Hacker Newsnew | past | comments | ask | show | jobs | submit | more throwawayjava's commentslogin

> The fact that the story can't be repaired does not rebut the ColanR's claim that deleting it is inconsistent with NPR's stated policy

The point is very simple: removal != retraction. Removing without retraction is erasing history. Issuing a retraction is not erasing history.

And obviously, in this context, "erasing history" means the thing it means in standard vernacular English. Not the thing it means in git terminology.

> Furthermore, the best policy would be to retain a complete edit history of the story online... I can't see a reason for not making the edit history public other than reducing embarrassment of NPR.

1. How does showing the edit history reduce embarrassment to NPR?

(e: Ironically, if reducing embarrassment were their goal, they would've found a way to keep the story as in-tact as possible. "Look, we weren't 100% wrong. We just had to change a few paragraphs". IMO a full retraction is far more embarrassing than trying to weasel out of it by keeping as much of the story in-tact as possible.)

2. Some real downsides to this approach are already debated down-thread.


> The point is very simple: removal != retraction.

Exactly. NPR should retract the story -- that is, admit mistake and add a clear statement that they do not endorse it -- but they should not remove the text from the internet. The policy ColanR quotes says content should not be removed, and everyone agrees that content can be retracted.

> Removing without retraction is erasing history. Issuing a retraction is not erasing history.

But issuing a retraction and and removing the text is (an attempt to) erase history in standard vernacular English, i.e. make it difficult for others to figure out what happened in the past by hiding evidence. It's true that silently removing the story is even worse, because it erases the history of the article's existence rather than just erasing the history of the article's content.

> 1. How does showing the edit history reduce embarrassment to NPR?

You misunderstand my claim. Showing the edit history would increase embarrassment for NPR.

> Ironically, if reducing embarrassment were their goal, they would've found a way to keep the story as in-tact as possible.

This is not true if the error was so egregious that it couldn't be defended with a straight face. Given the constraint that the story must be retracted (i.e., un-endorsed), it is clearly more embarrassing for them to leave the retracted text online than to remove it from the website.

> IMO a full retraction is far more embarrassing than trying to weasel out of it by keeping as much of the story in-tact as possible

You misunderstand. None of us think they should weaseling out of it. I am comparing what they have done to the better procedure of (1) retracting the story (in the sense of declaring that it is flawed and not defensible) but (2) leaving the text of the story publicly available with a disclaimer.

> 2. Some real downsides to this approach are already debated down-thread

This thread has dozens and dozens of comments, many of which are flawed, and I'm not sure what you're referring to.


I think the problem is that a link that says “RETRACTED: jessriedel is an axe murderer” has the potential to do more harm on an ongoing basis, both ethically and legally (as in, a court might see this as contributing to ongoing damages in a libel suit).

The best thing for both parties is to remove the original text and post a retraction notice explaining what used to be there and why it’s now gone.


If this were NPR's real motivation, they would just censor the subject's name.


> "erasing history" means the thing it means in standard vernacular English

I'm a native speaker of English and I'm not sure what that means.


> Publish a retraction. What the fuck is this!?

...this is a retraction?


> the policy that NPR pointed to (and that I quoted) is more general than that

I think the salient point is that there's a huge difference between a silent removal and a public retraction. The practices that NPR indicts in your original post obviously refer to the latter. Removing without retraction is certainly an attempt to "erase history". Publishing a public retraction and discussing it at length is not "erasing history".

NPR's executives have even explained why, in this particular case, they chose to remove the article: Turpin said, "In order to fix the problems, we would have essentially had to construct a different story," a process that he said would have been "intellectually dishonest. There comes a point that you can't fix a story appropriately." He said the decision was not made lightly; "we believe strongly we don't want to disappear material for any reason if we can possibly keep from taking it down."

edit: Essentially it's the difference between a ghost edit and an "edit: I changed xyz in my post". Do you really believe that silently /dev/null'ing an article is equivalent to a public retraction?


> What is the meaning of ownership if the land is not really yours?

You can't own the beach. And you can't de facto own the beach by buying all the land around it and then restricting access. These were both well-understood when Khosla bought the land. He knew exactly how California interpreted land rights in this particular case.

Land purchases regularly come with both private (contractual) and public (statutory) limitations/restrictions on use.

Landowners are not kings, and private properties are not sovereign borders.

> But then, the state should not have allowed the sale, right?

