Hacker News new | past | comments | ask | show | jobs | submit login
Prominent D.C. media firm implicated in fake FCC comments (gizmodo.com)
282 points by spac 27 days ago | hide | past | web | favorite | 72 comments

Sickening how easy it was for the telcoms to push their agenda through in spite of overwhelming public opposition.

I'm pretty sure most individuals following the story of net neutrality knew the fcc's ballot box was compromised.

It would be nice if someone could face consequences for the fraud put on by the FCC. I won't hold my breath though.

The thing is, "the ballot box" you refer to was just a way for random people on the internet to submit comments. It isn't intended, I think, as a way to guide policy. I expect that they switched to web-based commentary because, historically, they've accepted letters and the web-forms are just an "upgrade". Who's going to hire staff to open, read and respond to letters these days?

These fake "grassroots" operatives merely abused the web-commentary system in order to keep the FCC leadership (and Republicans) from being completely humiliated by the sheer number of people in favor of net neutrality. Turns out, they got sloppy and went a bit overboard.

What is really disturbing, however, is that the company named, "CQ" is in the business of automating "campaigns" with software and data analysis. To them, it doesn't matter WHAT the campaign actually is as long as they get a paying customer-- smears and disinformation is just another day of work for them.

There's so many layers of slime here, it's hard to know where to start.

> It isn't intended, I think, as a way to guide policy.

Public comment is explicitly intended to guide policy. It’s not a vote but the entire point of the process is to get feedback before a policy is finalized and most agencies attempt to address common points.

It’s reasonable to question the inherent assumption of good faith with the current administration but we’re also in an unusual time where the most cynical view of previous administrations are now the smart money view of how things work.

Public comment is meant to surface arguments to be taken into account to guide policy. As such, having 1 or 10 million people make the exact same argument in the same words doesn't change anything.

This is only true in the most simple cases. If there’s any portion of a proposed policy which is balancing different interests - which is almost always the case - the ratio of comments is a useful metric because it would surface problems such as e.g. a policy which big companies are comfortable with but hit many small ones with substantial compliance overhead.

In this case, having tons of AstroTurf comments allows them to say that there’s public support outside of large ISPs. Given how anticompetitive the decision was, they needed that fig leaf.

No agency would be on firm footing pointing to a large number of identical form comments received for one side as a factor in a balancing of interests. It would not survive the “reasoned decision making” requirement. To the extent you’re talking about balancing the needs of small and large companies, for example, you’d have to point to concrete examples and data. That could come in comments (small businesses talking about real challenges they’ve faced), but relying on a large number of identical form comments wouldn’t pass muster.

If you assume good faith, yes. In the environment they were in where they knew that oversight was in friendly hands, however, I think they were just looking for enough to say the process was followed knowing that it’d be done before it’d get any critical examination.

Exactly. This is why the continued focus on "fake comments" in this process is so tiring... They didn't mean anything to begin with. Just as the comments opposing net neutrality didn't matter either. The FCC does not pass regulations by open voting and never did.

If the fake comments didn't mean anything, why make them?

That's a very interesting question.

I suspect this incident demonstrates the scope of disinformation campaigns and the industry that enables them.

It is totally possible that cramming a couple million fake comments into the FCC was done solely for the purpose providing a politician a "talking point" in favor of getting rid of net neutrality.

When one dollar = one vote, I think it’s pretty obvious where this process ends.

This one always cracked me up:

The unprecedented regulatory power the Obama Administration imposed on the internet is smothering innovation, damaging the American economy and obstructing job creation. I urge the Federal Communications Commission to end the bureaucratic regulatory overreach of the internet known as Title II and restore the bipartisan light-touch regulatory consensus that enabled the internet to flourish for more than 20 years. The plan currently under consideration at the FCC to repeal Obama's Title II power grab is a positive step forward and will help to promote a truly free and open internet for everyone.

Barack Obama

1600 Pennsylvania Ave NW

Washington, DC

With so many rotten apples, maybe the tree should go. This is a gigantic, criminal subversion of Democracy that the FCC remains complicit in, something that only jail time can fix.


> Duby is one of the 24 people who signed the open letter in May 2017 demanding that her comment be removed by the FCC. Of 14 others who said their names were “used to file comments we did not make,” Gizmodo was able to duplicate the experiment 12 times. In each successful case, the comments were received by the FCC while CQ’s API key was in use, with the logs reflecting deviations in the timestamps roughly equivalent to the blink of an eye. (For reasons unclear, two of the signatories’ comments could not be located.)

This is alleging that CQ impersonated people to file comments with the FCC. I don't know if this meets the legal requirements to be considered some kind of fraud, but it sure seems like it should.

It isn't just that CQ stuffs the ballot box with false statements, it looks like the FCC is complicit in hiding it as well, then Ajit goes on record saying it is the Russians. How much lying can they do? That was rhetorical, I believe there is no upper bound.

The State of NY and the FBI cannot comment because of ongoing investigations... How about the FBI interviews Ajit Pai, who can't not lie, and they throw him in jail just like Martha Stewart.

What CQ did and what the FCC is help hide is felony impersonation in most states. Each occurrence (sounds like there was at least 8k from the WSJ investigation) could be a felony.

What exactly are you saying is a lie?

Exactly what he said was a lie.

The only thing he mentions is the Russian addresses, which I haven't seen anybody disputing. Do you think there weren't that amount of comments with Russian addresses?

Nobody's disputing it except for you.

> What exactly are you saying is a lie?

> I'd like you to point to a single lie, since you're asserting that he lied without pointing to any.

Again, what specifically is a lie and why do you think it's a lie?

Sitkack's comment was specific. Read it.

The "they" is ambiguous, and the statement it's referring to is ambiguous as well. That's why I asked for clarification multiple times, the refusal to do so is incredibly suspicious.

How about we enumerate the truth, a much more tractable endeavor.

You can make as many leaf comments as your karma allows but isn’t going to change anything.

I'd like you to point to a single lie, since you're asserting that he lied without pointing to any.

No, it's alleging that CQ submitted those comments. There's no reason to believe based on this article that CQ impersonated people - what's more likely is that they operated websites to gather comments and people submitted false names through there.

If you look at the linked WSJ article you'll see there were fake comments on many of the common templates, some associated with pro-NN groups and some against.

The investigation focuses on a specific anti-NN comment that was submitted, identically worded, hundreds of thousands of times, in alphabetical order by name. That's not "people submitting false names" through a website, it's obviously organized fraud.

If you're suggesting that CQ was just the unwitting middleman between the FCC and some other group that was submitting the fraudulent comments through their site, that would have been trivially detectable and filtered out by CQ:

> “Before we submit these comments (via the API) we remove any bad or questionable submissions,” he told Gizmodo. “On a technical level, a few of the things we do include running the email address through an email validator, eliminate duplicate records with the same email address, and remove multiple submissions from the same IP address.”

>The investigation focuses on a specific anti-NN comment that was submitted, identically worded, hundreds of thousands of times, in alphabetical order by name.

No, it's focusing on many different comments, including some from pro-NN groups and some from anti-NN groups. 14 groups in total were subpoenaed.

The WSJ article linked (can read it at http://archive.is/sp9Q7) says 4,622 people polled said they hadn't left a particular comment that was pro-NN, which is the majority of the 7,741 total responses that claimed they didn't leave the comment that bears their name.

>If you're suggesting that CQ was just the unwitting middleman between the FCC and some other group that was submitting the fraudulent comments through their site, that would have been trivially detectable and filtered out by CQ:

It's trivial for any teen on 4chan to defeat those measures, come on. Pretty much everything that's out there could in theory have been done by a single 14-year old for laughs. Connect to websites that are collecting names without extensive validation, use a few IPs, etc.

What's more likely - that 14 different groups, some on the pro-NN side and some on the anti-side, all chose to commit crimes by faking names and addresses, or that one bad actor connected to a few of them to troll?

I find it interesting that you keep referring to "the WSJ article linked" -- which now that you've actually linked to it turns out to be a 2017 article that as far as I can tell you're the only one who's linking to -- rather than the current article on Gizmodo we're actually discussing here, which describes an investigation that is in fact focused on the specific anti-NN comment submitted by (or through) CQ.

Even that WSJ article focuses largely on that specific blatant example, and notes that the several anti-NN statements they examined were "63%, 72% and 80% bogus comments" (the single pro-NN statement they examined "was 32% bogus.")

Astroturf exists and is a problem on all sides, yes, but this specific instance is far beyond the pale. Which is presumably why it merits this year-long investigation.

> It's trivial for any teen on 4chan to defeat those measures

That's... debatable. But also beside the point. CQ claims they vetted their submissions for "questionable submissions". If that vetting failed to notice that an identical comment was submitted through them that many times, with names in alphabetical order, then they either are lying about vetting their submissions for accuracy, or they are themselves the ones committing the fraud.

The article is linked from the OP here.

OP is narrowly focused on a particular one out of fourteen companies under investigation for unclear reasons. They don't say whether they've tried similar analysis of any of the other form comments. If I had to make a wild guess, I'd say they expect more clicks here than with a more in depth investigation of everyone involved, for reasons left as an exercise to the reader.

>Astroturf exists and is a problem on all sides, yes, but this specific instance is far beyond the pale. Which is presumably why it merits this year-long investigation.

14 organizations are under investigation, not just one.

>But also beside the point. CQ claims they vetted their submissions for "questionable submissions".

And also gives a few examples, all of which are trivial to bypass, and all of which are automated, so it's not clear why you'd think they are manually looking at any of these before submitting unless it fails the automated tests.

> 14 organizations are under investigation, not just one.

I did miss that, fair enough -- amend my statement to "Which is presumably why it merits this year-long investigation of these 14 organizations".

> it's not clear why you'd think they are manually looking at any of these

It's not clear why you'd think 818,000 (per WSJ) identical comments submitted in alphabetical order should pass automated tests?

>It's not clear why you'd think 818,000 (per WSJ) identical comments submitted in alphabetical order should pass automated tests?

There's no reason to think they were submitted originally in alphabetical order. The most likely possibility here is that CQ sorted them before submitting to the FCC.

It occurs to me that it's entirely possible CQ would sort the comments by alphabetical order of the commenter before submitting them in bulk, still not sure where it's from but that seems a likely explanation.

The submissions are individually timestamped, not pre-sorted and submitted in bulk.

Also, where are you getting the claim that it was submitted in alphabetical order from? It's not in OP.

It was widely reported at the time; here are a couple of cites, several of which were also linked in the OP: https://www.techdirt.com/articles/20170510/08191137334/bot-i... https://www.forbes.com/sites/janetwburns/2017/12/14/earth-to... https://www.zdnet.com/article/a-bot-is-flooding-the-fccs-web...

(Also, FWIW I personally witnessed it while trying to submit my own comment on the issue; the viewable posted comments were about 98% duplicates of that one "The unprecedented regulatory power the Obama Administration imposed on the internet" etc, and the names were somewhere in the E's.)

And do you have any reason to think they were submitted to CQ in alphabetical order, as opposed to the obvious possibility that CQ or their client sorted them prior to submission to the FCC?

They're timestamped.

Those timestamps are the time the FCC received them. If you read through the email chain you'll see they had many comments ready (the email says he was expecting to send over 250k comments/day) prior to submitting any or getting the API access set up.

This wasn't real-time submission, and this should be clear from OP.

If you read through the email chain you'll see that it occurred before the docket even was made public to be commented on: "The docket will be on internet privacy, it is not yet published. We believe it will be published later this month"; if they already had prewritten comments ready at that time, that confirms fraud by CQ, because the public can't write comments on something that doesn't yet exist.

Meanwhile CQ is repeatedly insistent that the submissions be recorded individually, and not by the bulk csv method; and when the API limits turn out to be too strict they go to great effort to do it anyway: "We will need to set up multiple servers to feed the API simultaneously to meet the delivery needs of our clients" -- 114 api keys compared to one or two for most other groups.

None of which sits well with your idea that they'd intentionally sort them first. Bulk submissions might be expected to be sorted. The assumption for individually submitted comments is that they're individually submitted; it exposes the astroturfiness very clearly when all the identical comments show up in strict order in the docket. (Or at the very least, if this was just a dumb goofup or excessive tidiness, why wouldn't they have simply explained that as the obvious excuse the moment it started being reported? There'd be plenty of supporting evidence, the stage space alone would have had to have been significant)

What's the difference between the bulk csv method and individual submissions on the FCC end? It's not clear from the emails.

Why hasn't Trump launched a high profile investigation into this? This guy was appointed during Obama's presidency and he could score some easy points.

I don't know if you're just misunderstanding things or wilfully spreading misinformation, but just in case:

1. Ajit Pai started working at the FCC under George W. Bush.

2. Ajit Pai was nominated under Obama for a republican seat at the recommendation of Mitch McConnell: no more than 3 FCC commissioners can be of the same party so the split is generally[0] 3 of the president's party and 2 of the opposition, Ajit Pai was a GOP pick.

3. Ajit Pai was made chairman of the FCC by Donald Trump.

Ajit Pai is a thoroughly Republican operative. Getting rid of Pai would not score any points with the republican party (he's doing exactly what he's supposed to do there), nor would it score any points with the democratic party (way too little, way too late, no chance of the damage done being repaired, and what happened was exactly what'd been predicted when he was made chairman).

[0] when that comes up / is possible: commissioners terms are 5 years (though they may serve up to 6½ if their replacement is not yet appointed) and the seats are staggered so one commissioner's term is up every year

He was McConnell’s nominee for one of the Republican positions during the Obama administration and was promoted to chairman early in the Trump administration. It’s safe to say that he’s not out of step with the GOP leadership when he has the backing of both key Republican leaders.

Moreover, Pai has resolutely stood for large companies being insulated from competition and being allowed to extract more money at the expense of the public. That’s Trump’s platform, not an aberration!

...not really? Trump appointed him as chairman, and Mitch McConnell picked him for the board in the first place:

> That’s not to say Obama shared Pai’s views—rather, the president was abiding by the convention that the minority party gets to nominate two appointees to the five-seat commission. http://fortune.com/2017/12/14/ajit-pai-net-neutrality-donald...

Who are prominent clients of CQ Roll Call?

Faking an identity online for political manipulation should be a felony, full stop.

That could end up outlawing Pseudonyms and fictional characters. The real problem is that the government would accept public comment through an unauthenticated web form.

Money quote:

an FCC spokesperson said comments from the general public are “generally not substantive, so thus have no impact on a rulemaking.”

There you have it. So the number of comments is useful only when it helps make whatever point they're trying to make.

Independent federal agencies are, by design not accountable to public opinion, except indirectly through the election of the President, who appoints commissioners. Agencies are designed to be guided by expert opinions and judgment. The purpose of APA comments, thefore, is not to poll the public. It is to ensure that the agency considers all the relevant “written data, views, or arguments.” 5 USC 553(c). Legally, a rule making cannot be challenged on the basis that it is contrary to the weight of public opinion. It can only be challenged on the basis that the order adopting the rule doesn’t adequately address the “data, views, and arguments” raised in the comments. https://digitalcommons.law.ou.edu/cgi/viewcontent.cgi?articl.... Accordingly, a large number of comments that raise the same views cannot have any impact in the rule making. The soundbite you quoted, therefore, is a completely unremarkable statement about regulatory law.

That also makes sense. Regulatory agencies micro-manage businesses to a greater degree than ordinary legislation, including managing prices. When you remove those things from the purview of the market and business judgment, you can’t subject it to the whims of the public. If you had a vote on it, electricity would be free and there would be no money to invest in power plants or the grid.

(Now, spamming an agency website with fake comments is a highly problematic action from a systems integrity and computer security point of view, no doubt. That’s a separate issue from the agency’s rule making policy.)

Everything you said makes sense and seems accurate, except for one important detail: while the agencies aren't required to consider or obey the weight of public opinion, on issues with significant public interest they ought to voluntarily consider public opinion (even if they choose to reject it).

After all, they serve the public despite the indirect accountability, and often there is more than one path which the experts deem viable. Public preference can be an appropriate in choosing among the viable paths.

In this situation, when the coordinated fake comments line up with the governing politicians' views, and when both run contrary to the weight of real comments from experts and non-experts alike, it casts doubt on the genuineness of the FCC's good faith in considering even the expert public comments, and therefore on the validity of the rulemaking overall.

Whether they ought to or not, regulatory agencies working on high profile actions do take public opinion into account. This happens via political appointees and the staff they hire, who are accountable to the politics that brought their administration into power. Ultimately, regulatory actions can become election issues.

The idea that regulatory agencies act solely on the basis of research and law is not believable, because you see regulatory agencies directly reversing rulemakings that were enacted only a few years earlier (not coincidentally, during the previous administration). It beggars belief that objective research and non-partisan legal theory has changed so much in just a few years (again: spanning a presidential election) that regulations have to be revisited and reversed by new rulemakings.

Look at it this way: if FCC rulemaking doesn't take public opinion into account, why did supporters of this rulemaking feel the need to stuff the comment box with a bunch of fake, identical, supportive comments? They should have had no reason to do that, since the rulemaking was in their favor and (supposedly) being done for an objectively right reason. But the truth is, they feared the politics of it and wanted to create a counter-story that blunted the appearance of how unpopular this rulemaking was.

I think we're agreeing and you probably meant to reply to the other person I was discussing with. If we're disagreeing, let me know and I'll reread.

Just adding on to/expanding upon your comment

Say you’re asked to head a team to build an operating system kernel. You will use your expert judgment, and that will include not just facts and data, but your philosophy. Agile or waterfall? Micro kernel or monolithic kernel? The decisions are not going to be a neutral function of the facts on the ground, but your overall philosophy. But you probably wouldn’t base those decisions on polling the general public!

Building an operating system kernel is not anything like setting public policy. Anyway, I'm not saying how things "should" be, I'm relating how regulatory policymaking actually works. Not that it's driven by public opinion, but that powerful participants are aware of its impact on public opinion.

Again: if that wasn't true, then why did people who supported the FCC rulemaking bother to pump in a ton of fake supportive comments? Why did people at the FCC try to cover up that that happened?

> Building an operating system kernel is not anything like setting public policy.

It absolutely is. You’re designing a complex and highly coupled system within a framework of rules and constraints. In the case of regulation, the rules and constraints come from economics, while in the case of operating systems they come from computer science. In both situations, there are no known optimal solutions—nobody has designed a mathematically optimal way regulate electricity prices, much less structure an electricity market, just like nobody has designed a mathematically optimal way to manage memory, much less structure a kernel. Thus, a lot of judgment and heuristics and trade offs are involved at every level, and the exercise of that judgment is going to depend heavily on personal philosophies. Margaret Thatcher approaches regulation very differently than Franklin Roosevelt, just as Linus Torvalds approaches operating system kernels very differently than Andrew Tanenbaum. If you hire Tanenbaum versus Torvalds to design you an OS kernel, you can expect a very different result, based on their expert judgment.

Unfortunately, a lot of people think of regulation as just picking from a menu of policy choices, and fail to realize that what you’re really doing is designing economic systems and markets.

> Unfortunately, a lot of people think of regulation as just picking from a menu of policy choices, and fail to realize that what you’re really doing is designing economic systems and markets.

A lot of these people vote and donate to candidates! Some even get elected to office. Therefore you cannot dismiss their impact on the regulatory process, even if you would like to.

Because it is ultimately the political process that determines which people--which personal philosophies--will have the resources and power to make regulations.

Agencies are required to exercise reasoned decision making. I’m not sure I’d agree that popularity is a “reasoned” basis for choosing one policy over another. Besides, the public does influence the agencies—by picking the President, who appoints Commissioners. Agencies are part of the executive. That’s who sets the broad regulatory agenda. I’m not sure it would be proper for agencies to end-run around that process and make policy based on public opinion directly.

Popularity isn't always a reasoned basis for choosing one policy over another, no. But sometimes the priorities of the public as expressed through comment can absolutely be an element of reasoned decision making, when the experts have validated multiple approaches as acceptable and the issue has as much direct impact on & interest for the public as net neutrality.

Phrased another way: if the experts say "doing A will work but have tradeoff B, while doing C will work but have tradeoff D", it's useful to know whether the country prefers B or D so that the commissioner can pick A or C accordingly.

The independence of agencies like the FCC is meant to be independence from any individual President, so that the commissioners are less responsive to the whims of the politicians than the regular executive departments. This is why there are commissioners from both major parties, why their terms of office are longer than a presidential term of office, and why they can only be removed from office for cause rather than merely for making a decision the President didn't like.

Your argument that they should defer to the wishes of the executive is exactly counter to the structure and purpose of how they're set up (even if I admit they're less independent in practice than in theory).

As a point of comparison, regular executive departments are headed by a single political appointee of the president's party, who customarily offers to resign when a new party enters the White House, and who can be fired with no notice for any reason or no reason.

> Phrased another way: if the experts say "doing A will work but have tradeoff B, while doing C will work but have tradeoff D", it's useful to know whether the country prefers B or D so that the commissioner can pick A or C accordingly.

I don’t see how that’s a reasoned basis for picking one versus the other. If a court did that, would you say that was a reasoned decision? (“We have these two competing interpretations of the 14th amendment. The majority of the public opposes interracial marriage, so we will pick the interpretation that allows states to ban it.”)

As to your other point, independent agencies are set up so that the President can’t micromanage specific policies. But the heads are nominated by the president precisely so the executive can set the overall regulatory agenda and philosophies. Remember the context in which the APA was created. The government regulated things like what routes freight trucks could run and the prices. You didn’t want direct executive control over those things. But broad agendas, such as ensuring more freight service to the Midwest? That was squarely within the administration’s power to influence the agency.

Courts do indeed reverse some of their rulings over time based on changing attitudes, sometimes with explicit reference to public opinion in their reasoning. So do regulatory agencies both independent and departmental.

The 14th Amendment example is about statutory/constitutional interpretation, whereas the type of ruling we're discussing here is not.

Agency regulations (and common law rulings by courts) often make policy judgments among multiple valid possibilities. Sometimes public attitudes appropriately affect the policy judgment, though of course not always.

Of course agencies and courts don't make popular opinion the primary driver of their rulings, even when they consider it. And I'm certainly not suggesting they should.

As for why the APA was created, that's a very complex question, but that seems more about systematizing the regulatory rulemaking process as delegated from Congress than about the difference between an independent agency and an executive department (both of which are governed by the APA). We're talking about the kind here that's supposedly independent.

Isn’t the FCC supposed to be a consumer protection agency? Seems silly to disregard consumer concerns.

The FCC was not intended to be what we consider a “consumer protection agency” (an agency that one-sidedly advocates for consumer interests, like say the CFPB).

The Communications Act of 1934 states that the FCC is created to regulate communications with the purpose of making available “rapid, efficient, nationwide, and worldwide wire and radio communication service with adequate facilities at reasonable charges, for the purpose of the national defense, and for the purpose of securing a more effective execution of this policy by centralizing authority theretofore granted by law to several agencies and by granting additional authority with respect to interstate and foreign commerce in wire and radio communication.”

If you look at the historical context of the FCC, including the “several agencies” that the FCC replaces, the focus was on coordinating resource allocation. Granting spectrum licenses, granting permits to build infrastructure, etc.

The AT&T monopoly exemplifies the sort of system the FCC was mandated to help create. It was a world class, gold-plated telephone network, available almost everywhere in the US. It wasn’t consumer friendly in many ways (it was expensive and restrictive), but it was certainly “rapid and efficient.”

The middle initial stands for Communications, not Consumer. It is theoretically meant to regulate shared communications resources, initially the radio waves but by extension also phones and the internet.

Paid for by.... Mr. Pai? Follow the money...

This is treason and I don’t see why it would be punished differently.

But...this isn't treason. Fraud, yes. Definitely fraud. But unless there is some other evidence that this fraud was committed with the specific intent of aiding a foreign power rather than just line the pockets of the telecoms, then no, not treason.

“the crime of betraying one's country, especially by attempting to kill the sovereign or overthrow the government.”

Maybe I’m being a little hyperbolic, it isn’t the government being overthrown but one branch of it and certainly what the public wanted was intentionally silenced.

You are being more than a little bit hyperbolic, and it's bad for the conversation when people do that.

Expansive interpretations of "Treason", such as the one you propose, are explicitly prohibited by the US Constitution due to their abuse by european monarchies to silence dissent. "Treason against the United States, shall consist only in levying War against them, or in adhering to their Enemies, giving them Aid and Comfort. No Person shall be convicted of Treason unless on the Testimony of two Witnesses to the same overt Act, or on Confession in open Court." Article 3 Section 3.

As a result of this strict standard, the last successful prosecution for treason in the USA was in 1952...

The Republican Party is the minority party and have most of the power in the US at every level.

By definition, most things that they do will be fairly unpopular, and some things they do will be wildly unpopular.

definitely a good look to call for death on a web page. good stuff here. and, as a sibiling commenter pointed out, this isn't treason kids, at all.

signed, someone who despises Ajit, the telecoms, and has followed FCC shenangins for years

Applications are open for YC Summer 2019

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