fweespeech

> Wow, I wonder if it's location based. Mine was heavily situated in one location, whereas yours was perhaps global? So it was always prime time for someone? Here is a graph of my usage for the past week (note lows of 30-40 and highs of 650 odd):

A) I'm not sure "one location" is a city, country, or province.

B) Yes, it was global as we had offices from China to the EU.

C) Given the OP is handling 15 r/s, 150 r/s is not going to be pushing 500 mbps unless something is very poorly designed.

D) Even at spikes of 40x, it would still only be 600 r/s which is still quite doable with a commodity server.

The only serious difference is it was very large batches of requests (20 * SKUs) so people were fine with waiting a few minutes.

Fair enough. xD The law isn't something I'm well versed in. It just seemed odd that loophole existed in Japan.

Except in this case, its almost certainly a violation of money laundering laws in the US.

Tbh, part of me wonders if it would be perfectly legal if they sold an actual product at a massive markup (say, "designer chocolate"), provided "bonus tokens", and allowed refunds of the chocolate even if it was purchased with those tokens.

I'm not aware of any laws that prevent the conversion of tokens to chocolate to refunds. Then again, a cryptocurrency is probably less hassle at this point than any of these fronts.

https://bitcasino.io/

And I guess I'm not the only one to conclude that.

> What is the most complete solution you are aware of that compresses & optimizes png, jpeg, and webp and can be operated on a server? It should not only be able to optimize as part of the build process but also in response to user-generated content.

Tbh the UGC side is just triggering the "build process side" as the upload occurs.

As far as best,

https://github.com/MediaCrush/MediaCrush/blob/b32e2752d1a19d...

I'd suggest you look there for some decent examples of how to go about it. They may be defunct but I use a similar approach (slightly different knob tweaks with the same binaries) and it works fine. May not be 100% optimal but its good enough imo.

Its too application centric to be honest to give good generics unless you've built something similar before.

> As a concrete example, say you have an ecommerce app. Read requests take 80ms while write takes 200ms.

An ecommerce site can be nearly 100% cached except for cart/session/search data. This gives it a massive advantage over everything else (since you can load the non-cached data via js and its relatively tiny).

I think the last time I benchmarked one of our front end servers it was ~500 r/s on 2 cores for our search page.

You really aren't going to know until you try it and benchmark it yourself tho.

> We're all assuming content comes in constantly, but that's not the case. I imagine they spike heavily at peak times, perhaps 5-6 fold (judging by my own site statistics). Serverless might be good for them because they can handle small loads with small costs and large loads with no effort.

No. I've pushed 1000 r/s (approx) on a price tracking tool at work for keeping track of competitors on one of those boxes. Their largest spike (even if you assume its 10x their average) would be at most 15% of that.

Using AWS Lambada created that cost structure. Using an alternative to AWS would not.

$150 (at most) dedicated server for instance.

Tbh, its only worth considering for pure "credit card" services where its essentially read-only already and any damage can be attributed to fraud you can perform a chargeback against.

That being said, if you run all your spending through your credit cards except obvious notables you update manually you are golden. (i.e. Rent/Mortgage)

Thanks. :)

> That works out at 2.95 years of computation for $370.

It is 2.95 years of computation on less than 1 CPU core.

http://docs.aws.amazon.com/lambda/latest/dg/lambda-introduct...

> AWS Lambda allocates CPU power proportional to the memory by using the same ratio as a general purpose Amazon EC2 instance type, such as an M3 type. For example, if you allocate 256 MB memory, your Lambda function will receive twice the CPU share than if you allocated only 128 MB.

https://aws.amazon.com/ec2/instance-types/

> m3.large 2 7.5 1 x 32

> Each vCPU is a hyperthread of an Intel Xeon core except for T2 and m3.medium.

1 Core per 7.5GB RAM. Or ~1/28th of 1 core.

(2.95*365.25)/28 = ~39 Core Days.

That is really only impressive when you are comparing AWS prices to AWS prices which is a trap people seem to frequently fall into imo.

I wouldn't call it a low price.

Maybe I'm just spoiled because I can put crawl/parse tasks in a work queue for each url without much concern for latency but I've gotten 1000 r/s (on async calls) out of $100/€100 dedicated servers.

Their setup really only makes sense for WAN facing public architecture that can spike because you have no control over the load. Otherwise, I'd suggest you just get some leased dedicated server(s) to meet your workload.

Is archiving a bunch of porn really safe legally? I always had the impression archiving a large number of copyrighted images wasn't.

I'm curious, have you ever been mailed a rotting dead rat in a plastic bag before?

Some people have legitimate reasons to not want to be found and the police aren't going to do much with that kind of harassment either.

Similarly, if there is legitimate evidence a "crime" was committed via the website they can seize the domain or the registrar can. "Anonymous DNS" information doesn't get you very far if there is a legitimate court case.

> A company which has is CCBill. They sign you up as a "high risk" Visa card merchant (which is an upstream notion supported by Visa), and then are much more hands on in the checkout process than a normal credit card processor, with a website people can use to look up their subscriptions to websites (so like, imagine if all Stripe purchases required you to use their web form and then showed up on the customer statement as only "Stripe", and if the customer could then go directly to stripe.com, enter their credit card details, see they had a subscription with you, and cancel it; to be explicit: PayPal does not have these restrictions, even though they might seem similar to you).

> (To note: it does bother me somewhat that PayPal doesn't go out of their way to support some of this stuff, but as long as the actual credit card processing firms support it and there exist banks willing to allow it, I don't get pissed off about it as it is totally possible for you to do it yourself... it isn't like you are locked out of the actual network effect of credit cards, which is what is important; and given that there even exist easy well-supported full solutions out there like CCBill, it barely seems like a problem.)

http://pandorablake.com/blog/2013/05/censored-by-ccbill/

> CCBill noted, "It is a violation of CCBill's AUP to reference 'force' in this context as it implies a non-consensual situation (fantasy or non-fantasy). In order to be compliant with CCBill's AUP, please ensure all references to forced acts are removed."

http://motherboard.vice.com/read/the-secret-censorship-of-on...

> The posting or display of any image or wording depicting or related to extreme violence, incest, snuff, scat or the elimination of any bodily waste on another person, mutilation, or rape anywhere on the site in a sexual or erotic manner, including the URL and meta tags.

The censorship guidelines they are applying are literally those of CCBill as per their terms of service:

https://www.ccbill.com/cs/client/policies/ccbill/acceptable_...

You can read that there. Its in section 3.

---

> Do you have a large percentage of people who legitimately buy stuff from your website later claim they didn't?

Do you not understand that chargeback rates are a calculable, actionable metric? Or was my reference to the fact I thought worked in comparable businesses not an obvious enough hint that I've worked for both physical goods and ummm digital products of an intimate nature?

There is no real legitimate reason for their behavior not being modeled entirely on actionable metrics visible to both parties. Worse, they know and anyone who has dealt with the credit card processing on the merchant end knows it. Its purely, at this point, little more than a political issue being wielded where laws are unpopular.

> This is not your standard fraud model: with porn you see a remarkably large amount of "oh shit, someone (usually my significant other with whom I share my finances) just saw this charge on my card, which is totally not OK for whatever reason, so I guess my best strategy here is to totally disavow the charge and pretend it was credit card fraud".

You think people don't do this shit with lingerie, sex toys, and anything else along those lines that might indicate an affair. K.

Every other non-credit-card-company pulls the products I've worked with based on return/chargeback metrics that discriminate this problem with perfect ease and shove the full cost on to me. So do credit card companies by the way for anything that don't deem "immoral" in case you were wondering.

The metrics exist, they simply do not use them for this purpose when they target things in this manner.

---

> A second issue porn has which most websites do not is related: customers actively do not want their statement to accurately state from whom they purchased the product, lest someone sees exactly what their fetish is simply from reading the credit card report. This drastically increases the chargeback rate from people who legitimately do not remember what they bought (particularly if the service uses subscriptions).

Yeah. that isn't just a porn thing. Its anything potentially sexual, including lingerie. Please let me know when the credit card companies start banning places like Victoria Secret. The difference here is lingerie is too "mainstream" for that to be a viable target. And if you think companies like Amazon doesn't pull lingerie for high rates of returns, claims of "defects", and/or chargebacks, you've never tried to sell the stuff. This stuff has an operational cost but for purely digital goods (porn) its quite a bit cheaper as well.

http://www.bedroomjoys.com/discreet-billing-shipping/

> For your privacy and protection when you place your order with BedroomJoys.com your credit card will be charged by Ryhma LLC and your order will be displayed on your credit card statement as Ryhma LLC.

https://www.girlielingerie.com/help-faq

> We do not include our name on packaging and/or shipping labels, discreet plain package displaying only “GL” and no reference to lingerie anywhere. Discreet billing displaying only “GL” and no reference to lingerie anywhere.

---

> If you want to accept payments for porn, you need a fundamentally different fraud model than for a standard website, no matter how weirdly digital or in high demand: this is not about scammers or about black card fraud, this is about basic lies and deceipt. I see no real reason to claim PayPal must go out of their way to support this.

I'm talking about merchant accounts with a real bank and/or dealing directly with credit card companies. The banks aren't the ones that place pressure on these people. Its Visa and Mastercard and so forth. Fetlife wasn't operating through Paypal but through something like you suggest and has used CCBill in the past and the consent/non-consent definition is part of how CCBill operates.

---

Yeah.

FetLife acts more like a porn/escort site than a social network in many respects. Its unfortunate since a sex positive kink community would be a great boon but they all seem to get seedy when they allow RL nudity/discussions.

I wonder if a no RL rule outside of things like clubs, etc. that are larger organized group activities with physical presences and business licenses would have been the "real" solution.

If that logic was the case, they'd simply have a fraud metric that if you exceeded you'd have your account disabled.

As someone who works at a company where we literally have thieves testing cards on our site on a daily basis, any site that accepts credit cards has a high rate of frauds as thieves test cards. Somehow, we manage not to pass these cards to credit card processors (and no, I don't mean we use Stripe. I mean we have our own in-house anti-fraud process since we deal with the processors through a merchant account at a major bank like FL does). Porn companies have filtering in place as well for the same reason.

It would be a solid metric instead of "morality" if the cause was what you believe it to be.

The "fair price" is what the current owner is willing to sell it for and not the price set an auction where only Zuckerberg and the original owner would be willing to bid.

This is basically employing eminent domain to strength Zuckerberg's little fiefdom.

> In my city, the homicide rate is still 5X higher than it was in 1950s. That's hundreds of people dead each year, unnecessarily. And that's after dramatic improvements in medical technology! I have a lot higher standards than a 50% decline since the 1990 peak. Let's get it back down to the 1950s level.

Yet you repeatedly, blatantly cherry pick data that is just more lies of omission like this one where you try to defend yourself. Now you move the goal posts and show complete ignorance of the way population densities have changed since the 1950s and the relationship between such density and crime rates.

> I am not arguing about the crime changes over the past 20 years. The original article was about crime changes in the past few years, based on changes to policy in the last few years. I think the changes to crime policy implemented twenty years ago were good, and good impacts on crime. But the changes in the past few years are bad, and will result in an increase in crime.

Nothing substantial has changed unless you are truly and genuinely ignorant of history. You think BLM is new?

The LA riots and numerous similar events have been caused by killing unarmed civilians by police, regardless of justification. Stop being a sheep and actually read the history of the past 50 years. BLM and other protests are pretty fucking peaceful compared to the LA riots and other events in the past with similar causes. Get back to me when these "riots" and "policy changes" involved 10+ deaths on a regular basis.

The only people claiming it is getting worse are salesman peddling fear and ignorance. Stop being so easily persuaded and actually educate yourself.

> And homicide rates have risen substantially in the last two years (up 10.4% in 2015, and projected for 13.1% this year). These are the biggest increases in a generation.

Pretending a 49% drop over 20 years is sustainable continuously with no outliers is...absurd. There is a reason these sorts of trendlines are generally drawn over 5-10 years of data, not 1 or 2.

