HNHacker News
TopNewBestAskShowJobs

anjc

1,414 karma · joined October 24, 2011

submissionscomments
anjc··on Ctrl Shift Face: A growing ‘deepfakes’ YouTube channel
Same here. Jim Carrey's face seems to morph into John Wayne's and James Dean's when he does impressions of them. I thought the same was happening with Bill. I'm aware of deepfakes but yet it took several watches with me going "what the heck is going on here" to notice the almost seamless transitions.
anjc··on Delisted Overnight: A Cautionary Tale for Indie iOS Developers
Is the functionality of your app already included in the Health app? If not, take a guess what feature will be soon included in the Health app. Apple have a history of killing competitors in the App Store, most recently with screen time apps:

https://edition.cnn.com/2019/04/27/tech/apple-screen-time-ap...

anjc··on Washington state lawmakers approve human corpse composting
> They don’t need to be viewed after they’ve been identified.

Says who? Do you have information about the grieving process being easier if you don't view?

anjc··on Modern SAT solvers: fast, neat and underused
Why are people ITT talking about reducing problems to SAT? As I understand it, the point of reducing a problem to SAT is because there's a proof that SAT is NP-Complete which in turn would prove that your problem is NP-Complete. But if you have some sort of industrial constraint satisfaction/ASP/scheduling/search/etc problem, you can solve it with whatever methods are available. You don't have to reduce a problem in such and such domain down to a big Boolean formula and solve it via SAT techniques. Am I wrong?
anjc··on Modern SAT solvers: fast, neat and underused
Is meta-learning still useful for SAT solving (e.g SATzilla)?
anjc··on When Doctors Thought ‘Wanderlust’ Was a Psychological Condition
Access to water in Ireland is assumed to such an extent that it isn't even listed in the ESRI's criteria for poverty. As such, it's very obvious that lack of bathing/changing clothes will almost NEVER be due to lack of access to water. Even temporary halting sites must have access to water considered by local councils.

> From a privileged, lower-middle class or above perspective

Very confused as to why you think 'water' might be limited to the demographic who work as teachers, Gardaí, and above, or why access to water indicates privilege.

anjc··on Source code for Zork, Hitchhiker’s Guide, and other Infocom games
Who's like us!
anjc··on Katie Bouman Accidentally Became the Face of the Black Hole Project
> On the other hand, assuming the Post article’s mention of an “algorithm” is meant to refer to CHIRP, it is in error, since CHIRP was not used to create the current images.

Yeah? You wouldn't find this confusing as a lay person? Many articles said she created the algorithm. As I said, she didn't "accidentally" become the face of the image. It's due to these journalistic errors.

anjc··on Katie Bouman Accidentally Became the Face of the Black Hole Project
My mistake, it was in the other article I linked to.

Her current/future employers, her colleagues, and every academic's employers and colleagues, want to know contributions. Some fields make you explicitly list them for each paper. No doubt feelings of inadequacy feed into this at some point, but it's part of the job and isn't saved for just measuring women.

anjc··on Katie Bouman Accidentally Became the Face of the Black Hole Project
I'm confused about this. People said her code contribution on the repo was minimal. Others in the field came out and said "yes but her algorithm was key here, not lines of code". Fine. Her own statement then says that "no one algorithm was key, it was a group effort", but her colleagues say her algorithm wasn't used at all. The obfuscation doesn't seem accidental at all.

Why is a person sexist and homophobic [1] for wondering why two major news articles say the opposite thing [1,2]:

[1]: "Katherine Bouman, a 29-year-old postdoctoral researcher who developed an algorithm that was key to capturing the stunning visual." https://www.washingtonpost.com/nation/2019/04/12/trolls-hija...

[2]: "While she led the development of an algorithm to take a picture of a black hole, an effort that was the subject of a TED Talk she gave in 2016, her colleagues said that technique was not ultimately used to create this particular image." https://www.nytimes.com/2019/04/11/science/katie-bouman-blac...

anjc··on Geek Social Fallacies (2003)
> This is a shining example of old web content we've lost in the current era of walled gardens and social graphs.

Five AMAZING geek social fallacies you never learned in school. You wont believe number 4!

anjc··on Google’s new AI ethics board is already falling apart
> But I looked, and could only find her expressing concerns about restrooms and sports teams.

> If that's not "openly hostile to trans people" I don't know what is ... There's room for reasonable discussion on issues like the inclusion of trans folk in competitive athletics

Discussing trans in sports is which one? Hostile or reasonable?

anjc··on How Sitting For Long Periods Affects Your Metabolism
Thanks for reply. Regarding alpha value: you seem to say that with a small sample size, it'd be possible to see no effect between groups. I think you were implying that they should've been testing p<0.0001 or something? But I'm not sure how to choose this value, without knowing the difference between means that you expect for the medical intervention.
anjc··on How Sitting For Long Periods Affects Your Metabolism
Ah thanks, I understand this aspect. As I say, they simply failed to rejected the null hypothesis, and their approach here seems perfectly valid.

The authors did overreach in their interpretation of the result. But OP said ANOVA is wrong, and that the alpha value is wrong. I don't understand how they can say this without understanding this domain, the intervention, and knowing what effect size could be expected. Maybe α=0.05 is perfectly reasonable here.

anjc··on How Sitting For Long Periods Affects Your Metabolism
Can you explain this common mistake further?

> They're claiming there is no difference between groups because they were unable to show that, if there was an inherent difference, the chance they'd see these results is at most 5%.

It seems to me that they're claiming that they failed to reject the null hypothesis, the end. I.e. according to this statistic, there was no significant difference in the means between all groups in the ANOVA. This was a small sample size, but the analysis seems to include multiple trials for each participant.

