Palantir Announces Confidential Submission of Draft S-1 | Hacker News Reader

Palantir Announces Confidential Submission of Draft S-1

businesswire.com

195 pointsxoxoy194 comments

As someone who worked in defense consulting/IT contracting many moons ago, let me clue you in on a simple fact: there is an INSANE amount of money wasted on these contracts.

The project that we were working on was projected to maybe go live in ten years (when the tech would be laughably outdated), and we were the sub-sub-sub-contractor of the massive company at the top. It was nothing more than a vaporware webapp built by 22 year olds right out of college who were white/American enough to get clearance immediately, and our team got $100 million for it.

The money in these DoD contracts is truly nuts. And it never gets slashed because how politically terrible slashing defense is, and it never gets audited because too many people are making money.

So as it relates to Palantir, sure, they're scammy and unethical as all hell, but they certainly know where the bread is buttered in America. Even in a (likely) Democratic adminstration.

What’s the best path to get on this gravy train?

A master's degree. Body shops can bill more for contractors if they have one. This often means they make it a requirement.

https://beta.sam.gov/

https://www.usa.gov/become-government-contractor

Get a security clearance. Higher the better.

Military is a common way to get one, but plenty of civilian jobs will sponsor, too.

Clearances cost a bunch of money to do though, so they usually pre-screen heavily.

yes, democrats and republicans are very much alike (only their speech differs)

I understand somewhat people saying this in 2015.

I do not understand people who say it in 2020.

Maybe you don't understand anything... just wait and see when Biden gets the seat. (only look at actions, and not at what they say)

He/she means it in the sense of holy grail budgeting with respect to Defense (and Social Security, Medicare/Medicaid (Holy Trinity?)), no one would dare cut it regardless of party. It’s important to note since it’s one of the largest beneficiaries of tax money allocation.

Social Security spending is significantly larger than defense spending.

Thanks, you’re right. Still standing by my point, no one touches that stuff.

That's not true, though. The current administration is attempting to get the ACA struck down, which would boot millions off Medicaid.

does that include the black/secret budget?

Because populist politics has little to do with operational politics, and operational bureaucracy.

The Tweeter in Chief (and his followers) has absolutely zero interest in any of the actual mechanics of government, and almost everything material flies completely under the radar of false Twitter wars is 'same old'.

The dynamics of awarding contracts likely has not changed, and frankly is probably more regional (i.e. we want that factory or HQ in our county!) than ideological.

Edit: By 'followers' I mean literally 'Twitter' followers. Not supporters in general.

Who do I vote for if I want universal healthcare?

You can be as cynical as you want about whether they “really mean it”, but it’s right there in the Democratic party platform: https://democrats.org/where-we-stand/party-platform/ensure-t...

That doesn't say what you think it does. Additionally, the Democrats had a super majority for two years under Obama and ended up passing a healthcare bill written entirely by insurance lobbyists. The ACA skyrocketed health insurance costs, by every objective measure. They had 2 years already. I don't see how the exact same elected officials won't squander another super majority (which will be hard to get).

In my opinion, a D house and R senate is likely -- which means no universal healthcare.

> Additionally, the Democrats had a super majority for two years under Obama and ended up passing a healthcare bill written entirely by insurance lobbyists.

Super...ish. Loss of a single senator would've killed the bill.

Lieberman strangled the public option in its crib.

It took 3 seconds of skimming to see it does not remotely say that. It clearly says to have a public option and medicar for 55 plus only. This is a nudge of the status quo. Par for the course for Democrats.

Leftists, at the local level.

Democrats, at the national level, and repeatedly, if you want much hope of it in the next decade or two. It'll take some movement of the Overton window, and Democrats tend to vote in one election, then go "welp, I already did my part" two years later.

The Democrats. They've been trying for a while, are easily misled, and are generally a clusterfuck, but at least they're tring. Though there have been some Republican attempts -- attempt in the loosest sense -- believe it or not.

The Hil-Dog was talking about universal healthcare during her husband's (Slick Willy's) administration in the 90's. Didn't go anywhere -- not with Newt G in Congress -- but was a key item on her platform running against Trump. And most of the Dems went in on Obamacare, though several balked at the Public Option (I beleive it was one particular).

Mitt Romney kicked off Romneycare in MA, and Obamacare is essentially just a knock-off of that. Both were based off of a Heritage Foundation white paper about free-market universal healthcare (which Heritage has since disawovwed, as has Romney; see below). Unfortunte that Romney didn't embrace that while running against Obama, but that's probably the only way to score a GOP nomination.

https://en.wikipedia.org/wiki/Clinton_health_care_plan_of_19...

https://www.heritage.org/social-security/report/assuring-aff...

I don't know I feel like there's a lot of votes that differ too

> (likely) Democratic adminstration

I am not taking any sides here... but don't count your chickens until they hatch.

These comments are a useful tool for confirming the dominant bias of the HN community. It is frustrating to see smart people downvoting intelligent observations because they would be miserable if the observations came to pass. The downvote button should not be an indicator of your personal annoyance.

I'd argue HN shouldn't have a downvote button at all. I can understand when using it to downvote misinformation or comments that break guidelines, however, in discussions it often is used to quietly say "I don't like your opinion, period". Then again, not having a downvote button would probably lead to more flamewars but I'm just speculating.

It's a shame seeing Reddit-like behavior on HN (I understand comparing HN to Reddit is blaspheme).

Nearly 100% of the time I've seen "dead" comments, it's because they said something that goes against the grain. This feature is pretty much pure abuse by power users. Same for downvotes.

Agreed. The downvote should be used to indicate a comment is off topic or doesn’t add to the discussion, not whether it aligns with your opinion.

In this particular case, however, I’d agree it’s off topic (but I personally didn’t downvote it because it’s not that out of bounds).

> As someone who worked in defense consulting/IT contracting many moons ago, let me clue you in on a simple fact: there is an INSANE amount of money wasted on these contracts.

I've always wondered, considering massive amount of waste in Western military budgets, it makes me wonder what the real amount of money being invested compared to other military powers such as Russia and China.

Russia has its own "leaks"(money wise).

The Chinese military is historically beset by corruption and "leaks". The "Century of Humiliation" started when the Chinese navy was utterly destroyed by a newbie regional power Japan at the Battle of Yalu River partly because corruption had deprived the Chinese of enough ammunition for gunnery practices. Also, take any figures coming out of the current Chinese military with a grain of salt. The Chinese military budget is likely to be much, much bigger. Part of Xi's reform is to get rid of the fat in the Chinese military. It may be a while be China sees the fruits of those reforms.

But is there real reform? I know they're "killing chickens to scare monkeys" (杀鸡儆猴), but the incentives for corruption remain unchanged.

https://en.wikipedia.org/wiki/Kill_the_chicken_to_scare_the_...

Yeah I'm pretty skeptical as well but I've learned that just because someone is a competitor, we shouldn't deprive them of the virtues that we like to believe we have. It is very possible and I would argue probable that Xi is a genuine Chinese patriot (likewise I would say for Putin). Being a patriot, however, doesn't mean one can't also accept some amount of corruption if that furthers the long term goals. As people have already pointed out, even the US defense industry has some level of corruption too but no one can reasonably doubt the superiority of the US military. If we are frank with ourselves, the US is somewhat corrupt too yet it's always managed to cope with that.

In terms of force capabilities, those are much more difficult to hide. The amount of say, aircraft carrier groups is always going to be public knowledge, they're too big to hide. But in the world of nuclear ICBMs, the strongest nation states don't ever fight each other directly outside the smallest skirmishes. So the military for China and the US is mostly used to ensure compliance and power within their sphere of influence.

> it makes me wonder what the real amount of money being invested compared to other military powers such as Russia and China

Probably a lot, but it's hard to do apples-to-apples, because the economies are very different.

US soldiers/sailors/marines/etc. need to get something on-par with US wages. Ditto for defense contractors -- those programmers need to get paid alright, otherwise they're going to piss off to SV, NYC, Denver, etc. Average salaries in China and Russia are going to be lower, as are costs of production of hardware, so you may get a lot more relative to the size of the investment.

It seems that the military complex in the US is a country-wide employment program paid by taxes.

This is true, and goes back to the settling of the Western territories.

Amusingly, many of them allied with Southern Democrats (this was in the 19th century before the party platform switch) in proclaiming their desire for independence from the oppressive government, while in reality their livelihoods were entirely dependent on the government paying out money (railroad subsidies, agricultural subsidies, homesteading programs, etc). It also had a lot to do with for the anti-black sentiment of the South aligning with the anti-native and anti-Chinese sentiment of the west.

This is also true in particular of the rise of California, which at one point accounted for nearly half of the country's military spending. If California were a startup, it's angel investors were the military and those seeking gold (or more precisely, those seeking to establish railheads to ship it back East).

Around that same time, something like 80 percent of all federal jobs were at the Postal Service.

This is quite literally true given the amount of money and jobs the defense base (not just direct contractors but the entire supply chain) injects into local economies. Most of American manufacturing has been off-shored. Regulations prevent defense base procurement of even raw materials (such as aluminum, titanium, and steel) not made in the US.

Presentations are made to congresspeople of the first, second, and third order economic effects this has particular to their district down to supplier street address (dealing with .shp files will one day kill me) with heat maps of surrounding economic activity driven by the location of these businesses. These four lunch places entirely sustained by this job site, those jobs helping drive a couple service industry jobs, etc.

I’ve come to view the defense contractor space as pretty much a 1:1 economic response to the hollowing out of American industry during the 70s, 80s, and 90s. Onshoring back the dollars in the only mutually politically viable way. Social spending has become verboten for Republicans. “Big business” is the same for Democrats. But they both know they need this massive injection of cash into the American economy to avoid the political situation becoming factually untenable. I think Americans probably need significantly more discontent and mistreatment by the state in order to become politically incoherent but if this spending (which is essentially welfare due to its excess post Cold War) disappeared it would be guaranteed.

When I heard recently about the last person who got pension money from the us civil war, I asked myself if the us military is the start for something like an universal basic income.

Every way in which a government spends money somehow is. In its essence it's a way to direct your citizen's productivity at something.

That something could just as well be advancing renewable energy, modernizing infrastructure or educating people, but ... yeah, if you really want to you can spend it on weapons, I guess.

Money sharing with corp. decision makers.

Your judgment of an entire industry using an anecdotal sample size of 1 is poor form to say the least.

I've worked on more than one contract supporting numerous others and I concur with OP. I guarantee you'll get the same answer from most people that have been in the contracting trenches.

I work with a number of contractors who say exactly the same thing as GP. There is so much cash that is just going down the drain with no accountability. That being said, venture capital is pissing away huge amounts of capital with no measurable output either. Capital allocation is bad across the board.

The difference is that venture capital getting pissed away is on the VCs; they took a risk and maybe they lose out or win big, but it's on them.

The out of control DoD spending is your tax dollars, which could go to fixing bridges or to schools or hospitals, etc.

I mean, I think the personal sharing of any story implies that someone else's views or experiences may vary, no? The fact that it concerned you to the extent that you called it out is certainly notable.

Other people may have had different experiences. But that is absolutely not what I saw, not from my colleagues, not from my friends at other firms, and not from the general consensus around DC at the time.

It doesn't matter who is in office, both the democrats and the republicans agree that the DoD needs more money. Always more.

That's because the DoD pork barrelling rains infinite cash on every state whether red, blue or purple. To vote against it is to vote against free money and jobs for your state.

Consider the possibility that the entire point of many of these projects is spending money and not what you think you’re trying to build.

As someone in defense software I agree. I always joke that I make Google salary since I'm paid $1,200+ for 5 lines of bash that takes a week of testing / verification.

> As someone who worked in defense consulting/IT contracting many moons ago, let me clue you in on a simple fact: there is an INSANE amount of money wasted on these contracts.

Am orginally from NoVA, prior military, and did contractor work for a while. Totally agree with OP.

I recall being in a bar in Reston, VA while a bunch of drunk-ass contractors were mouthing off about how much overtime and bonus pay they were getting, and if they bumped up to a role with TS/Lifestyle-Poly it would double. Meanwhile a friend of mine was in the FBI and got hurt at work, while working for a comparitively unimpressive yearly salary. Was already pretty burnt out but that's one of the moments that made me head to the other coast (and, eventually, leave the country).

Lemonade and Agora's IPOs in the last two weeks both doubled right out of gate and have kept going (1)

Bankrupt HERTZ pumped 500%+ over a couple of weeks in June, to such a degree that their board nearly got a $500m unprecedented "worthless" equity offering out into the public market before the SEC shut them down. (2)

Tesla is trading at $1,400. Their latest competitor Nikola - who has no sales and no actual consumer product - was briefly worth more than ford (3)

I would assume that board rooms & investment bankers are SCRAMBLING to get their IPOs pushed out in Q3 or early Q4 2020.

In the meantime hold onto your hats, don't risk what you can't afford to lose :)

(1) https://www.cnbc.com/2020/06/18/bankrupt-hertz-terminates-co...

(2) https://techcrunch.com/2020/06/17/all-bets-are-off-as-hertz-...

(3) *Was for a minute - off to an $18b valuation vs $25b for ford today haha - https://www.fool.com/investing/2020/06/09/as-stock-price-mor...

Don’t fight the fed.

Amen.

The trend is your friend until the end

It’s actually a poorly done job by the bankers when stocks pop at ipo. Means their client left money on the table.

It also means that their real clients - IPO subscribers - got the stock at huge discount.

As my financial advisor from firm of WSB likes to say, “J Pow make money printer go brrrrrr”.

Honestly none of this makes any sense, record unemployment, skyrocketing covid cases, people trapped at home shopping groceries online. Yet $TSLA is at 1400, and to quote my advisor “is going to the moooon”.

Sigh.

It's all in interest rates. Far-future speculative cash flows become worth more when interest rates are slashed, and the fed's response to the whole coranavirus situation is to cut interest rates to zero.

Or maybe $1400 just isn't worth anywhere near as much as it was a year ago

It kind of makes sense, $2 trillion of the stimulus wound up in commercial banks' deposit slips because people who didn't lose their jobs had nowhere to spend it. It's not surprising to me that people saw the markets tank, remembered '08, and thought "let me put my dollars to work a little bit more."

Now Robinhood rolled out fractional shares and anyone can invest in anything. Chaos, in aggregate.

Or maybe bitcoin was never worth $17,000. Speculation is speculation. Stocks like tesla have such a retail following that they are no longer grounded in any fiscal reality. To properly invest in these stocks, one needs to find a financial advisor specializing in reading tea leaves.

Since it was 1,000 USD or so and was raising quickly in early 2017, I thought a good price for 1 BTC was a solid 10k USD. Happy with more, but w/ its current usage and interest, it seems to be the #

I suspect they are worried about Biden winning. A big part of their sales is from police forces for doing surveillance work.

Given the movement to cut back on police spending across states a Biden presidency would pretty much guarantee a massive cut in police related revenue.

Historically Biden has always voted for police spending. Hell, Kanye West has a better political record of ameliorating the incarceration industrial complex than Biden.

From the long line of rounds of raises (what are they on now, their L round?) my guess is that Palantir is a money fire and is vastly overinflated in valuation.

It'll be good to actually see their numbers now in Q statements.

Without knowing what they actually do, Palantir is a terrifying name.

The company lives up to its name, I promise.

Do you say that as somebody who doesn't actually know what they do? Because it's exactly what they do. You can only admire the honesty.

All the Lord of the Rings inspired startups give me the heebie jeebies. Anduril literally translates to "Flame of the West."

Knowing a little bit, but undoubtedly barely the tip of the iceberg of what they actually do, Palantir is a terrifying name.

It's what they call crystal balls in Lord of the Rings

JFC it just hit me that it's named after the seeing-stones in LOTR.

It's the logo too.

Their campus in Palo Alto is also referred to as "The Shire", IIRC.

Thiel actually explained this in a talk when I worked there. It comes from Lord of the Rings which Peter Thiel is very fond of. Apparently the Palantiri aren't inherently good or bad, it's just that power's impact reflects the people who use it. The ancient elves or wizards or whatever used them for good. But Sauron got his hands on one and used it for bad. The sinister potential was a deliberate choice, to remind people to be sure that power isn't abused. Actually kind of a good message for the tool Palantir builds, but probably not good for PR.

Also Thiel is very consistent on using Tolkien's pronunciation. It's not PA-lan-tir. Or Pa-LAN-tir. It's pa-lan-TIR with emphasis on the last syllable. There's actually a written accent in the books: http://tolkiengateway.net/wiki/Palant%C3%ADri

Regardless everybody at the company sad PA-lan-tir.

I find it hilarious that Thiel is a fan of Tolkien, yet he seems to embody Saruman's ideology (if not worse), while Tolkien's philosophy is more of Gandalf's. Tolkien was generally against technology, but especially against something so easily abusable (if not abused already) by malevolent power.

Is Thiel seeking to build an army of ressurected elves to destroy all life in the planet? And what did you have in mind for "if not worse"?

Note that I meant Saruman, not Sauron. Unlike most of LOTR characters, Saruman is a very modern and enlightened kind of person striving for "Knowledge, Rule, Order", but using Machiavellian means to achieve this purpose. Maybe Sauron was also initially led by good intentions, one would have to look it up in Silmarillion.

It only gets worse when you actually find out what they do.

If there were any good in markets this IPO would fail. We are without conscience and this despicable enterprise will continue.

Markets don't have ethics. That's why we have regulation and democracy.

Markets in-and-of themselves do not have ethics, but the outcomes of a market reflect the ethics of the consumers within it.

and so far, the markets have overwhelmingly voted in favour of profit over ethics.

Shareholders only care about profits, nothing else.

Certainly many of them do, but there are also plenty of people who want to focus on investments that also have a positive impact: https://en.wikipedia.org/wiki/Socially_responsible_investing

One specific example of a robo-advisor that does this: https://www.wealthsimple.com/en-us/feature/socially-responsi...

If you feed a bunch of vegetarians beef stew and they don't realize it, are you allowed to call them unethical hypocrites?

As long as there was no attempt to misrepresent the content of the stew then the vegetarians are likely not actually as concerned about their diet as they may have claimed.

Ex-Palantir engineer here. What exactly do you think makes the company despicable, apart from them working with some entities you don’t like?

They don’t even source or store data themselves... they just provide a glorified federated search engine on top of data organizations already have, with fancy visuals and user interface on top.

Actively enabling mass surveillance.

The data is often collected illegally (because there are no consequences for doing so, and it takes the legal system years to catch up with the new data being collected, and then that is either legalized, or the agencies just move on to the next illegal source).

So just working with entities you don’t like, then?

If you think your best hope for addressing government behavior you disagree with is to try to take their tools away, you may as well stay silent completely because it isn’t going to do anything.

If your grievances are with the government, take your complaints to the government, no?

Because Palantir inherently works with third-party organizations, you're able to take every conceivable response and say:

"So just working with entities you don’t like, then?"

Not really sure if this will be fruitful, honestly.

Do you share the same feelings about Google and Facebook? The level of user tracking undertaken by the aforementioned companies is staggering. Google and Facebook know practically everything about everyone. Both Google and Facebook have federal contracts and have for years.

exactly. FB and google collect data. Palantir doesn't. Just because a company runs warm and fuzzy super bowl commercials doesn't mean it's more ethical.

>What exactly do you think makes the company despicable, apart from them working with some entities you don’t like?

I mean, this is certainly enough for me.

The Gotham system and who Palantir decides to provide it to borders on dystopian. [0]

>The LAPD uses Palantir’s Gotham product for Operation Laser, a program to identify and deter people likely to commit crimes. Information from rap sheets, parole reports, police interviews, and other sources is fed into the system to generate a list of people the department defines as chronic offenders, says Craig Uchida, whose consulting firm, Justice & Security Strategies Inc., designed the Laser system. The list is distributed to patrolmen, with orders to monitor and stop the pre-crime suspects as often as possible, using excuses such as jaywalking or fix-it tickets. At each contact, officers fill out a field interview card with names, addresses, vehicles, physical descriptions, any neighborhood intelligence the person offers, and the officer’s own observations on the subject.

This is terrifying.

[0] https://www.bloomberg.com/features/2018-palantir-peter-thiel...

It’s just a search engine... and don’t get me wrong, it’s a really robust and expressive search engine, but all it’s doing is data analysis. As scary as that is (and it is absolutely scary), it’s not anything that someone couldn’t do with tableau, or with a whole lot more patience and data manipulation, straight elasticsearch.

Palantir makes that easier, sure, but at what point is a tool sufficiently generic so as to not be held accountable for what it’s used for? And if there is no amount of generalizability that absolves the author of software from what its used for, how do you square that against open source software?

Should the elasticsearch authors feel guilty if their software is used for something unethical as described here? No money is changing hands is one difference, but for a decision of morality does the payment matter?

To be clear, I don’t think there’s one right answer here. Different people will absolutely have different opinions, and that’s not the type of disagreement that’s likely to be swayed on a forum, I just wanted to provide you a different way to think about it, if you’re open to doing so.

Palantir is not being misused. LAPD is using Palantir for its intended purpose.

> A major theme of palantír usage is that while the stones show real objects or events, they are an unreliable guide to action, and it is often unclear whether events are past or future: what is not shown may be more important than what is selectively presented. Further, users with sufficient power can choose what to show and what to conceal: in The Lord of the Rings, all uses of palantíri influence the action through deception or misreading of what is shown. Commentators such as Paul Kocher note the hand of providence in their usage, while Joseph Pearce compares Sauron's use of the stones to broadcast wartime propaganda.

This doesn't sound like the kind of device that should be sold to law enforcement, or anyone.

The entire point of this literary device is to show that the user does not actually understand what they are seeing leading to an incorrect assumption which ends in disaster. It's a cautionary tale, not something to be aspired to.

I realize this is a technology forum, so it's popular to think about it in a very laser-focused kind of way where technology is neutral, and it's only the WAY that people use it is a problem, but over and over in society we have seen that this is simply not true.

Nuclear fission is politically and ethically neutral. An atomic bomb isn't.

A hunting rifle is politically and ethically neutral. A fully automatic assault rifle isn't.

A search engine is politically and ethically neutral. The way that Palantir packages and presents it and the use cases it optimizes for isn't.

Now, I realize that there are a million different grey areas and exceptions with all these subjects. You can murder people with a chef's knife, after all. I'm certainly not out here claiming that if something CAN be misused it should be banned.

I'm not even necessarily saying that something shouldn't be built OPTIMIZED for a particular use that some would consider unethical. There is always the very excellent argument that technology can't be repressed, and if you don't build "the thing", someone else will. You can't put the genie back in the bottle.

But we all know that there are lines that are too far. I don't want to jump straight to Godwin's Law, but I would say we as society universally agree that it was indefensible for corporations to provide tools and technology to facilitate the holocaust and any form of genocide.

I am not saying that ICE or certain US police forces are comparable to Nazi Germany's organizations. But there is a spectrum. And somewhere on that spectrum there is a line. And the line needs to be incorporated and considered from a humanistic point of view - one of global ethics. (vs. national ethics such as "If the US does it it's not illegal/unethical").

And from that standpoint, myself, and many others, think Palantir's ethics and chosen way of business is way over where we draw the line.

> A hunting rifle is politically and ethically neutral. A fully automatic assault rifle isn't.

First, a hunting rifle is definitely not neutral. Vegetarians and vegans would like to have a word with you.

Second, an “assault rifle” implies select-fire automatic. Otherwise there is no difference between hunting rifles and rifles just painted black to look scary. https://en.wikipedia.org/wiki/Assault_rifle

Look, gun nuts can debate semantics until they're blue in the face.

I'm drawing a line between UTILITY and MURDER.

Vegetarians and Vegans may be absolutely right that in an evolved elevated humanity we shall no longer harm any other living creatures, but in the grand scheme of things there are ways to hunt that are humane, sustainable, and efficient. And there are weapons/guns optimized for this utility.

On the other hand there are weapons out there optimally designed for indiscriminate slaughter of other humans at close range.

> debate semantics until they're blue in the face.

> I'm drawing a line between UTILITY and MURDER.

The irony is so thick here, especially when you're talking about a search engine.

a search engine who's utility is abuse by fascist-adjacent paramilitary police organizations to round up immigrants, discriminate against minorities, and create a pipeline of non-violent drug offenders for the private prison system.

Yawn. A gun is designed to shoot bullets. The vast majority of guns are bought and never murder. A small minority of them are used to murder animals and an even smaller minority are used to murder people. Handguns make up the majority of that latter group.

I own a handgun to shoot bears and mountain lions that might try to put the moves on me when I’m deep in the Rockies. One of my friends hunts with an AR-15 body gun with no intention of ever using it on people (and it has no automatic mode). He has it because of the handy attachments.

Whatever classification you think exists to draw the line does not in reality.

>It’s just a search engine

do you have any comment on this?

https://www.theverge.com/2018/2/27/17054740/palantir-predict...

How do you square using 'predictive policing' technologies on predominantly minority populations to forecast who will commit crime, which sounds somewhat like Minority Report, with "just a search engine"?

complete bs

all the evil companies use microsoft excel as well, does this make microsoft despicable?

> apart from them working with some entities you don’t like

In principle, isn't working with entities you don't like sufficient to call a corporation despicable? Like if you think ICE is a despicable organization, and Palantir directly enables ICE, isn't Palantir partially responsible?

I also think it's morally problematic to run a company which directly profits from government waste, especially if you're simultaneously lobbying for increased government spending in your industry. I don't know if that description fits Palantir (I don't know much about the company), but some people in this thread seem to think it's not too far off.

A lot of people in this thread are raising questions about what Palantir actually does. I worked as a software developer at Palantir in 2014, so maybe I can shed some light on this though this info may not be up to date. Back then they had two main products: Palantir Gotham and Palantir Metropolis. Accompanying this was a big infrastructure to ingest data from various sources. First the two frontends:

Gotham is focused on displaying entity based data. Entity based data was often things like people, reports of events, invoices, etc. Gotham let you do things like click on a person, and see all criminal records associated with them. Or their relatives and known associates. This was called "search around" and was apparently a very big deal. One sample use case for Palantir Gotham was to search for people that made more than one purchase of the same Schedule 2 drug, and plot those purchases on a map. A common method of finding pill mills was to plot these drug purchases on a map and see when there's a bunch of purchases along a highway route. Big indicator that someone is buying in bulk, but splitting up their purchases to fall under thresholds which traditional alerting mechanisms relied upon. Palantir Gotham is the product that people usually associate with Palantir - it's the one used by a lot of three letter agencies.

I don't know too much about Metropolis. It was more about quantitative analysis. I think it was popular for insurance companies. IIRC it could do things like plot the frequency of adverse weather events on a map, and then insurance adjusters to fine tune premiums. I think there was also a Palantir Metropolis product aimed at small businesses.

Now, the above covers the frontend. A huge part of Palantir's operations is data ingestion. Basically, this consisted of moving data out of government or enterprise databases and moving them into Palantir's format. This usually involved forward deployed engineers that would work with the government or business and work out a way to ingest the data. An un-glamorous job but one of the really important ones.

Another huge component of the business was customer specific customization. When I worked there something like 40-50% of software engineers worked on business development. They rest worked either on infrastructure of UI platform (basically building the components of the UI that aren't customer specific). Business development at Palantir when I worked meant "software development for a given business (or agency)" not sales. If you worked in business development you were a developer for a given business or agency. This was also a lot of grunt work. One customer wanted to bucket items based on the phase of the moon. I distinctly remember this ask - maybe the FBI was investigating a werewolf. Implementing features and customizations like that was the role of business development.

Is it a powerful tool or is it snakeoil? Well, given that the workflows it's replacing are often whiteboards and excel spreadsheets it may actually be a very significant gain. Ultimately what Palantir has is a flexible way of displaying and finding relationships between data. Flexible enough to ingest very different types of data. Perhaps more importantly they have a technical workforce capable of rapidly extending the product to meet new customer needs. And if customers are happy with the product I guess it's worth it.

I just hope they've moved off of god-damn java swing for the UI.

Here's a few demos to actually see the product in action.

* This is Palantir Gotham: https://www.youtube.com/watch?v=jZug8kV28RU

* This is a more generic talk. Skip to the 22:00 mark: https://www.youtube.com/watch?v=XyDEHCcQ_bk

* Also gotham: https://www.youtube.com/watch?v=_Q8bwhAW2Mg

* Using gotham to detect money laundering: https://www.youtube.com/watch?v=XCR_LBzur-k

I want to find a demo of metropolis but alas I can't seem to find one.

So basically it's an Orwellian surveillance tool that justifies it's own existence based upon remnants of an unjust war on drugs created by Nixon? I'm so sick and tired of Palantir's advertising tactics of doing good.

We should enable the government to create massive databases of peoples lives and communication, then link those ... so that we can stop kids from doing ecstacy or "catch terrorists".

Give me a break. Palantir is basically like PRISM ... except it came out of the minds of Silicon Valley (from whom I would expect much more) ... than the NSA/CIA

The main use case or illicit drug detection that I as aware of was detecting fraudulent pharmacies selling shitloads of legal opiods. Which primarily affects rural white people - I don't think that's part of Nixon's drug war which primarily targeted inner city people. And the drug detection use case is only one example. IIRC anti-terror work was the first big use case for Palantir Gotham.

As far as whether or not it's an Orwellian tool, one thing to be aware of is that Palantir doesn't collect any data or have access to government data. The ingested data is still administered by the agency. I don't think they can even legally transfer it to 3rd parties.

Palantir builds tools to browse and make correlations with the data that government agencies already have. It's more of an investigation tool, not a surveillance tool. There's a substantial latency between something happening and being able to actually see it in Palantir - it needs to first be logged and recorded, and then go through the ingestion pipeline.

Sure, it's a valid position to be against government collection of things like pharmacy purchases and financial transactions. But that strikes me more of a moral position against government data collection rather than something specific to Palantir.

That's a weird comment. From what the OP described, the software is a data visualization tool. Palantir the company doesn't hold the data based on what the OP said. The govt or whoever gathered the data would be the owner of the data. It doesn't appear the Palantir software itself does the original data gather.

Right. But the data tool is designed to be integrated across municipalities via a shared schema -- this fundamentally changes the efficacy of the data -- it makes the data more valuable.

Palantir Gotham is used by the LA police department and the data integration creates new capabilities -- even if they keep the data. At the end of the day there will always be a trade-off between crime and surveillance. Palantir explicitly sides with surveillance.

It really doesn't matter who owns the data etc. The moral position of Palantir is very clear: that it supports surveillance.

Not sure why you're specifically against Palantir. Tableau basically does the same thing. Excel did it before either product. Police, NSA, CIA, FBI use Excel, AWS, GCP, Oracle, twitter, tor, bitcoin (insert any tech product) etc... If Palantir supports surveillance because their product can be integrated with municipal data, then virtually every tech company supports a surveillance state because any tool can be leveraged by these law enforcement entities to surveil a population more effectively.

what type of double speak is this? They literally sell data integration products with a specific focus on defense and police departments.

What you're implying is that someone that sells uranium to people that build bombs is just "a chemical supplier" .... like yeah the essence of what they are doing is the same, but the implications are totally different.

Using the example presented, Palantir is helping stop the opiod crisis which causes over 50,000 unnecessary deaths in the US each year. Hardly Orwellian.

It’s basically saying: forget the systemic issues (big pharma) and focus on the petty ones by surveilling everyone.

It's all there on the website as well for anyone interested... Palantir isn't all that secretive about their products:

Gotham: https://www.palantir.com/palantir-gotham/

Foundry: https://www.palantir.com/palantir-foundry/

This was long overdue. Palantir has always been a darling of the valley. I often wonder how much of Palantir is actually owned by DoD and if that information would be made public after they IPO.

Edit: Looks like In-Q-Tel, the investment arm of the Central Intelligence Agency, already owns some stake in the company. [1] [https://www.nytimes.com/2020/07/06/technology/palantir-techn...]

It’s hard to watch the stock market right now as an underpaid employee at a pre-IPO unicorn. Who else here is desperately hoping the next S-1 post is about their company?

From the other thread: Forgive my ignorance: For most other companies, I've seen a direct link to an S-1. Is sending in a confidential draft first normal and we just don't hear about it, or is the confidential draft sent but not announced usually, or is this unusual?

It's pretty common to submit a confidential draft. This started with the JOBS Act [1], which loosened disclosure requirements generally for "Emerging Growth Companies" [2] when they are going public and introduced the notion of a confidential draft registration statement. You originally had to meet certain requirements to qualify for confidential filing, but that privilege was subsequently expanded to all companies [3].

Exactly how common it is to announce your confidential filing is harder to assess, but it's not _that_ unusual. They likely decided their filing would leak and thus it was better to get ahead of it.

[1] https://en.wikipedia.org/wiki/Jumpstart_Our_Business_Startup...

[2] https://www.sec.gov/smallbusiness/goingpublic/EGC

[3] https://www.skadden.com/insights/publications/2017/07/sec-pe...

Very bold to do it in the current landscape but I'm sure they've got their reasons to do so

On the contrary, anything tech related is sky rocketing. The Fed is proving they are willing to buy up just about anything, market highs know no ceiling. Now is the time to pile in with absolutely any tech related offering.

So much this. Most people aren't privacy activists so Palantir is just a high profile contractor to invest in.

Bold in which sense? I can think of a range of reasons why any company, and Palantir specifically, could be hesitant.

If I were Palantir I'd be worried about the funding from my primary customers (defense, law enforcement, ICE) in the likely coming very Democratic federal government. There's been a lot of noise made about changing how much money is being spent on law enforcement, which would directly impact Palantir's bottom line. If they can IPO and gracefully transition to private sector work, then they'll be golden; but afaict private sector hasn't gone as well as they'd hoped yet.

Hence the rapid IPO. ;)

I'm expecting they'll have a pretty good IPO, but I'm considering going short for that very reason. They're almost a caricature of "evil" startup companies. Peter Thiel's association only makes it worse. Not committed to this thesis though, since it's possible they could be swept up in TINA mania for the next ?? months/years.

Scott Adams claims evil businesses tend to do the best.

What issue do you take with Thiel? I think it’s a bit of a stretch to call him evil.

Well he _is_ a founder of Palantir.

He secretly funded a lawsuit against gawker which was pretty shady of him.

>He secretly funded a lawsuit against gawker which was pretty shady of him.

You can't be serious. In a court of law, Gawker was found to be (extremely) wrong, and outing people ain't cool.

It was anything but shady.

The shady part was strategically structuring the damages for Gawker's insurance to not be eligible and ensure bankruptcy. If I remember correctly (don't quote me on this), they lowered it enough for it not to kick in.

He thinks women shouldn't vote.

Even if the military budget was cut by 25% I suspect it would still be the largest item in the federal budget or close (have no numbers, would love to know if its true). Point being, Palantir is gonna have plenty of money available to harvest under any administration.

In the 2019 budget, Defense spending as a whole was $676B which is a distant second to Social Security at $1T. If you cut Defense by 25%, it'd be ~$500B and drop behind Medicare which is $644B.

You have the wealth of human knowledge at your fingertips, don't spread misinformation.

Ref: https://en.wikipedia.org/wiki/United_States_federal_budget

Most people buy the meme that the US has an overblown military budget because it's a prevalent trope in movies/shows and because when they see the raw number they fail to account for how stupidly enormous the US federal budget is. If you told them it was only around 11ish percent and that most of that was boring stuff like troop salaries they wouldn't believe you. They simply see the cost of a new drone and fail to appreciate the vast sum of money that could be recouped from the tech.

Edit: Since the point score on this comment has oscillated up and down fairly quickly, here is a link from the World Bank detailing both total military spending and spending relative to gdp for various/most countries. A lack of consensus is fine but opinions should be informed when possible: https://data.worldbank.org/indicator/MS.MIL.XPND.GD.ZS

The comment didn't even say "overblown" which - as an opinion - could be debated. Their comment was "the largest item in the federal budget" which takes near-zero effort to determine if it's true or not.

This is true, I muddied the waters to no benefit.

Memes often have at least a grain of truth to them and the US military budget is no exception. ~3.2% of US GDP is spent on the military compared to a worldwide average of ~2.1% [1] That means, even relative to other countries , the US is spending 50% more.

Anecdotally, you may have seen stories like the military paying 10k for a toilet cover.

It's accurate to say that the US has a overblown military budget. In my opinion, this money would be far better spent on universal healthcare like other reasonable countries do.

[1] https://data.worldbank.org/indicator/MS.MIL.XPND.GD.ZS?most_...

[2] https://www.washingtonpost.com/business/capitalbusiness/the-...

As an American Id love to cut the defense budget by half and pull all our troops out of Europe. They can handle it themselves just fine. Some serious savings there.

But the point that Palantir will still have a big pool of money to play with regardless of who is in power is completely valid. Over 80% of Democrats (in both the House and Senate) voted to approve the 2020 defense budget. I believe the numbers were similar in 2019 too. They may talk a big game about cutting funding, but I will believe it when I see it.

Why are you comparing defense spending to a pension fund (SS)? Hardly seems appropriate. Having a bunch of old people around is expensive. Are you suggesting that we should go back to letting them starve when they get too old to work or die young because they can't afford medical care?

Better comparisons are Police ($100B) and prisons ($80B) but really you should be looking at spending as a percent of GDP. Defense is about 3% of GDP.

So we spend (nominally) 7% of our GDP to take care of our elderly. We spend (nominally again) 3% to police the world. No other country comes close to us in military spending but we probably make more from our empire than it costs for us to maintain it. At least so long as we don't get into too many stupid wars (cough cough Iraq).

I can see what we get for SS and Medicare and I'm fine with the taxes to support it. I'd feel a lot better about the military spending if less went to crazy boondoggles like the F-22. It feels like a giant waste -- we shipped all our factories to China and none of this crazy fancy tech is going to help us if we get in a fight with them.

Palantir feels like it might be another boondoggle but I've seen a bit of it at a large company I work with and it actually solves some real problems. It appears to be a slick integration of Spark and Git. It has some potential to solve problems that they haven't been able to touch in the 10 years I've been contracting for them.

The only thing I "suggested" was $1 Trillion > $676 Billion.

On the flip side, Social Security and Medicare have dedicated revenue streams which are meant to cover their costs. In the 1980s, a budget deal was arranged where Social Security would move from being a pay as you go arrangement to actually collecting more money than necessary with the deal being that the Federal government would borrow money from that SS surplus to cover, among other things, Reagan's tax cuts. Because of the way the accounting is done, the surplus from SS payroll taxes is then converted into Treasury bonds. A similar situation exists with Medicare. Because of this, an argument could be (and has been made), that SS/Medicare are not really a part of the Federal government.

The statement that defense is the largest discretionary part of the federal government is an indisputable fact.

> The statement that defense is the largest discretionary part of the federal government is an indisputable fact.

Your position appears to boil down to "if the author had said something different, they'd be right!" - yes, I'll agree with that. It doesn't make the original statement true.

Isn't both Social Security and Medicare paid for by their own specific taxes though (and the trust fund if it exists in case of SS)?

Spending on the overall “war machine” is split up into a number of buckets, specifically so that people like you can misleadingly point at one of them and say “that’s it, that’s the whole thing”, when in fact the military industrial complex milks taxpayers for well over a trillion each year in total: https://www.thenation.com/article/archive/tom-dispatch-ameri...

> If I were Palantir I'd be worried about the funding from my primary customers (defense, law enforcement, ICE) in the likely coming very Democratic federal government.

Why? Palantir grew to be a unicorn (not just a unicorn, but a decacorn) under a Democratic federal government.

The political environment is different now, with so many democrats running under a platform of LE reform. It may amount to nothing, or may even be a boon; but it’s uncertain which way it will go for now, just that it’s likely to change. One way to lessen the risk of a negative change would be to IPO now while the numbers still show great growth.

I wonder if the looming possibility of Trump not winning the upcoming election is one of the reasons.

I don't know anything about the legitimacy of this IPO, but if they have a competent ML team, then what terrible things they can do with their data are [unfortunately] worth killing for if you're an authoritarian government.

Congrats to a wonderful company.

Is this sarcasm?

No. Most of the deriders here have no idea of the amount of good they do.

Perhaps you can enlighten us?

Its about helping organization utilize their data - nothing more, nothing less. Palantir does not own the data. They do not fuse some secret magic data source with other secret magic data sources. They provide tooling and platforms to allow companies to make sense of their data.

Look at https://www.palantir.com/palantir-gotham/platform-features/ for instance. It lets you draw connections between data, it allows you to spin up graphs to look at relationships, or plot things on a map, or perhaps make a nice web-based pivot table. But it is not some draconian evil thing, where moustache twirling villains sit around, finding new ways to pry into your personal data. Nor are they some scary shadow government-corp that seeks to entrench itself in the military-industrial complex in order to control things from within. It's tooling, its refinements on data, and its your typical software as a service company. The value proposition is giant players are good at having lots of data, but bad at using it, or finding insights, or visualizing it, or whatever - so they offer a one-size-fits-all set of tools and solutions.

I am very tired of the misconceptions around them because of their cool name. Some of the most intelligent, humble, and hardworking people I know work there. Their mission is misunderstood.

lol

[redacted]

And the other founder, who is the CEO, absolutely hates Trump...

For what it's worth, Thiel isn't supporting Trump like he used to so far. https://www.theverge.com/2020/7/2/21312154/peter-thiel-trump...

I wouldn't take anything that Peter Thiel "says" he's thinking about doing too seriously. He also said this in 2016:

>I think one thing that should be distinguished here is that the media is always taking Trump literally. It never takes him seriously, but it always takes him literally. … I think a lot of voters who vote for Trump take Trump seriously but not literally, so when they hear things like the Muslim comment or the wall comment, their question is not, ‘Are you going to build a wall like the Great Wall of China?’ or, you know, ‘How exactly are you going to enforce these tests?’ What they hear is we’re going to have a saner, more sensible immigration policy.

The claim that Peter Thiel's connection to Trump has somehow benefited Palantir is completely absurd. According to Alex Karp, it takes ten years to build the kind of customer base with governments that they have built, the majority of which was built under the Obama administration.

What does Palantir do again? Serious question.

Build software to do data analysis + work with clients. From what I know.

Basically what it did in the LotR books and movies. Provide technology to questionably-ethical people that gives them insights into things they wouldn't ordinarily be able to discover, accompanied by longer-term downsides that aren't readily apparent at the time the technology is deployed and used.

This analogy really works for me. Palantir is a well-named company.

Data mining for primarily government customers. Named after the seeing stones from Tolkien's mythology.

Their site should have some info about that, let’s see...

“At Palantir, we build software that lets organizations integrate their data, their decisions, and their operations into one platform. Our software empowers entire organizations to answer complex questions quickly by bringing the right data to the people who need it.”

Whoops! Never mind, I have no idea what this is describing.

It's vague on purpose. Their target demographic is CIOs and CTOs at large corporations and bureaucrats in governments. That language actually makes sense to them.

>That language actually makes sense to them.

Maybe they think it does, but it can't actually.

It is reasonable to think a CIO may have some concrete data sources in mind that they want to integrate and build business intelligence on top of.

The language is abstract since Palantir develops bespoke applications.

But it does. You see, these high level executives often have a goal in mind, but the best way to go about it is with this sort of ambiguity because it gives them room to inflate the expectations and plausible deniability when it goes wrong. If you look at the PowerPoint therapy know as consultant-driven product development, you will see the same language. Vague, ambiguous when documented.

It means they apply current data science tools for legacy clients who have tons of data, but few modern engineers.

Part of their modus operandi is to eschew data science in favor of human in the loop pattern analysis. This is born from Thiel's experience at paypal where human in the loop was much better at catching human fraud than machine learning and other automated approaches.

Human-in-the-loop, definitely.

My point being that they still deploy a lot of low-level data science against data sets to prepare a view for human interaction.

And their typical customers have limited ability to do that prep in-house.

If you have to ask, you're not their target demographic. It's pretty safe to say they are not looking for potential customers stumbling upon their website and deciding to try them out.

Software consulting similar to Accenture and SAP. They have similar customers as well, banks, corporations, governments, etc.

call their developers "forward deployed engineers" which sounds cool to bureaucrats, who give them ungodly amounts of tax money to put their data into databases

My understanding is it's mostly just BI-style dashboards for peoples' personal information, plus marketing.

missing the key detail that their first and (as far as i’m aware) still biggest customers are the military and law enforcement

And Saudi government "diplomatic teams" with bone saws...

I figured everyone already knew that, my point was just that the tech itself may not be very impressive.

Bossware, but also for governments.

"Palantir software is a Java-based platform for analyzing, integrating, and visualizing several types of data, including structured, unstructured, relational, temporal, and geospatial. In addition to accessing all natively connected data sources, such as ... Palantir's import process allows Warfighters to easily draw in any other types of data, such as unstructured message traffic, structured identity data, link charts, spreadsheets, phone data, and documents.

Palantir has a link analysis tool similar to Axis-Pro, Analyst Notebook (ANB) and the Distributed Common Ground System-Army (DCGS-A) Standard Cloud (DSC), which are currently used in theater. The Palantir platform uses advanced database technologies, combined with a sophisticated front-end GUI that enables extraction, transformation, and linked of varied datasets. Palantir uses a revisioning database, also known as version control, to documentall modifications performed on objects and entities in the database. Users of Palantir can perform dynamic modeling scenarios in which they manually create links between unrelated objects and entities that update across all datasets in the repository without permanently changing any of the original data"

Essentially a graph database that plays well with geospatial polygon and route searches along with time series', useful for trawling through all your surveillance data and putting together who-did-what to decide who to, erm, target

There's a few screenshots and also pretty interesting user reviews in this Operational Assessment Report published by US Army April 2012

Actually here's my favorite user report on how they use the software, page B-6: "Look for bad guys. Find out who and what those guys are connected to."

https://www.wired.com/images_blogs/dangerroom/2012/08/1830_0...

So Neo4J (I think that has versioning?) with AirFlow?

Seriously, I have never quite understood what all these unstructured data companies do. Autonomy had quite a big product...which, as far as I can tell, was largely about spending heavily on sales and trying to get companies to pay you for warehousing data (with very little of the AI wizardry they claimed to be selling).

I think there a publicly listed company that also does this kind of thing...I don't get it. I understand why you can charge an unsophisticated company to hoover up all their useless data, I am not sure how much actual tech is being deployed here beyond that database.

EDIT: I think I am recalling Elastic, which I think has an unstructured data business...maybe that is just my imagination.

Not Neo4J, so much as a relational database with user-defined schema. They actually build the relational semantics on top of cassandra so that it scales like a beast. And they open sourced it - https://github.com/palantir/atlasdb

Now take that, add a few more database technologies to do remote indexing and on the fly imports on federated databases, and some geospatial hotness, give it a swanky UI (now web based, not java), and that’s pretty much the platform, or at least was as of 2014 when I left.

And the thing is, they’re remarkably open about all of that... it’s just that people like the boogeyman narrative more, so all the openness about their tech is completely overshadowed, but there’s a reason it and Pinterest were the two hottest internships for college CS students in 2012/2013.

Oh god, it looks like the schemas are defined in Java. Cold sweat emerging, ORM flashbacks...no, no, I can't, sources of truth everywhere, they are multiplying.

Srs, thank you. I will actually take a good look through that as I find this kind of thing interesting. I think there are a few projects that do the converse too: key value on top of relational. I will be honest, I am not sure that I immediately understand the benefit of relational on top of key value...surely that comprises the benefits of going key value? But I don't know, I guess I will need to read more.

These companies have more than just tech - it's a relationships and sales company. They make a product, but that product is heavily customized to a particular client, and each client gets their own customizations.

The product isn't mass market enough to be an off the shelf solution. So i would argue that the tech is less important than relationships/connections to build the trust that the product can be delivered as required.

Palantir is just CGI or Accenture or Booz Allen or IBM, but focused on Defence, Security etc.. They have devs with 'clearance' and the 'vibe and BS' of a Silicon Valley firm.

So their BS is more focused than Accenture BS.

To be somewhat less cynical - almost all companies have difficulty with IT. I worked at a large handset maker that did some seriously important things, and our IT was delerious. It was as bad as a random cracker company. To do internal reporting, data warehousing, etc we were incompetent.

Even companies that are able to transition their core product offers to being more 'tech' still lack operational technical capability.

It's hard, and there's so much churn it's difficult to even tell 'who is good, who is not, how much it should really cost'.

Reply on news.ycombinator.com