps out in randomized controlled trial
I'd never heard a single strain could affect constipation, so these results unsurprising.

Where attention to the gut biome has really saved lives is when a person has massive Clostridium difficile overgrowth – which is more associated with uncontrollable diarrhea than constipation. This state of extreme illness is typically only possible because prior antibiotic use cleared out the natural variety of commensal bacteria that normally keeps this particular strain at bay (or totally absent).

In such cases, people near death can bounce back after a full-spectrum "fecal microbiome transplant" (FMT) where all of the hundreds-or-more of distinct bacterial species are transferred from a healthy donor to the patient.

It's quite hard to package all those sensitive, beneficial, adapted-just-to-the-human-gut strains for reproducible evaluation & administration – so FMT is hard to fit into standard medical studies & FDA approval processes.

But sometimes a mere dot's worth of a healthy person's feces can be a miracle cure for someone else! So, a company has driven an approval of their particular $10k-per-treatment shit-pill through FDA processes, and partially as a consequence, the FDA has been cracking down on people arranging FMTs in less-formal & far-cheaper ways – despite a great record of safety and effectiveness.

A well-informed & highly-opinionated rant on this topic: "A Monopoly on Poop" https://stephenskolnick.substack.com/p/a-monopoly-on-poop

gojomo··on Pro bettors disguising themselves as gambling addicts
Unironically, "blockchain fixes this".

Decentralized prediction/betting markets, on public consensus ledgers (blockchains), can't exclude the best bettors. And, match bettors and clear markets with far less overhead/"vig" – resulting in fairer, market-set odds.

That leaves less room for a small clique of enterprises whose profits are driven by problem betting, and thus capable & incentivized to engage in manipulative marketing, and misleading product offerings, to find and drive the vulnerable to "extinction". (That's a term of art in gambling design, see https://alum.mit.edu/slice/play-extinction-research-reveals-....)

The regulation and licensing of privileged franchises to state-approved entities has made such entities more able to exploit compulsive gamblers than a free and open system would.

Many states have even gotten into the business themselves, with lotteries offering atrocious odds, promoted with deceptive advertising, overwhelmingly patronized by the poorer & more-innumerate.

Go figure!

gojomo··on Two new books on John Calhoun and his rodent experiments
Gwern notes Calhoun hardly piblished anything formal (much less peer-reviewed) about his research - mostly interpretation-heavy summaries of what he wanted to emphasize. Not enough detail rigorously 'replicate'!

But Gwern does mention a few followups that tried similar setups - and didn't really see the same major takeaways. So "nobody has tried" isn't really accurate. Nobody has succeeded!

gojomo··on Two new books on John Calhoun and his rodent experiments
Gwern's done a deep dive into some of Calhoun's work & how it's popularly interpreted – the abstract:

> Did John Calhoun’s 1960s Mouse Utopia really show that animal (and human) populations will expand to arbitrary densities, creating socially-driven pathology and collapse? Reasons for doubt.

https://gwern.net/mouse-utopia

gojomo··on Independent directors of 23andMe resign from board
I don't think you're properly modeling a truly ambitious person whose basic needs are already fully-met for life.

Anne Wojicki is already sufficiently wealthy – net worth $150m+? – that maybe what really interests her is playing hardball for (say) $10B+ instead of just a few hundred million more? Or for the control & glory of shepherding forward some breakthrough cancer treatments that the other investors might simply treat as financial options to sell early?

And, perhaps she's got the votes & de facto IP control & legal budget to think she's got a good chance of winning, and even a loss can't cut her out?

Isn't this just bare-knuckle "founder mode"?

gojomo··on Hezbollah pager explosions kill several people in Lebanon
Yes, but I believe the post to which I was replying used "the initial attack" that Israel "failed to see coming" to refer to Hamas' October 7th attack.

Hence my reference to Hamas' likely chaffing of Israel's intel, and Israel's false-alarm fatigue, before that "surprise" attack.

Of course Hezbollah likely uses similar tactics - but if as deeply pierced as this latest attack implies, they'd have to wonder if any comm successes they thought they had were just Israel leading them on. I doubt Hezbollah has surprised Israel recently!

gojomo··on Hezbollah pager explosions kill several people in Lebanon
It looks like a trigger that can only be pulled once.