This seems like a standard use of ANOVA to me. So how can you say that the alpha value is wrong just based on this one sentence from the abstract? It seems to me that you'd have to know the domain, know what effect to expect, and so on

anjc··on There is no reason to cross the U.S. by train, but I did it anyway
Unreadable and bizarre article. Congrats to the author on rubbing shoulders with the hoi polloi and getting a...train.
anjc··on Treat Medicines Like Netflix Treats Shows
> I don't want to put words in their mouth, but I think it might be a bundle vs a la carte argument. For issues that tend towards short term treatments (eg a handful of antibiotic pills to mitigate the risk of wound infection) then you don't want to pay the same total price as someone who is treating some chronic condition.

Yeah this is what I meant. The article says that the Australian government offset the high cost of new Hep-C drugs by negotiating a flat rate of $7.5k per person per year for any Hep-C drugs, and then talks about the high cost of penicillin and epipens.

Nobody would use Netflix if the subscription fee had to account for films that cost billions in R&D and only 10 people were going to watch.

anjc··on Treat Medicines Like Netflix Treats Shows
Am I missing something:

>A study just published in The New England Journal of Medicine found that on average, Australia pays $7,352 (U.S.) per course of treatment. It has been able to treat seven times as many patients as it would without the agreements.

I guess the reason we don't treat medicine like Netflix is because most people don't want to spend $7352 for a course of penicillin....?

anjc··on Sinclair ZX Spectrum Prototype
I can't see if it's an emulator or uses a real z80 etc. Looks cool though
anjc··on Alleged Coinomi exploit shows how easy it is to have Bitcoin stolen
Between people losing their keyphrase, to software wallets being hacked, to hardware wallets being compromised...at what point is it more safe to just keep your coins on a reputable, insured, exchange? I think the odds of Coinbase doing a Mt. Gox are a lot more slim than the odds of a random person screwing up their own storage solution.
anjc··on A sold-out city? The fight to save Dublin’s nightlife
This must be a route-dependent thing (e.g. Sandymount routes would have lots of students on them). My routes rarely/ever have trash or liquids on them, hit branches etc.

Again re: seats and engine....these are the same seats and engines used in Germany and England also.

Still, I prefer the trains to buses in Dublin.

anjc··on A sold-out city? The fight to save Dublin’s nightlife
> There's buses. They're ugly, loud and extremely uncomfortable

Ireland didn't invent these buses just to annoy you, the Volvo buses used in Dublin are used worldwide, including London and Berlin. They're as modern as you can get.

anjc··on Some Details of My Personal Infrastructure
It's obvious that these people are not stupid and understand that interfaces need to be documented. They are instead confused. I've had micromanagers like SW before, and have been confused many times at the state of something precisely because the micromanager took it upon themselves to "do it themselves" on the fly, just like SW did here.

Admitting that it's clearly bad and saying you'll fix it by tomorrow would be a neverending pursuit, because the micromanager's meddling is constant and touches everything.

No doubt somebody else will now be confused that this third parameter is suddenly documented when it was a task delegated to them, and somebody else is confused because they were told to deprecate the parameter.

anjc··on You Do Not Need Blockchain: Popular Use Cases and Why They Do Not Work
Any arguments like this are immediately nullified because we have real-world examples of blockchain doing astounding things which can't be done without it. E.g. cryptocurrencies such as Bitcoin.

Can you transfer value across the world without Bitcoin? Yes, you can use banks, you can use wire payments, you can stuff gold in your butt and hope it doesn't show up in Airport scanners.

Can you verify that goods have been delivered without blockchain? Yes, you can insert to write-only dbs, you can hope that everyone in the supply chain is honest, etc.

There being an alternative to blockchain for a task doesn't mean blockchain isn't highly suitable for that task.

These same anti-Cloud-hype arguments were spewed throughout the 2000s, yet look how the world has changed because of it.

anjc··on Which Face Is Real?
It's interesting how, aside from a few infrequent and slight glitches, the artificial faces look perfect, but yet we still intuitively know which one is real based on lots of other cues (the background, the pose etc).
anjc··on Disinformation and ‘fake news’: Final Report published
> If there were significant negative side effects of vaccines - of an order which negates the benefits of those vaccinations - I would argue that we'd know that by now.

New vaccines are created and may have side effects. E.g. H1N1 vaccine causing narcolepsy in Europe recently: https://www.cdc.gov/vaccinesafety/concerns/history/narcoleps...

Using the phrase "the science is settled" is silly, given that the creation and manufacturing of new vaccines isn't settled.

anjc··on OpenAI Trains Language Model, Mass Hysteria Ensues
> saying "X is good/bad" 3 times as a persuasive argument

I didn't say that it's a persuasive argument, I said that it can be persuasive IN arguments. There's nothing sad about an AI learning it, or people being happy with it, it's very impressive.

anjc··on OpenAI Trains Language Model, Mass Hysteria Ensues
The parts you criticise are the parts I was most impressed with. These sorts of repetitions can be persuasive in writing/arguments, and it's impressive to me that a model learned this type of writing.
anjc··on AAAS: Machine learning 'causing science crisis'
If you've faked your data to get these results then it could mean the end of your career, hence must be a very uncommon practice. This is a different from cherrypicking data, or putting a positive spin on your findings. But again, it's expected now that data and code are open, so you wont get away with much these days in high-impact conferences/journals.
anjc··on AAAS: Machine learning 'causing science crisis'
Luckily the academic system will typically weed out unethical behaviour (e.g. faking results).
← PreviousPage 8 of 33Next →