Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
eli
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
38 ms
·
661.
▲
by
eli
3y ago
Fewer police = suffering?
662.
▲
by
eli
3y ago
You can report phishing sites really easily here https://safebrowsing.google.com/safebrowsing/report_phish/?h... Or alternatively report an abusive google ad here https://support.google.com/ads&#x
663.
▲
by
eli
3y ago
Isn't it the biggest after godaddy?
664.
▲
by
eli
3y ago
Does Mike Rowe have some particular expertise in AI that would make his opinion more valuable than the thousands of other commentators saying it will ruin and/or save the world?
665.
▲
by
eli
3y ago
But can't you already hyperlink product.zip to any URL?
666.
▲
by
eli
3y ago
I'm having a hard time imagining users who would be fooled by https://wellsfargo.zip but not fooled by https://wellsfargo.inc
667.
▲
by
eli
3y ago
So this is bad because .zip sounds vaguely like something related to attachments? I dunno that doesn't sound much worse than wellsfargo.info to me.
668.
▲
by
eli
3y ago
I get the concern at a high level, but it seems weird to not have any specific examples or a PoC. Couldn't you already send someone an email or direct them to a webpage where the text "financialstatement.zip" is linked... any
669.
▲
by
eli
3y ago
Who are the unbiased truth tellers here? House Republicans?
670.
▲
by
eli
3y ago
Was this the most damning fact uncovered by the House investigation into this report? That there was a researcher working on it who wasn't properly credited? That in itself doesn't seem like reason to discard the report, let alone
671.
▲
by
eli
3y ago
Yup everyone who disagrees with you is taking payoffs as part of a big conspiracy. What other explanation could there be for why people disagree with you?
672.
▲
by
eli
3y ago
No, I mean a majority of all experts around the world. And I don't find it particularly damning that the director of the National Institute of Allergy and Infectious Diseases was involved in the commission of one of many studies to det
673.
▲
by
eli
3y ago
Because of all the conspiracy theories, it is not going to be pretty if there's a lockdown in the future for any reason whatsoever. Indeed, I think governments will be very reluctant to order a lockdown even if it's the right thin
674.
▲
by
eli
3y ago
The evidence for the lab leak theory is weak. You're assuming facts that most experts dispute.
675.
▲
by
eli
3y ago
The NIH, the scientists doing the research, and most other experts in the field disagree. Partisan politicians are the ones yelling about how dangerous it is. I think I'll listen to the experts.
676.
▲
by
eli
3y ago
Compared to staring directly at the LED panel?
677.
▲
by
eli
3y ago
Biden called for the legislation in his State of the Union speech, so of course Republicans feel obligated to obstruct it.
678.
▲
by
eli
3y ago
it's a New Relic competitor
679.
▲
by
eli
3y ago
It took a lawsuit though. The FTC should just mandate the fees be shown up front.
680.
▲
by
eli
3y ago
I like Mastodon, but "too complicated" is a legitimate concern and maybe undersells it. It's hard to use and puts a lot of burden on the users. Also seems like it's hard on admins, who have to deal with bad actors on the
681.
▲
by
eli
3y ago
Political vigilantes? Decentralization cannot possibly increase diversity if it doesn't provide strong user safety measures. And decentralization makes that infamously hard problem much harder. I think you would end up compromising on
682.
▲
by
eli
3y ago
“Censorship” is popular because it works better than the alternatives. The unfiltered internet is an endless stream of spam and junk and crypto scams. Few people want every message sent to them to land in their inbox.
683.
▲
by
eli
3y ago
Right I’m saying I don’t particularly care about the governance structure of the protocol underlying my social media network. A centralized network run by a nonprofit or a b-corp and a relatively open API sounds great to me. User safety is
684.
▲
by
eli
3y ago
Strong user safety features are at odds with a federated protocol. IMHO federation is a mistake.
685.
▲
by
eli
3y ago
I don't follow your point. "Dark" is obviously a much broader word that can mean many different things. It is less precise than "deceptive" and here it's being used metaphorically while "deceptive" is
686.
▲
by
eli
3y ago
The IETF for one https://github.com/ietf/terminology/pull/9 My intent wasn't to appeal to authority though. I think Deceptive Patterns is just plain better. A term being popular doesn't prove it is
687.
▲
by
eli
3y ago
I don't think you or OP or anyone else here is trying to offend. Nor do I think most people hearing the term would find it offensive. But like I said: the alternative is both more descriptive and avoids the issue. This feels like an
688.
▲
by
eli
3y ago
"Deceptive patterns" is now the preferred term for what used to be called "dark patterns" It's both more descriptive and avoids equating "dark" to "bad". I've also heard "hostile patter
689.
▲
by
eli
3y ago
I'm not sure about that, but even if true: public transit can scale up to support a whole city using current technology. Driving can't.
690.
▲
by
eli
3y ago
The point of the article is that the per-mile cost of cars is actually much higher than you think if you include the externalities. Autonomous cars that work as you describe are not anywhere close to being ready. And I don't think the
More ›