Thus, choice of the optimal time could be influenced by a lot of things:

- knowledge of other Hezbollah imminent action making comms disruption right now of great importance

- recognition that the vulnerability had been discovered and was about to be remediated

- via other "eyes on" prime targets, knowledge that just one or two top leaders were briefly in especially-vulnerable positions (like sleeping alongside their pagers)

- etc

And, there will be a "long tail" of damage to Hezbollah's usual communications practices & trust in devices/suppliers. Some marginal recruits may even be deterred from joining a battle against an opponent which can carry out this sort of attack – though of course, others may be emboldened.

gojomo··on Hezbollah pager explosions kill several people in Lebanon
And you thought "Key Escrow" or "Chat Control" was bad!

Future maximum-security states may just require every mobile phone to have an explosive charge pre-installed, with the detonation-codes only available to the authorities, of course.

Then when they detect a device in active use by a [terrorist|child-pornographer|subversive|tax-evader], ka-blewie! Problem solved at the press of a button.

See also: 'Black Mirror' s3e6, 'Hated in the Nation'

gojomo··on Hezbollah pager explosions kill several people in Lebanon
Not really: https://www.nytimes.com/2023/11/30/world/middleeast/israel-h...

They just didn't realize how real this particular plan was. It's quite likely Hamas is regularly bluffing similar attacks in many potentially-compromised channels. If they then see Israel react, they've received info about Israel's capabilities & tactics – and maybe even ways to misdirect Israeli forces with future feints. And any real plans may stay hidden amongst dozens/hundreds of other "false alarms".

gojomo··on The Internet Archive has lost its appeal in Hachette vs. Internet Archive
This histrionic concern trolling about "mortally…damaged" is nutty, given that IA already settled the damages portion with the plaintiffs over a year ago for what seems to be some symbolic amount: https://news.ycombinator.com/item?id=40203774

Perhaps diligent but conservative government archives should be run by your gray ideal of "boards of archivists and librarians".

But the Internet Archive was founded to be an activist organization, in deed and legal strategy. When you start, fund, or staff an archive, you can run it as risk-averse as you'd like.

gojomo··on The Internet Archive has lost its appeal in Hachette vs. Internet Archive
What is this "very sensible middle ground" you think existed as an early-exit option that also served the IA's goals?
gojomo··on The Internet Archive has lost its appeal in Hachette vs. Internet Archive
This is an absurdly strained & non-factual summary of the history.

>one man's seemingly fanatical conviction against the law

Reviewing the large number of amicus briefs on the Archive's side, from the get-go through appeals, refutes the idea this was a solitary crusade: https://blog.archive.org/2023/12/29/friend-of-the-court-brie...

Those supporting IA's position range from the American Library Association (the world's oldest & largest library advocacy group), to individual libraries of all kinds, to expert IP law academics, to public-interest advocates like the Center for Democracy & Technology or Public Knowledge, to fellow open-culture organizations like Wikipedia, Creative Commons, & Project Gutenberg. Also: lots of book authors, including those with commercial success & titles inside the IA's lending program.

The IA was in the leading position, sure – but taking the arrows for a very large group of like-minded organizations sharing a stance against copyright maximalism. Personalizing it as one man's radical crusade is odd.

>asked to stop… tried to open a dialogue

Saying no to the copyright maximalists, even through their claims of absolute control & threatened or actual lawsuits, has been essential in establishing the actual settled law around copyright.

What sort of 'dialogue' can be had when the sides have incompatible views of the law: one believing in a permissionless right to do an exact something (supported by reasoning & precedent) and another asserting an absolute right to prohibit that exact same thing (supported by other reasoning & precedent)? Each side needs to enact their beliefs then resolve it in the courts.

HathiTrust - a major consortium of university libraries – was the named defendent in an earlier lawsuit by some of the same copyrightholder interests with regard to Google Books scanning. (It's also an ally of the Internet Archive in this fight.) Should HathiTrust have rolled over when "asked to stop" scanning by rightsholders? Absolutely not: they won in court & on appeal.

If Sony hadn't appealed the Betamax decision to the Supreme Court, VCRs & everything since that let people record their own copies of TV programs could've been "illegal". A mere 'dialogue' with TV broadcasters or moviemaker trade associations couldn't have done anything: the issues had to be ruled on by legal authorities.

>In addition, there has been real collateral damage to the many noble aspects of the Internet Archive. Legal fees and judgements have diverted resources away from the Wayback Machine, the library of public domain works, and other IA programs that provide real value to society. I truly hope the organization can survive.

I agree that the overheated rhetoric from both the plaintiffs (about giant but never-proven sdamages) and defendants (about how central these principles are to IA) may have created that impression in some coverage – but the idea this was ever existential for IA, in legal costs or potential damages, is pure paranoid fantasy.

As a non-profit, the IA files detailed form 990s with the IRS showing income & expenses. I challenge you to find any hint of legal costs changing other operations in the years since the lawsuit was filed (2020) and appeals launched.

I suspect, but have no inside info, that much of the costs were borne by other advocacy & legal organizations/donors that wanted to pursue a ruling on these particular essential issues. That is: this battle was fought with resources targeted for this program and these legal principles, not resources diverted from other programs.

As part of private settlement with the plaintiffs in 2023 – not any court monetary judgment against it – the IA agreed to make some undisclosed payment but ALSO had permission from the plaintiffs for IA to continue to pursue appeals (like the one just ruled-upon) on the issues important to IA, at no risk of further damages.

That's hardly the "scorched earth" plaintiff behavior implied by some hyped coverage imagining an IA bankruptcy, or other threats to its ability "to survive".

This was always a dispute on some copyright principles; it will be a loss to the public if IA's vision of format-shifted digital lending is ultimately ruled illegal, but no impact to IA's other long-established programs.

Finally: this may not be the final chapter & ruling on these issues. Sony had to appeal all the way to the Supreme Court before getting the Betamax ruling in 1984. Google had to appeal all the way to the Supreme Court before getting a ruling that API reimplementation could be fair use in 2021. I don't know if IA will judge it as worthwhile to appeal. But they might! And before those other historic final appeals, the preceding judgements seemed pretty definitive and bleak for the ultimate victors.

gojomo··on Escaping from Anaconda's Stranglehold on macOS
Well he didn't win a Nobel Prize in Python best practices, that's for sure.
gojomo··on The vagus nerve orchestrates the mind-body connection
You may be able to find a local community college course that offers free/cheap observation or participation in human cadaver dissection.

This 2022 LessWrong post mentions a $60 5-hour option at the Sf Bay Area's Merritt College (in Oakland):

https://www.lesswrong.com/posts/5PBWgHiCiiJHjPRSn/what-it-s-...

gojomo··on Intel SGX Fuse Key0, a.k.a. Root Provisioning Key Was Extracted by Researchers
Some of Signal's designs for contact privacy, including in the new usernames feature, rely on trust in SGX.

If anyone (including Signal) can pretend to be a secure SGX environment, you're back to trusting Signal's personnel/operations, rather than Intel/SGX, for some of the metadata/contact privacy they've historically touted.

More info (2020): https://medium.com/@maniacbolts/signal-increases-their-relia...

gojomo··on Cautionary tale on using Chase bank for indie business
Crypto solves this. Well, eventually, after initially making it worse, because all your crypto-adjacent transactions in the meantime make legacy banks extra-hair-trigger suspicious.
gojomo··on Eyechat
Ultimately it will likely be easier to simulate eye-contact with live restyling – that is, synthesizing the view from a virtual 'camera' using one or more other cameras nearby – than physically hide a true camera inside a monitor. (Simulation could also signal eye-contact with any point on the screen, not just a single camera location.)

Nvidia, Apple, & others already have software for this, which will only get better.

gojomo··on NASA says Boeing Starliner astronauts may fly home on SpaceX in 2025
Even if up for 8 months – and returning to a US with a different President, perhaps even a different party-of-the-President, they'll not match the experience of Sergei Krikalev – who traveled to the space station Mir for the USSR, & was for a while stuck there when the USSR dissolved, only returning 311 days later:

https://en.wikipedia.org/wiki/Sergei_Krikalev

He later became the 1st cosmonaut to fly on the US Space Shuttle:

https://historycollection.jsc.nasa.gov/history/shuttle-mir/p...

gojomo··on Boeing's Starliner proves better at torching cash than reaching orbit
Hey now! It got two astronauts into orbit just fine. They just haven't yet figured out how to safely bring them back.
gojomo··on Wells Fargo Bet on a Flashy Rent Credit Card. It Is Costing the Bank Dearly
That'd be a good plan, if you could count on the loot flowing fairly & proportionately to any commoner who buys shares on the public markets. But in a rentier racket this big, that doesn't happen.

Large amounts go to management insiders, or are kicked-back to politicians & their allies or pet causes via donations or sweetheart deals to favored projects – which might show up as "losses" on "failed projects" eventually, but hey, it can all be covered out of the privileged rolling take from everyone on the outside.

gojomo··on Agricultural drones are transforming rice farming in the Mekong River delta
Ultimately drones may enable purely-mechanical pesticide-free removal of each individual insect/weed.
gojomo··on Comma Gets a Cure (2000)
So: this would be an ideal passage to read to help a voice cloning system get a full sample of an English speaker's accent/tone?
gojomo··on Wells Fargo Bet on a Flashy Rent Credit Card. It Is Costing the Bank Dearly
Yes, because money is fungible, and Wells-Fargo (& its "systemically-important" peers) receives via its privileged positions a de facto non-stop rolling bailout in addition to any occasional "crisis" larger headlingebailouts.

This fumble – at "just" $120mm losses per year – is still pretty small compared to its billions in income. A drop in the bucket!

But the reason they can take, and survive, any number of such unwise swings at no material risk to their insiders & shareholders is that their overall income is largely a function of their oligopolistic power & bottomless access to cheap federal money. They just have to avoid drawing on it "too much", in a "too unseemly" manner.

They can push that advantage a little more when they lose more elsewhere, as long as their coarse risk and overall results look similar to their peers – who are similarly pressing their special advantages around the edges.

So there won't be a specific measly transfer from public funds, "this is covering your failed Bilt program", of course. But they can just lean on their advantages a little more, and "keep up with the Joneses" in their peer banks in coarse indicators, with everyone at the trough (executives, politicians, politicians' pet projects, major shareholders) not facing any even marginal negative feedback.

And in some next major macro reversal which puts all banks of the same class in danger, it'll all be papered over again – bringing them up to some level of stability without regard to how many extra hundreds of millions leaked through Bilt-style errors or sweetheart deals to favored groups. So de facto, retrospectively, all those leaks were "free" to the insiders.

gojomo··on Booking.com ignores twofactor, lets everyone email-login without a password
I see the top comment in your top link makes the same recommendation I made in another comment here (https://news.ycombinator.com/item?id=40720789).

If it works, you should definitely solve the problem for yourself even before booking.com eventually suffers enough to address the problem more generally.

gojomo··on Booking.com ignores twofactor, lets everyone email-login without a password
Sounds like they're under active attack due to some poor initial practices & having a hard time getting in front of it.

I suggest changing your email with booking.com to something the attackers wouldn't know.

Using the Gmail option of extending your normal username with '+' something – eg use ACCOUNT+unguessable-string@gmail.com in place of ACCOUNT@gmail.com – might be enough. With luck (if the site hasn't been too dumb), then when they hit the site with your old/plain address, no email will be generated.

gojomo··on Wells Fargo Bet on a Flashy Rent Credit Card. It Is Costing the Bank Dearly
As one of the "systemically important" (too-big-to-fail) banks, Wells-Fargo can make many bets that, when they lose, will ultimately be covered by taxpayers in the next round of bailouts, but if they win, will reward insiders & shareholders in the periods between bailouts.
gojomo··on Wells Fargo Bet on a Flashy Rent Credit Card. It Is Costing the Bank Dearly
For those people who don't "make mistakes", the card issuers are still earning:

• part of the 2-3% added to every charge

• fees from promotional marketing, guided by the info in your purchase history, to those customers

• fees from related services the same customers may purchase from the same bank

Even when the high rewards cards marketed to conscientious low-credit-risk customers send back some of that 1-3% as "miles" or "points" or "cash back", the other information/marketing values about high-spending, reliable-paying customers remain interesting.

← PreviousPage 5 of 34Next →