Please, just stop pushing your agenda on HN.

https://mises.org/blog/fbi-us-homicide-rate-51-year-low

> Homicide rates were considerably higher in the United States during the 1970s, 80s, and 90s, but over the past 25 years, have fallen nearly continuously:

> As Pew has reported in recent years, in fact, the American public is "unaware" that the homicide rate in the United States has fallen by 49 percent over the past twenty years. And while Pew doesn't report on it, it's also a safe bet that the public is also unaware that homicide rates have collapsed as total gun ownership in the United States has increased significantly.

Never turn it down.

If nothing else, you can always come back with "I have outside projects and I'll need a blanket exemption for projects that I started on before working at Amazon."

I don't think you understand "substantial" in that context.

The fact there is a consistent 50+ year trendline of compensation growth is simply caused by a combination of productivity gains and inflation.

You'd see a 5%+ jump in a single year if a legitimate shortage existed.

https://www.bls.gov/lpc/prodybar.htm

2007-2015 : +1.3% (nonfarm productivity gains)

2007-2015 : +1.8% (manufacturing productivity gains)

2007-2015 : +1.2% (total compensation)

Notice how similar these numbers are? That is a sign of normal compensation growth.

> I don't know what's a fair price in every area of the country for all of these jobs, so beats me if $12 is good or bad, but beyond this one anecdote, people have been saying for years that there was a shortage of skilled labor in the country. So aside from whether this one guy is paying fairly or not, it does appear to be a real issue.

If I say something and I make $N per year for believing that.

Why would they stop saying it?

The truth is, the "shortage" is largely in the poor quality of hiring screens and and a desire to pay the real market rate (rather than the lowest # they can hire the bare minimum of workers to function and work them 50 hours a week).

If a shortage genuinely existed, you would see substantial wage growth.

We don't have such wage growth.

-------

Similarly, when labor organizations get too powerful you see the opposite effect that creates an artificial shortage:

https://members.aamc.org/eweb/upload/The%20Complexities%20of...

> The rapid expansion in medical education was halted twenty years after it started as researchers, policymakers, and the educational community reassessed the need for further growth. This change recognized that the doubling of medical school graduates would lead to a growing supply for at least 35 years. In 1980, the Graduate Medical Education National Advisory Committee (GMENAC) reported (based upon the number of physicians needed to provide “necessary and appropriate” services) that a surplus of 70,000 physicians would be likely by the year 2000.20 With the promotion of tightly controlled managed care in the early 1990s, many groups reaffirmed the belief that the nation was facing a surplus. The national Council on Graduate Medical Education (COGME),21 the National Academy of Science’s Institute of Medicine (IOM), the Pew Health Professions Commission,22 the AAMC,23 the AMA24 and other national physician associations expressed concerns about an impending potential surplus of physicians.

> The medical education and training community responded to these recommendations. The number of graduates from U.S. medical schools remained virtually unchanged between 1980 and 2005. Despite this stagnation in medical school enrollment, the doubling of medical school enrollment during the 1960s and 1970s led to a steady stream of new physicians entering the workforce at a rate greater than those leaving the field. Consequently, the ratio of physicians to population increased steadily from 1980 onward. However, few health policy experts in the year 2000 would have argued that this constituted a surplus of physicians, even though this number grew from 202 per 100,000 in 1980 to 276 per 100,000 in 2000. 29 Bureau of Health Professions projections had estimated that physician demand - not a defined need - would increase over time. In fact, per capita utilization of medical services did rise as they had suggested it would.30 Other analyses had assumed major cost and utilization controls in health care.31

https://www.aamc.org/newsroom/newsreleases/458074/2016_workf...

> Washington, D.C., April 5, 2016—Under every combination of scenarios modeled, the United States will face a shortage of physicians over the next decade, according to a physician workforce report released today by the AAMC (Association of American Medical Colleges). The projections show a shortage ranging between 61,700 and 94,700, with a significant shortage showing among many surgical specialties.

-------

> I do think the purely political forum that a minority of users here seem to want, would best be served by a different site.

Privacy issues, opinions on hiring practices, etc. are all topics that have heavy polarized sides with little to bridge the divide and would likely benefit from a shift in separating those topics from the solidly tech related ones.

Although, to be honest, I'm more concerned with the fact you are oblivious to this than I am about the fact you call change ignorance. I guess it does not really matter, I'll be off HN for a long while.

Yeah, it honestly seems like HN should split into two forums effectively.

Tech (No politics, etc.) / Not Tech.

The trying to force it all into a single view seems to be creating some friction.

If you find it interesting:

https://www.sec.gov/news/testimony/2011/ts120111rsk.htm

> There is no express statutory definition of the offense of insider trading in securities.3 The SEC prosecutes insider trading under the general antifraud provisions of the Federal securities laws, most commonly Section 10(b) of the Securities Exchange Act of 1934 (“Exchange Act”) and Rule 10b-5, a broad anti-fraud rule promulgated by the SEC under Section 10(b). Section 10(b) declares it unlawful “[t]o use or employ, in connection with the purchase or sale of any security . . . any manipulative or deceptive device or contrivance in contravention of such rules and regulations as the Commission may prescribe as necessary or appropriate in the public interest or for the protection of investors.”4 Rule 10b-5 broadly prohibits fraud and deception in connection with the purchase and sale of securities. As the Supreme Court has stated, “Section 10(b) and Rule 10b-5 prohibit all fraudulent schemes in connection with the purchase or sale of securities, whether the artifices employed involve a garden type variety of fraud, or present a unique form of deception,” because “[n]ovel or atypical methods should not provide immunity from the securities laws.”5

Its very, very broad anti-fraud laws.

> See here for one example [0]. The SEC does have rules against it, so it's not entirely on a case-by-case basis [1]. It's just the SEC isn't supposed to write law.

It isn't really.

https://www.sec.gov/news/testimony/2011/ts120111rsk.htm

> There is no express statutory definition of the offense of insider trading in securities.3 The SEC prosecutes insider trading under the general antifraud provisions of the Federal securities laws, most commonly Section 10(b) of the Securities Exchange Act of 1934 (“Exchange Act”) and Rule 10b-5, a broad anti-fraud rule promulgated by the SEC under Section 10(b). Section 10(b) declares it unlawful “[t]o use or employ, in connection with the purchase or sale of any security . . . any manipulative or deceptive device or contrivance in contravention of such rules and regulations as the Commission may prescribe as necessary or appropriate in the public interest or for the protection of investors.”4 Rule 10b-5 broadly prohibits fraud and deception in connection with the purchase and sale of securities. As the Supreme Court has stated, “Section 10(b) and Rule 10b-5 prohibit all fraudulent schemes in connection with the purchase or sale of securities, whether the artifices employed involve a garden type variety of fraud, or present a unique form of deception,” because “[n]ovel or atypical methods should not provide immunity from the securities laws.”5

Congress wrote some very broad anti-fraud laws, never amended them, and left it up to the SEC to exercise its discretion on what qualified as such practices.

The failover guarantees with RDS Aurora are substantially better as long as you have 1+ read replicas.

That alone is worth the difference imo.

> Would anyone please tell me which of Linode, DigitalOcean and Vultr have cost ceilings? I looked at their pricing pages but couldn't figure out. They all claim that they have monthly billing caps for the hourly rates, but meanwhile, both DO and Vultr have per-GB charges if the transfer quota are exceeded, and Linode is silent on this on its pricing or FAQ pages. Can the data transfer charges be capped too? If so, what happens when the quota is reached?

They don't for traffic. So you do run a small risk of something happening.

However, Linode at least pools your VPSs so if you have 100 of them and 20 of them "go over" the cap you still are often okay because of the other 80 that didn't "go over".

The truth is none of these providers provide truly hard caps. The difference is with Amazon/Google/Azure/etc you can realistically get hit with a 4 figure bill if something goes seriously wrong.

DO/Linode/Vultr I've never seen accidental "mistakes" causing that sort of thing and even an active dos/ddos attack that would cost you more than $100 in overages before they started null routing you.

Billing alerts that are simply a notification you exceeded $X is not any sort of limit, particularly when they may take hours to arrive.

>2) Yeah. Trump and Russia are doing it, too. It's just as nefarious there. Why is the standard defense of Clinton whataboutism?

If I stab you and you try to defend yourself, it is whataboutism now?

Politics is about offense and defense. You can't fight a guy with bot networks pushing a false narrative without one of your own.

You honestly think unilateral disarmament is the way to go?