I guess that's one way the state could react to these sorts of lawsuits. That would be unfortunate, because it'd probably end up meaning that one asshole prevented a whole bunch of reasonable people from buying beachfront views with their hard-earned money.


> meaning that one asshole prevented a whole bunch of reasonable people from buying beachfront views with their hard-earned money.

I don't understand this point. You don't have to purchase the road leading to the beach to own beachfront property.


But in practice many developments do have the effect of blocking off access to a piece of beachfront. The state doesn't interfere with those developments, working off of the assumption that the owners are reasonable folks and will follow public access laws.

If the only way to get people to behave reasonably is to not issue building permits, then everyone loses. What you're suggesting is that California should choose the defect/defect quadrant of a game. But there's a more optimal solution -- just tell people like Khosla they're not allowed to defect on this question.

In other words, why should I be prohibited from owning certain types of land or having the chance to purchase into certain types of developments just because Khosla's being an ass about restrictions he knew about when he bought the property? What you're proposing is the "if I can't have it then burn it all down" solution. Which I guess is a way to vindicate Khosla in a sort of pyrrhic way, but everyone else loses.


> And all it would take is ten years of shared, modest pain and discipline.

Modest for who?

Social Security: There are an absolutely insane number of people who didn't plan for self-sufficient retirement because they planned on social security checks. They shouldn't need it, but they do. Those people are a far larger problem and constitute the bulk of avoidable liability. The problem with SS is people who need it but shouldn't have, not the people who actually don't need it.

Military: Here's a truth that doesn't win elections: blind patriotism is the only palatable way to sell any absolutely enormous jobs program to the US public. A lot of these folks don't have the education or skills required to operate in a private sector unbuoyed by gov't spending. And to the extent that they do, they'd be entering newly saturated job markets with low barriers to entry.

Gas taxes: Carnage for the lots of folks who can't afford to true cost of driving but already sunk 30+% of their annual gross into a car as a 10+ year investment.

The policies you're suggesting -- at least, most of them -- would leave an enormous amount of human suffering in their wake.

Still justifiable policies / good ideas? Well, that's another discussion. But let's be realistic about the impacts.


controls and sensing.


If I'm a patient being diagnosed by a black box, I'd rather have 90% accuracy. If I'm a doctor or researcher trying to effectively treat a type of cancer, I'd much rather have the explanation.

Come to think of it, 90% is pretty low for a course of totally unnecessary chemotherapy that cold've been avoided by a human doctor noticing, as humans often do, how dumb the provided explanation was. So maybe even as a patient this isn't an obvious choice.

More generally, I'm super amused by the idea that the Master Algorithm will fall out of the brilliance at few big tech companies, but only if they could have access to my purchase history or porn viewing habits from the past decade.


I’d want both the 90% as well as the explanation. While some parts of these NN are black boxes, many parts are not. I’d want to know what data the model was trained on; if there were any biases in the patients tested so far; were there any issues with batch effects, et cetera...Like you said, if the question of chemo was on the line i wouldn’t want to go with just a black box.

That being said, at some point in the future, when these types of problems with NN can all be ironed out, I think going with the NN will be the obvious choice. The massive numerical complexity of cancer and the human body is too large for any human to understand and treat optimally.

NN + doctor for now; just NN in the (relatively near) future.


If the human filtering the results improves the 80% to 85%, then it's still better to use the 90% unfiltered machine. You assume the human will actually improve the accuracy at all, not reduce it, which means he's smarter than the machine in some aspect. If the AI is very good, he might more often degrade the accuracy by rejecting things he doesn't understand or that he has an incorrect bias/opinion about.

Remember that AlphaGo game where the commentator thought it must have made a mistake because it did a move that was obviously wrong but then it surprised everyone by winning because of that?


I think the larger point he's trying to make is that it's foolish to regulate the potential out of emerging technology (NN) and make it harder to discover new technology (the penicillin and X-ray of their time).


Most class definitions are classist.


> software engineer salary somewhere else is usually more like 20-25% less rather than 50% less -- someone making 140k in SF might make 110k-120k in places like Boise or Denver.

Worth pointing out that this function doesn't work for inputs much higher than $150k or so, and doesn't even work at points like $140k depending on the particular candidate.


$/sqft isn't a good metric. The correct metric is "what am I actually going to spend on housing?"

This way, the answer begins to change as your family size grows. Which I think is accurate.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: