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

So then why does no one offer 99.999% uptime guarantees in writing?

It should be low risk to offer such guarantees then.


Well, (a) why would they? (b) "uptime" has shifted from a binary "site up/down" to "degraded performance", which itself indicates improvements to uptime since we're both pickier and more precise.

Are we really questioning why cloud providers would offer better uptime guarantees?

Yes, I'm asking why they'd lock themselves into a contract around 5 9s of uptime since the parent poster mentioned that they won't do so. Of course, AWS actually does do this in some cases and they guarantee 99.99% for most things, so it feels a bit arbitrary - 5 minutes vs an hour, roughly.

So then its clearly not as trivial to achieve as you made it sound.

You can certainly sign a contract for five nines SLA with cloud providers.

You just won't like the price.


Then it’s clearly higher risk?

If you are asking this question you don't understand what it takes to hit 5 nines in a real life measured system.

Who said signalling would be limited to just 1 thing at a time?


Can you link a source for it? That sounds too absurd to be true…


It’s not that absurd and happens all over the world in university systems. I had a Comp. Sci. Professor that taught assembly and graded on a curve. As you might imagine the one guy that was a wizard at assembly caught flak from the unwashed masses.

I had another professor that not only did a curve but dropped statistical outliers to prevent this problem, he literally explained his system on Day 1 of the course. This was 15+ years ago and by no means a new idea.


The future is not evenly distributed.

I tried to search for it, but even the 2 documents that superseded the one from around the time my daughter was at school at not available.

I mean, the site doesn't even have a valid secure certificate so...

In the site below (In Spanish) you can search for 10/2019 and a cursory translation of the document title will show that this is the proper document (For 2019 onwards, the replaced doc 04/2014 isn't available either)

https://koha.chubut.edu.ar/cgi-bin/koha/opac-search.pl?idx=k...


How do you reconcile the fact that many people in Anthropic tried to hide the existence of secret non-disparagement agreements for quite some time?

It’s hard to take your comment at face value when there’s documented proof to the contrary. Maybe it could be forgiven as a blunder if revealed in the first few months and within the first handful of employees… but after 2 plus years and many dozens forced to sign that… it’s just not credible to believe it was all entirely positive motivations.


Saying an entity has values doesn't mean the entity agrees with every single one of your values.


The desire to force new employees to sign agreements in total secrecy, without even being able to disclose it exists to prospective employees, seems like a pretty negative “value” under any system of morality, commerce, or human organization that I can think of.


That's a perfectly fine belief to have. I might even agree with you. But you're not really advancing a discussion thread about a company's strong ideals by pointing out some past behavior that you don't like. This is especially true when the behavior you're bringing up is fairly common, if perhaps lamentable, among U.S. corporations. Anthropic can be exceptional in some ways while being ordinary in the rest.

(I have no horse in this race. But I remain interested in hearing about a former employee's experience and impressions about the company's ideals, and hope it doesn't get lost in a side discussion about whether NDAs are a good thing.)


You dont believe it increases the probability that Anthropic may be hiding other unsavory things too?

I can see a very charitable person only seeing a small increase, but a literally zero change, and therefore zero relevance, seems absurd.



Are you confusing me with someone else’s comment?

This doesn’t address my question on what you believe.


Read the beetle example in that article. It's exactly on point.

You believe Anthropic is a rare subspecies of beetle (an "unsavory" company) based on a certain pattern on its back (certain NDA-related behavior). I and several others here have noted that lots of companies have that pattern on their backs. Which means that you are basing your conclusion on weak evidence. If you use Bayes Theorem to calculate the actual probability, you'll find that "[trying] to hide the existence of secret non-disparagement agreements" barely moves the needle at all. Does it move the needle? Sure. But much less than you think.


Even if it only moved the needle a tiny amount… that’s still a non-zero amount?

And therefore a non-zero amount of relevance?


Your original point carries an infinitesimal amount of weight. Yes, you win.


Win what? You haven’t even advanced a coherent argument yet… hence the original reply.


Lots of companies do it. Doesn't make it right, but HR has kind of become a pretty evil vocation, these days. I don't believe that they necessarily reflect the values of their corporations. They tend to follow their own muse.


Okay — but if Anthropic is typical banal evil in that regard, why should we believe they didn’t also compromise in other areas?

The exact point is that Anthropic is unexceptional and the same as other corporations.


I thought there were systems designed to effectively negate users that submit too many misleading posts.


Your parent post isn’t suggesting it’s always the same user submitting, just that users submit a lot of posts from this person.

Can’t say I agree, though. I don’t recall ever having seen one of his posts on HN, and a cursory search suggests they’re not even upvoted that much. Highest I found was under 30 points. But my methodology is flawed, as I basically searched for the name.


Stripe needs all that byzantine fraud prevention, on top of what they had a decade ago, because they are a huge concentrated target.

A smaller firm could be way simpler. Because they simply wouldnt have enough money to provide a decent payday for dozens of malicious geniuses going at them 24/7/365.


Is this true? I would expect most of Stripe's fraud overhead to be statutory in nature, not something they hire for because they're a concentrated target.

(They certainly have more staff because more volume, but the actual regulatory requirements I'd expect to be roughly the same for the service they provide.)


When we used Stripe, we opted out of all their fraud prevention stuff to save money (not sure if that's still an option). As a b2b SaaS where payment happens after a free trial (not at signup), we're just not a target for fraud, so it was totally fine.

I can't speak to why Stripe's fraud protection is so expensive. Is it because they're a target? Or maybe because they realized people will pay for it (it seems valuable for something like ecommerce)? I dunno, but I can confidently say that as of ~5 years ago, it wasn't required by any regulation, and my business was perfectly fine without it.

Now we use Paddle, and they also try to sell us a bunch of stuff we don't need at ridiculous prices. We're just using them because we wanted a merchant of record (where they handle taxes and stuff), but no, I'm not going to pay a % of my revenue for basic dunning emails, fraud prevention, vague "optimizations" that "increase conversions" (lol no they don't), etc.


Look at what happened to, say, Cards Against Humanity: You don't have to be a really bit store for some random card tester to ruin you.


what happened with them? I'm not aware of it


Oh, that makes sense. I was thinking fraud as in AML requirements, not fraud as in scammers and card theft.


Stripe was already a big target for basically anyone and anything 10 years ago. Fake merchants, card testers, the works. People were selling guides to defraud Stripe. And we are not even counting just losees due to nonsense like the Fyre festival.

You really don't have to be that big a payment processor for dozens of malicious geniuses to decide that they want to fleece you. If anything, the ROI is better in less sophisticated companies. Most ways to trick a payment company are, if anything, standardized. The smaller company can often be attacked by just changing the API calls, but otherwise taking basically the same actions you would to try to defraud a bigger fish.


I meant geniuses in the real sense, not colloquial sense.

If someone could reasonably expect a cushy 40 hour/week seven figure job, even a malicious personality wouldn’t risk criminal fraud without a much much bigger payout.

And to have dozens focusing on one company…

So anyone handling under a few hundred million per day are safe from that kind of coordinated attack.


> Stripe needs all that byzantine fraud prevention, on top of what they had a decade ago, because they are a huge concentrated target.

This is not true. Every payment processor needs this effort because as soon as you broadcast that you're a payment processor you're going to get about 3-5 scammers a day.

As an aside I really think Mercury bank should audit their onboarding process.


This obviously cannot be true, otherwise Costco would have been sued to oblivion for “dumping” their rotisserie chickens.


Forget about Costco, if some people here are so convinced this behavior is illegal they should be going after every fast food company that offers anything like "get a free/cheap xyz with any drink purchase!" Where the subsidy is obvious.


Costco gets to sidestep a lot of regulations because they technically are a private club with paid membership. The US anti-monopoly laws are also unusually weak.

In other countries, selling a $7 chicken if it's subsidized by the sale of other goods can indeed be illegal.


Do you have some countries in mind where that's illegal?


In Germany, selling goods for less then the one bought them for can be illegal if its used to push competition out on a large scale.


That would most likely be illegal in Finland. You're not allowed predatory pricing. And the same is true for the EU as a whole, although you may have to be operating in an international market, not just a local one. See Abuse of dominance in: https://en.wikipedia.org/wiki/Article_102_of_the_Treaty_on_t...


The US, the Sherman Antitrust Act prohibits predatory pricing.


So if I could charge less for chicken because I use that as a doorbuster to get you inside the store, but I charge more for other items, that is predatory pricing and against US law?


If you now acknowledge there is a carve out… couldn’t there be two carve outs? Or more?

It seems like you havent thought this through at all.


Some people genuinely believe the european copyright system (and La Liga and the Spanish judiciary) has more than 0% legitimacy… is it truly that hard to imagine?


Collective punishment is such overreach that it's a violation of the Geneva conventions. You do that and you no longer have more than 0% legitimacy.


I meant even after the fact they still believe to some degree of legitimacy.


Believing that an action is legitimate when it isn't simply means that they're in error.


How is imposing your worldview on them any better… than them imposing their worldviews on possible discount buyers of those unsold products?

You haven’t actually written the argument yet.


I would say Catalina in 2019 already had enormous issues, there were hard faults in Console pretty much daily that Apple never bothered to fix. (Plus hundreds of minor faults per day)

I had to downgrade to Mojave so the wheels likely came off internally around then.


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

Search: