NSA’s top talent is leaving because of low pay, flagging morale, unpopular reorg
washingtonpost.com
washingtonpost.com
Pretty sure a top cryptographer or security expert should be worth several times that, especially if they are also dealing with classified information and safeguarding all the data collection (e.g. if the NSA is going to spy on us, wouldn't you like that to be as secure as possible?).
This is another reason why we're going to enter an era of corporate feudalism. There has been a cycle of:
1. perception of gov't incompetence
2. normal people ask 'why are we paying people so much for doing a bad job?'
3. gov't pay scales fall further behind private industry
4. rise in relative mediocrity
5. repeat
I think voters largely don't realize how much competence actually costs in the market now and will vent about their justified perceptions of inequality by preventing equalization of pay between the gov't and private sectors, but that will paradoxically just make things worse in the long run as the private sector snaps up all the talent and we end up with a barely-functioning federal bureaucracy.
It's not just a 'perception of gov't incompetence': it's a widespread phenomenon. Government employees tend to excel at working organisational politics (i.e., in serving their organisational customers) rather than in serving their taxpaying customers. Government work is a kind of kabuki theatre, in which everyone agrees to praise the emperor's new clothes and ignore the boy who points out that he's naked.
The unfortunate thing is that government work is really important: it requires the best and the brightest, not the lazy and the uninspired. Patriotism should of course be a sine qua non for the civil service, but so should excellence. We should pay market rates, but we should pay them for market quality.
Raising rates without working to improve quality would mean getting mutton when we pay for lamb.
First of all, let's make sure we look at the full scope of civilian life with a sober eye. Anyone working the counter at your local Post Office could easily work the register at any grocery store in your town. And any medium to large company has plenty of internal politics. Don't make the mistake of comparing the worst case of government to some idealized notion of private enterprise.
Second, take a look at what happens to people who do leave the government. Very often, their salaries go up in the private sector. This is especially true for specialists like the folks who work at the NSA. Seems like they can hack it.
Folks who have not worked for the government, or very close to it, have a hard time comprehending the fundamental differences between government and private enterprise. For one thing, when you work for the government, many rules of your job carry the force of law. Being too flexible might not just get you fired, it might get you prosecuted.
And there are so many rules. I'll give you one example scenario: a person leaves their job, and a subordinate steps up, takes on their workload, and does great.
In a private company, you'd just promote the subordinate. You already know they can do the work, you retain institutional knowledge, and it's easier to hire junior positions. Efficient and effective.
But in most federal agencies that would be illegal, as it might permit some form of corruption. The subordinate would have to keep doing the extra workload, while the supervisor requisitions a new position to replace the old one, and then runs a (highly regulated) open public application process. The subordinate can apply too, of course.
This is not because the supervisor is some terrible inflexible or dumb person. The rules are just stricter and more burdensome in the government than in private enterprise.
Why? Because everyone loves to believe the worst about the government, so it's easy to sell voters on the necessity of a shitload of burdensome rules.
So: if you want people to act smarter in the government, you have to give them permission to act smarter. Firing has little to do with it. You have to give people the opportunity to try things and make mistakes in good faith.
Have you ever worked for government? I'm a civilian contractor in a software architecture role to a public safety agency in a state government. Half the "senior application developers" haven't written code in 15 years and don't know what a unit test is. There are maybe 2-3 individual contributors for each manager. Not team lead, manager, and is in a $105-110k+/yr position that does annual reviews, on call scheduling, etc.
The floor I'm on has 25-35 developers, DBAs, and BAs, and I can count on one hand the number of people who could get hired into a junior technical position at my employer. There are people who are unironically counting the days they have left until they can retire and draw their pension when they have more than 5 years left.
> Anyone working the counter at your local Post Office could easily work the register at any grocery store in your town.
And they'd take a 60% pay cut to do so, wouldn't they? Anyway, this is about folks in technical roles. Running a register and handing out stamps isn't particularly technical.
> Second, take a look at what happens to people who do leave the government. Very often, their salaries go up in the private sector. This is especially true for specialists like the folks who work at the NSA. Seems like they can hack it.
Well that's sort of tautological, isn't it? "The people who get a civilian job are capable of getting a civilian job." These folks are in the "Patriot, not inept" section of the diagram.
> Firing has little to do with it.
Firing has a lot to do. I've spoken candidly about my colleagues here with higher ups in the agency. They've openly bemoaned their inability to fire anyone, especially the people who they know for a fact do nothing all day. Those people are soaking up hundreds of thousands, even millions a year, in taxpayer dollars, doing little work of substance or value, and preventing their positions from being filled by someone capable.
1. tolerance to short-term career uncertainty varies greatly between individuals (e.g. changing jobs, moving, having money in the bank to cushion the cashflow turbulence)
2. the gov't largely underpays in dollars in return for high career stability
3. if level of tolerance to short-term disruption is strongly positively correlated to general 'ability', then the government will end up with on average less capable workers
Not saying this is necessarily true, and certain fields are probably exceptions (do people who choose to become cryptographers prefer a quiet peaceful life?) but certainly the movers and shakers of industry are people who can stomach enormous amounts of short term risk.
And there are also heuristic arguments to think (3) might be true on average. Like, if people with more money in the bank can take more short term risks, and it's usually the most capable people with that cash available (because they had more earning power in the past), then 'valuing career stability' might be a negative signal for candidate ability.
There's also a fundamental difference between you being cavalier with your money, and me being cavalier with our money.
Being tax payer funded changes accountability, investment horizons, and access to capital. The CEO of Twitter gets to put whoever he wants in wherever, and eats the costs if it works bad, but to ensure responsible fiduciary practices with our collective wealth we, the taxpayers, demand less aggressive and efficient practices to limit corruption and incompetence.
This is a conscious social decision, to move slower and more carefully, and will hopefully be minimized with increasingly robust Civic Tech. Alternative federal structures would historically point towards oligarchies and widespread corruption.
Not for me, I moved to London as soon as I graduated, but later moved back to a medium sized city, as a compromise.
That "someone else" is often a hardworking, loyal employee who does their best to earn their pay (thank you, to those people!!). But their attitude is ground down over time by watching those other people do literally nothing, and eventually get promoted "because seniority."
So I fundamentally disagree with your premise that you could transplant that population into a business. I'm certainly familiar with freeloads in business, but I'd struggle to come up with an example company where 1/5 or more of employees do no work at all over multiple decades and then retire with a pension.
I disagree that firing isn't a big part of it, but you can't just have more firing. You need to let managers do their jobs in a way that isn't just following a checklist in a binder, which also involves hiring latitude, pay latitude, etc. All stuff that could be abused, but some abuse is better than uniform shit quality at high expense.
One easy example from both government and big industry: how much money do we spend on sending people to get mediocre extra degrees from mediocre schools that don't actually have anything to do with their job responsibilities but is a requirement for a certain promotion level?
There is no comparable position in the private sector (there isn't even a comparable position in different land agencies), and there is no way you could pick someone off the street and expect them to be versed in that legislation without a similar timeframe (and no agency will ever foot the bill to train you on it either. They'll just leave the position empty and the work to pile up for years or decades until a similarly skilled person becomes available).
Unfortunately, the biggest issue that this particular agency is facing has been a multiple-decade campaign to hire externally to the agency rather than promoting internally. The result has been a body of upper management from the public sector that neither understands said legislation or how it applies.
I worked at NASA for 6 years and almost fell out of my chair when I read this. Switching from NASA to Google was the most stark contrast of incompetent people late in their careers having political infights to keep "their" funding to people who didn't care about funding and just did the fucking work.
I can think of 10 people I worked with at NASA who absolutely could never hold a professional position in a place that requires revenue or useful contributions.
I'm not aware of any scientific studies, but I spent a decade on active duty, and seven years as a reservist so far. When it came to technical jobs, almost everyone was terrible at their job, which was usually an incredibly narrow and overly simplified version of what the job would actually be in the civilian world. Everything is compartmentalized to the point where even if you excel at your individual responsibilities, you would still be terrible in the broader field. Almost every marketable skill I learned during my time in the military was something I did on my own time that was completely unrelated to my day job.
After I left the military and transferred into a different kind of technical role, eventually I started helping with interviews. 15-20-year government service veterans with incredible resumes and sr. titles are regularly found to be unfit for even an entry level job.
There are exceptions, but the rule in the military is incompetence. Its not that there's something inherently wrong with people, its just that the military recruits intelligent people, trains them very poorly, and then never lets them do anything remotely interesting or challenging. They just follow whatever steps they were told to complete in training over and over again for the rest of their careers.
As with anything, there are exceptions. Some tiny parts of the military do interesting work and have developed a culture that's reasonably conducive to innovation and fostering talent. From what I've heard, parts of the NSA used to be like this, but the lack of ethics among the organization's senior leadership ended up being partially responsible for the reorganization. Additionally, there are other people that manage to become really good at something in spite of being constantly strangled by the government's attempts to hold them back. Some of the most brilliant people I've ever met started out in the military, but the overwhelming majority of people that serve in technical roles are unfit to do anything besides put asses in seats for the defense contracting industry (which 95% of the time isn't any more capable of specialized technical work than the military).
Not really. Don't get me wrong: I like the people in our post office. But they'd be killed in the private sector.
At my local grocery store (Safeway), they'll add a new cashier if the line becomes long. The Post Office? Nope! During busy times, I've seen the line go beyond 15 or 20, and there's this only 1 slow clerk taking his time, chit-chatting, in no hurry at all.
Or the DMV. Visit the SF DMV if you want to see government at "work".
Well obviously. If the employee is competent, then of course they are more likely to leave. This exact method of adverse selection, where the good people leave, is the exact problem.
That said, I think the difficulty--or at least the consequence of the difficulty--is exaggerated. As a practical matter every large organization must contend with red tape, much of it self-imposed. Most government agencies, much like most corporations, impose more red tape than the applicable law necessarily requires. Consistency and repeatability demand process. Red tape is merely an epithet for process that appears wasteful, especially out of context.
IMO, government salaries suck because of the politics: Congress is more likely to increase expenditures if that money is laundered^Wfiltered through private contractors. Bureaucratic inefficiency is just a convenient narrative for why Congress prefers outsourcing. Over the past 50 years almost all the increase in expenditures for Federal work has gone to private contractors. The size of the federal civil service has increased by about 20-30%[1] since the 1960s, but budgets have increased several times that[2].
[1] https://www.opm.gov/policy-data-oversight/data-analysis-docu...
[2] See, e.g., https://obamawhitehouse.archives.gov/sites/default/files/omb... from https://obamawhitehouse.archives.gov/omb/budget/Historicals
I don't know if this is a problem in the US, but something else worth mentioning is that for many jobs, public sectors pay much more (higher pay, better pensions, more sick leave, better job security, ...). So it isn't just about firing people for incompetence. [1]
If the govt should be increasing pay to match market rates, then it should also be decreasing pay to match market rates.
[1] https://www.fraserinstitute.org/studies/comparing-government...
I worked in military medicine, and the IT department at one of the hospitals was entirely staffed with (in some cases well-meaning), but wildly unqualified staff.
Mostly this was people who were previously department secretaries and who turned out to be just generally incompetent, unmotivated or abrasive. As they couldn't be fired, they were promoted to IT. It was a dumping ground.
Their "Database Administrator" literally didn't know what SQL was and that's how it was for everything. One of them told me to my face (I was a contractor there to train them on an app): "I'm only going to be here for 5 more years, so I don't really want to learn all this."
Disgraceful.
In my limited experience this depends on the area. For example I know scientists who work for the government because environment and policy is their passion and for better or worse the EPA is/was the place to be. First choice not second.
But I've also met developers that work for state government and... I don't think they could handle a private industry job for the most part.
The southern states where I grew up and went to college both lauded people who were able to secure government jobs. No matter how grueling or toxic a work environment becomes and no matter how inefficient workers become -- everyone stays in those jobs for the pension.
In cities and areas where education in literacy is higher it's as if the market "pushes" that desire out, and the perception around having a government job is largely negative (excluding contractors).
I wish there were a way to increase competition for gov jobs and change the perception of them without resorting to privatization.
How do you know? I know a couple of talented, hard working statisticians that work for the census. They tell me a lot of good work gets done there. Well, at least until this year when they have suffered complete demoralization after a Trump political appointee crashed their china shop.
You've just described every large institution I've ever worked for.
IME, large companies that were well insulated from competition were way more dysfunctional than the government, too.
I think that any large bureaucracy, including ones in big corps, attract people like this.
So you are saying that government employees are greater at serving the people who actually matter. Not the people who should matter, but the people who do.
Imagine a startup that focused on its customers' customers (maybe once or twice more removed) instead of the people paying it. Would anyone be surprised if it gets out competed by a different startup focusing directly on their customers, even if at the cost of their customers' customers?
This isn't unique to government. I've seen private organizations where employees will make choices that benefit their own department at the cost of the overall organization. I sometimes wonder if a natural selection-esque drive behind this is the reasons that inefficiency grows as organizations grow.
> We should pay market rates, but we should pay them for market quality.
I think this would fix a lot of problems, but implementing it is going to be a nightmare because it isn't just the rank and file employees who you need to do this for. You almost have to start at the top, and from what little I've worked with governments, you are looking as a massive cycling of employees and a large loss of institutional knowledge. Also, given that much of how any government agency works is dictated by some of the same people who would be replaced under this, you are looking at having to pay people much more than even going market rate because of the extra costs of working the position.
For example, hiring good government IT is not just going to mean having to pay market wages, but also having to account for paying people to work in dead languages and eldritch abominations built by poorly managed contractors who often still have rights to the software that prevent the government employees from doing major changes (lots of contract security).
Maybe some groups within the NSA need the best and brightest, but the government in general doesn't. A lot of government work concerns accounting and administrative tasks that are neither intellectually challenging nor do they have a clearcut equivalent in business. Most government agencies cannot be run like businesses, because there either is no market to speak with or the markets for their 'products' are not free and open.
All you need is people who get their work done.
> but we must also make it easier to fire people (or manage them downwards) for incompetence
That only works if you start at the very top. If you get what I mean.
Is there any basis for this claim and the others? It's repeated by many, but that doesn't make it so. It doesn't describe the people I know that work in government.
No, they don't. Billing rates are fixed based on the experience and education levels of the people working on the contract. If the government says they won't pay more than, say, $250/hour for a security expert that is all the contractor is going to get, and that in turn limits the amount the contractor can make. Pay at government contractors is better than the government itself, but it is still shit compared to the rest of the tech industry. This is particularly true when you look at variable compensation (which is virtually nonexistent in government contractor land).
The employee’s compensation is more about the number of pimps^H subcontractors between him and the contract.
Another example are service based contracts where the government pays a fixed price for delivery of one or more outputs. Those outputs might just happen to align with what particular employees do. These things are generally aligned to make the contractor more money vs the employee. But the employees are well compensated, and the overall deal is good enough that they don’t lose a lot of talent.
Which means that your managers are still getting paid civil servant wages, no matter how well paid the contractors working under them are. Consider how well run a tech firm would be if it paid less and less competitive salaries the higher up in management you go.
(Assuming 40h per week, 47 weeks per year because systematic staff shortage and EU-duration holidays because that’s what I’m used to).
NSA leadership has clearly acknowledged the impediment:
The big change these days is there’s a supply-demand imbalance between the outside and the inside
Total compensation for [SV and other cities that tech start-ups call home] jobs can reach $200,000 or more, meaning even relatively junior cyber professionals in the industry can make more than top officials at the NSA.
We have some of the best talent in our country. We pay them less, but a lot of people prefer working in the public sector for a variety of reasons. Some are idealists who want to build a better society, others like the flexible hours and the paid sick leave for their children, others again like the relative stability or safety or the fact that they get to work on big projects.
I think the bad image comes from the fact that our scandals are public. When the banking sector does a big project that fails you typically won't hear about it. When the government fails, you'll be reading news articles, blogposts and the likes as well as having researchers from prominent universities trying to pour their take on what the government needs (typically to ask said researchers).
The public sector does fail more though. I think there was a study in the 00's that showed around 75% failures on major IT systems in the private sector and around 85-90% in the public sector. But then we have to digitize an un-digitizable law, so I think this is only natural.
In my municipality we don't do complete failures often, but we don't always succeed in implementing as successful as we wanted, leading to a lack of benefit realization.
That being said, we operate around 370 different systems 50-70 of which we've build ourselves, and most of these are working fine.
I think I'm a nice example. I get headhunter a lot, and I could be making a lot more money in the private sector, but I prefer the public sector because I have Children and since my job comes with paid lunch breaks I work two and a half hours less per week. On top of that, I'm not expected to work 40-50 hours a week despite being in management. I sometimes still do, but then it's typically for interesting projects such as heading the digital voter registration process during elections.
This may be very different in America, but I suspect that the NSA would be able to attract the best of the best on image alone.
NSA has a booth and so does In Q Tel and a few others, but these are very sparsely attended. Where you'll see a line of 200+ for Facebook and maybe 1 person for NSA by accident sort of thing.
A number of the 3 letter agencies have very nice job propositions (you really can do work you can't do anywhere else) but their processes and bureaucracy (not to mention pay etc) really leave quite a bit to be desired.
For example a number of the NSA Presidential Fellows are ridiculously amazing, but for some reason (maybe my lack of exposure) I've not found them to stay in their role, or to necessarily be super positive about it.
Similarly when we were getting offers even with a lot of the red tape out of the way (e.g. already having a high level clearance), it would take months for paperwork - in the meantime you could get a few offers and deadlines for many of the well known companies.
Again similarly there were likely a handful that went to these agencies over the last 6yr and they were never our best + brightest; if you were drawn to that sort of work it made more sense to do a PhD with a NDSEG/NSF fellowship and do an internship at one of the military labs/NSA then to spend years there.
I worked for a well known (in this area) science non-profit agency, and while we did attract good talent, after I moved to the private sector (and literally earning twice the salary) that I found much more brilliant people working at the high paid private sector jobs.
So while I agree that public sector jobs can attract good people, I think that most of the best people are in the private sector where they can earn more money.
I have the opposite perception about this in the US: government jobs seem to be 'butt at desk 9-5' where private sector provides the opportunity for flexibility.
Having worked in public sector (for a small gov't agency in NYC), I know for the fact that most voters or taxpayers don't hear much about most of botched projects or millions of dollars wasted extravagant IT purchases that incompetent gov't career bureacrats make. Or why the last president of my organization was "forced" to resign years ago after a major IT f'up. Normally, there would be no firing, but the scale of the f'up was so gross that the president had to resign, but, despite his public standing as the head of the second largest library system in the country (funded by taxpayers and private donors), nobody got to hear what really happened.
I have a friend who worked for the NSA. We had a heart to heart over drinks about a year ago. She asked what I thought about her working there. I responded, honestly, that if I didn't know her personally I'd be starting to wonder if she was, for whatever reason, unable to pass muster at Google et cetera. Not the best decision to drop that after a day of drinking together. But I'm glad I did--it was apparently the nudge she needed to quit.
Going to work and feeling like you're working on secret, cutting edge kit with the smartest people in the country is tremendously important to many people. The NSA lots its veneer of whiz kids confidentially cracking codes for good with Snowden. It now feels like another federal bureaucracy.
Postscript: She now works in the private sector and is thrilled with how much closer to her intelligence her new colleagues are. If you work for the NSA and are not a numpty, you're selling yourself short.
the NSA is a bit overhyped there
If you're newer than 3 years at facebook or google, there's no way in hell you're going to have any work life balance. And even if you make it past the 3 year mark, it still depends on which team your on.
So she stopped being an indirect servant of popular will, and started selling ads. Great.
> She now works in the private sector and is thrilled with how much closer to her intelligence her new colleagues are.
Not only richer, but less moral and more likely to pat herself on the back. Even better.
You do realize Google does more than sell ads, right? Search and Maps alone are tremendously useful tools, not to mention the dozens, hundreds of other products. Drive, Docs, Photos...
https://news.ycombinator.com/newsguidelines.html
You break the HN guidelines quite a bit like this, so we'd appreciate it if you'd clean up your act.
The problem is that a normal tax payer doesn't interact with the R&D scientific side of the government. To them, their local BMV office (state govt) is representative of the entire (state+federal) government bureaucracy. When reports of government pay leading/lagging private industry come out, their first thought is that BMV worker, not the PhD engineer building mine sweeping robots.
The federal government doesn't make this any better by keeping NASA on GS tables. If NASA made the jump off the GS system (the Air Force and other DoD agencies have with lab-demo [1]), I think it would go a long way in separating scientist/engineer salaries from bureaucrat salaries. Once the separation is in place, you'd think it would be easier for the R&D pay tables to get higher bumps than the normal GS tables. OPM does have special pay tables for engineers at the lower end of the GS spectrum [2] so there is some precedent for giving scientists/engineers pay priority.
The talks of a federal pay freeze in 2019 have not helped morale at all for the younger S&Es I know. It's miserable to get a 1.85% COLA increase when your peers get a 7-10% raise. Patriotism only carries you so far and it seems like the NSA is figuring this out the hard way.
[1] http://www.wpafb.af.mil/Welcome/Fact-Sheets/Display/Article/...
[2] https://apps.opm.gov/SpecialRates/2018/Table041401012018.asp...
$166k/year isn't bad, by any means. But it isn't competitive for a PhD scientist or engineer with equivalent experience.
I honestly don't understand why someone would work for the NSA given the choice.
Believing in what the organization does (which could include several different narratives about why its work is good and/or legitimate and/or necessary)
Recruitment via channels that civilian firms aren't using and offering people an educational and career path they didn't expect to have
Recruitment via family, friend, and teacher channels
Wanting to get to know lots of secrets
Wanting to work with smart people
Wanting to work with fancy hardware (for some subspecialties)
Wanting to have one's unusual abilities appreciated and put to use
Wanting to be paid to work on cryptography at all before the 1980s
Wanting to be in a spy movie/novel
Wanting to get to perform real attacks against real people and organizations and not go to prison for it
I know I've met people in all of the first three categories, and I imagine I've met people in the other categories and not known it.
I was amused by this as I have worked in a lot of DoD environments where a tie is "overdressed" and makes others uncomfortable. Likewise for a lot of environments where the uniform is a hoodie.
There is a type of work and a class of problems that can only be legally pursued in a federal agency like the NSA.
A little off topic observation here, but people who can't handle adhering to a business professional dress code are exhibiting a low resilience personality. Not a good thing.
Not to mention the reality that a button down shirt, slacks and a tie, when properly fitted, are not constrictive at all.
LOL @ the idea of anyone in private industry giving me the opportunity to do that.
These are some of the smartest, yet most normal people in the world. They don't go apeshit about free food or Silicon Valley startups. You'll never get fired, and you can raise a family perfectly fine, especially with the great schools in the area. After about $70k in salary, people are generally happy anyways. They're perfectly happy with dealing with pure tech challenges, because it's interesting to them.
It should also be clear that the NSA does everything per law. The NSA doesn't spy on US citizens. Remember the Snowden leaks revealed filters that removed US citizens communications. Why would a TOP SECRET program have filters to remove US citizens communications if they were trying to skirt the law?
I can't recommend the NSA enough. NSA caused me to understand that nothing is impossible and that everything in tech is small stuff and manageable. After the NSA, I rapidly did impossible things. (want me to list them?) I honestly don't understand why anyone would do obvious & routine things at Google, Facebook, Microsoft, etc., when they could be doing more awesome & surprising things at the NSA. It's always fun to see Google come out with something the NSA did years earlier.
Finally, and it should be obvious, going to work in a tie is a lot cooler than going to any job where people wear hoodies... but the NSA doesn't make you do that.. I do hate badges though.
It's the opposite of Silicon Valley, and it's great.
Now there’s no way to pay a government employee market rate without a scandal. The public views government employees as conniving sloths, but doesn’t understand that they pay 3x for a contractor (including sales, legal, compliance, risk premium, profit).
If you make over $80k a year -- you are doing well in NSA [i]. The 150k number is pretty much fantasy land. With NSA on your resume I am sure a penetration testing firm would love to have you at 100k.
If I was an American, I would deny the premis. If the NSA is going to spy on you, wouldnt you like your fellow country men to refuse to work for them.
I often wonder if I'm the only person on this list that wouldn't work for them simply because I feel its WRONG.
https://www.opm.gov/policy-data-oversight/pay-leave/salaries...
> Pretty sure a top cryptographer or security expert should be worth several times that [~150K x several times]
On the other hand, I often wonder whether folks who are pulling in nearly a half million dollars in compensation in the private sector (as individual contributors) are really worth that much or if they're just benefiting from a relatively temporary windfall caused by scarcity in tech talent.Keep in mind that government salaries still do come with decently good benefits and vastly more job security than start-ups. That is attractive to lots of people especially late in their careers. Cutting edge IT-related positions aside, non-cutting edge jobs in government do not face such stark competition with the private sector.
https://www.seattletimes.com/nation-world/private-contractor...
The fellowship is really good money for grad school, about 36k, and you get good interning in the various national labs and good-ish mentoring (if you want it). My sibling was telling me about the dinners though, a real hoot. The old grey-beards from the pre-Apollo days really are clueless as to why all the fellows are leaving for the tech companies. They really just cannot understand why 'millenials' don't want to do the work. My sibling had to inform them that many (likely all) of the candidates are lying on their applications when they say that nuclear security is important to them. In addition to that, the ability to make more money, not have a black hole on the ol resume, and be able to talk about their job to a Tonder date are important. From what I was told, Tinder really blew their minds (pun not intended). I guess they don't get out from behind the fence much?
- Switzerland: Switzerland is undoubtedly one of the richest countries. I contribute this largely to the fact that they have a direct democracy. All laws are voted upon by the people which makes lobbying much harder. Who profits from lobbying? Almost exclusively BigCos.
- Singapore: Singapore pays it’s top officials a competitive salary. The Prime Minister earns close to 2m dollars. As a result, Singapore attracts higher qualified staff for state positions.
- Estonia: Estonia’s growth since its independence some 25 years ago has outperformed nearly every Eastern European country. The reasons for this growth are their lean government, a high degree of digitalisation (nearly all government services can be accessed online), and an innovative culture at the state level.
In brief, a lot of the inefficiencies in government agencies come down to the facts that in many countries these agencies are intensively lobbied, risk-averse and - as stated by you - do not reward performance (i.e. efficiency improvements and innovation).
Also Switzerland is notoriously bureaucratic. Even very simple tasks require visits to one of the many, many small government offices that dot the landscape and each interaction always costs hundreds of dollars ... even when the act itself is basically "print out a piece of paper and give it to you".
But I fully agree that direct democracy creates better quality of government in general.
It still might be challenging to get a decent raise short of leaving for competition though. Pay is very easy to address in the grand scheme, the work environment, the politics at many levels, the organization, the idea that you could and will go to prison if you were to violate a policy or blow a whistle, inability to talk with your spouse, peers, or even coworkers about work, etc.. those things are hard and they probably suck there.
I am not aware of any exceptions to the law, unless you mean the Special Executive Service. You can - eventually - make it into the Special Executive Service (SES), which pays somewhat higher (around $200k per year, depending on specifics). But that typically takes 20+ years of service and is reserved to less than 1% of the total workforce of a lab or agency. It is extremely rare for someone to be hired directly into the SES, unless that person is a presidential appointee.
Which means that under ordinary circumstances, my lab cannot offer a Nobel Laureate more than $200,000 per year.
But you should not compare compensation for the top tech experts. It is not in general the function of the government to develop technology. There are some exceptions (for which there are opportunities for sweetening the pot), but in general government manages things, buying tech when needed.
Majority of it's bureaucrats are very expensive, non-fireable paper pushers. Non-fireabilitg leads to a huge waste. Raising the GS rates will make government even more expensive. To fix things you need to completely reorg promotions, allow for easy firing, do a RIF and give those remaining significantly more authority and autonomy.
Current schedule is pretty generous already once you factor additional pay scale multipliers and benefits. And if you must raise pay for top tech, at least go for the NH schedule, not GS. My 2c.
I disagree. The US Government is the single largest investor in R&D in the country, and does much of that R&D work in house. The National Institutes of Health, by itself, does the vast majority of basic and applied biomedical research in the US. That research feeds commercial R&D. The commercial guys can't afford to do the decades worth of basic research needed before a new medical treatment can be tested in human subjects.
NASA, DOE, and DoD research labs basically invented the modern world. NASA and DoD developed spaceflight, satellites, satcomm, GPS, and digital computer networking. NASA was the single biggest purchaser of, and investor in, integrated circuits right up to about 1970. NASA did much, if not most, of the grunt work to make commercial aviation safe enough to be commercially viable. The US Navy did much of the early research work in radio, due to the need to communicate with a worldwide fleet, and is co-listed with MIT as the primary US innovators in radar.
> Pretty sure a top cryptographer or security expert should be worth several times that, especially if they are also dealing with classified information and safeguarding all the data collection (e.g. if the NSA is going to spy on us, wouldn't you like that to be as secure as possible?).
I don't want to deny your argument, but as I often write on HN, the "interesting salary metric" is not the salary per se but "salary minus cost of living [in the respective region where you work]". I would bet that the costs of living are much lower near Fort George G. Meade than in the Bay Area.
The reality is that the compensation package for government workers primarily appeals to those with average to mediocre ability and a strong aversion to risk (often also a tolerance for boring, repetitive work and/or bureaucratic infighting). When was the last time you met a bright, creative superstar working at your local DMV office?
That's why the US Digital Service is actually a pretty good idea. Tours of duty are preferable to lifetime career hires. I'd like to see much more of the federal government adopt that model.
I feel like this has to be the biggest issue. When I worked in the DoD space as a constractor the amount of clearance required to work in an NSA SCIF would net a contractor a base of $160k a year (with benefits as you'd be working for a contracting company). With government employees making less to manage or do the same job as contractors, why would you stay on as a government employee? Most of the government employees I knew ended up transitioning to private companies.
For a comparison, in the 90s Singapore tried targeting a rate of 2/3 of private salaries for their public servants. I haven't seen evidence that this is still a format target, but their public services are fantastic. (Of course it's a small island country, so hard to compare with larger places)
For the kind of work that some divisions of the NSA does, it's the only place you can do it legally. If that's your passion and you don't want to risk fines and jail time, well you have to take the pay disadvantage.
so the first step is to pare down civil employment so that the general populace feels that they are getting a good investment. government needs to start selling itself on being a good investment instead of a jobs program. its not hard for the general populace to get this idea when there is news all the time about bloated pensions and high debt in their cities and states because of government pensions.
so since the 60s the size of government at all levels has tripled where the population hasn't. part of the explosion and state and local levels is from Congress funneling money to hire people outside of the Federal level but instead at local. then when the funding dries up the jobs tend to stay anyway
Pension, and other retirement benefits are awesome though. Over a period of time that actually makes up for everything else.
Lots of highly competent people out there in the market making big market wages ... and they fail all the time. Even just paying isn't enough.
Therefore I think government jobs shouldn't even compete on pay. In some areas they are higher than the free market though, and this is actually weird.
But let's not forget the NSA's own self-inflicted wounds as well. Let's see, here's what I remember:
* mission creep
* overreach
* LOVEINT
* massive collection of bulk intelligence with dubious justifications
* withholding information about 0 days it wanted to stockpile for its own use
* misleading technical committees so as to weaken encryption and security related standards
* introducing flawed security protocols
* paying money to security companies (RSA) to introduce weaknesses
* (thus with the above four items) actively sabotaging its own mission of helping to secure our nation's systems
* allowing its documents to be leaked en masse
* allowing possibly its most powerful and dangerous (though we don't know) sets of tools to be leaked as well.
Did I forget anything?
* We need a powerful spy organization to compete on the world stage.
This is an unpopular truth that people in our circles tend to forget. Rome was an influence solely due to the success of their military. What do you think will happen when ours becomes ineffective?
And the NSA is a key part of our military. Intelligence is everything.
I think it's good the overreach was exposed, but I was worried about exactly this outcome.
Guess: what country spends on military several times more than any other country in the world?
But not at all cost. There has to be effective oversight, checks and balances and public discourse about its limits. Not sure about the first two in the US, but here in Germany they fight that tooth and nail. And their certainly wouldn't be any discourse without Snowden.
They certainly did that mostly to themselves and i hope the only lesson they learned isn't "next time we shouldn't get caught"
I disagree or at least don't agree to the extent I might have in a more embattled time (i.e. any time in the past). I do agree intelligence serves the military well, I just don't agree with military intelligence scope growth at a time when military use is shrinking. I'm not sure history repeats itself here as much as perpetuates the justifications being used.
"Anyone who clings to the historically untrue -- and thoroughly immoral -- doctrine that violence never settles anything I would advise to conjure up the ghosts of Napoleon Bonaparte and the Duke of Wellington and let them debate it. The ghost of Hitler would referee. Violence, naked force, has settled more issues in history than has any other factor, and the contrary opinion is wishful thinking at its worst. Breeds that forgot this basic truth have always paid for it with their lives and their freedoms."
- Robert Heinlein
This quote has always stuck with me, and I posted it here a couple years ago. The reaction was almost universally negative. But after wising up a bit, I think offensive ideas need to be faced squarely if we're to evolve as a society.
I don't like it either. But I'd like it even less if it were a truth we conceal, rather than a falsehood we suppress.
It's hard to tell. But one thing is certain: If you read 100 Decisive Battles, you'll see the supremacy of intelligence in warfare. It matters more than any other factor. And the NSA would be our key way to get it.
That being said: diplomacy is almost always better than war. But the ability to wage war must not be ignored.
The military is worth paying for as long as they protect our life and liberty but not when they are aiming at us.
There are plenty of services that are needed that high value employees aren't rushing to get in to.
I just wish they could be more careful. It's a dangerous power they have.
Perception is everything. I say this as someone who dreamt of working at the NSA for most of my high school years, but would now never even consider it.
Edit: although I bet a lot of it is the 4x-5x salary jump you can get in cyber security jobs by having the NSA on your resume. :-)
As for perception: These technologies exist, and it's up to us to make sure that they're used wisely. Right now, the institutions of the US and allied governments are the caretakers and safeguards. They're not perfect, but I'll take them over the competition every day of the week.
It's likely we didn't get the really juicy bits, but the vast majority of the stuff we learnt about was just operationalizing things we already knew.to be possible.
And maybe it's not in their charter, but failures to promote security among private industry leading to hacks like equifax, or home depot, or many many others, looks to me like leadership is off chasing political objectives while at home we get pecked to death and damaged by a million small cuts of fixable security problems.
One of the first three technical books I read was Bruce Schneier's "Applied Cryptography". I loved crypto and wanted to work in it.
Fast forward to my 4-year degree when looked around and assessed my potential career paths. NSA salaries weren't competitive to what I could make as a generic software engineer and the NSA was starting to get a bad reputation, and that was well before Snowden.
Fast forward to today and it's hard to see myself working as an FTE for the NSA under almost any circumstances, both practical and ideological.
a major problem with NSA employment is the mandatory 5 year contractor like status period with half-pay. the NSA expects seasoned professionals from the private sector to join its ranks with a 33k salary? get real. The agency also prides itself on an almost obscene reliance on Polygraphs, a technology no more accurate or scientifically sound than a babushka shuffling chicken bones and reading tea leaves.
I also wasnt exactly blown away by the interview questions. my technical screening for a Linux position came from former members of the military, who grilled me on 2 unix questions before asking if they could switch to windows, confessing they 'didnt know much linux.'
I can't find out why I see smart people repeating this all the time. It's not true. Polygraphs are very flawed, and they are especially vulnerable to preparatory countermeasures, but they have non-trivial predictive power.
When I ask people to justify their dismissal of polygraphs, they generally point to the fact that they are not allowed in court, but this is a terrible reason. Hearsay isn't allowed in court either but that doesn't mean it isn't useful.
they don't. like any other military/police agency in the world they want them young and dumb.
The stark reality was that at Amazon, in 2 or 3 years I would be making double what CSEC was offering. Early career pay in one was end-career pay in the other. They seemed nice, and I suppose what they were doing felt important, but the money was simply too good to turn down. (No regrets either- I worry if I'd gone with them, I might have wound up living in Russia with a roommate named Ed.)
When you have job security and low expectations placed on you, it's wonderful. When everyone does, it's horrible. There is relatively little room for the lazy and useless in the private sector.
Spoken like someone who has not spent much time in the private sector.
Sure bad apples are harder to eliminate, but that doesn't mean everyone is bad.
Of course our choices are going to be Hillary vs Trump if we pay the top leader in the country less than a mediocre software engineer. Of course Hillary will give secret speeches to Wall St firms - how else is she going to pay her bills? Of course the government is going to be incompetent and corrupt - we can't hire good people at the salaries we're paying unless they use their influence to obtain side gigs in the private sector (often after their service, but that means they won't do anything to annoy their future bosses).
Why does Sundar Pichai make 500x more than the President? Do we, as a society, value running Google well 500x more than running our government well?
Government salaries need to be raised significantly (2-500x) across the board, or the government will continue being the mess it is now.
<<According to Title 3 of the US Code, the US President "shall earn" a salary of $400,000, along with a $50,000 annual expense account, a $100,000 nontaxable travel account, and $19,000 for entertainment.>>
Where can a mediocre software developer get $450K+?!
Keep in mind this is in SF, and I don't know any senior software engineers here making less than $300k/year, more than everyone in the government except the President!
Why?
> Why does Sundar Pichai make 500x more than the President? Do we, as a society, value running Google well 500x more than running our government well?
Why must the Presidency be the most highly-paid position for the government to run well? That is a ridiculous and silly assumption.
That's up to Google, not me. Unlike you, I don't seek to dictate what organizations should pay their managers.
> The point is that the pay for participating in self-interested capitalist endeavors leads to outsize rewards compared to those dedicated to serving society.
You are making two implicit claims here, both highly suspect and probably false:
- Working for Google contributes no value to society.
- Working for the government is synonymous with 'serving society'.
Google's flaws notwithstanding, the government has a terrible track record when it comes to incentives, inefficiency, corruption (particularly graft), accountability, and so on. Equating 'working for government' with 'serving society' is baffling.
> If we value those who serve society, perhaps we should...
Therein lies the problem with your rhetoric. When you say 'we', you pretend to speak for all taxpayers, but you do not. Your claim that higher pay would yield better results is probably true for some sectors and false for others, yet you make no attempt to distinguish between these and speak as if a universal raise would lead to better outcomes rather than more graft.
Of course, the President doesn't really have bills. He or she doesn't exactly pay to rent the White House.
In the US, we've chosen the private sector to remunerate our officials after they've left public service, and our lovely government is the result.
Keep in mind that, as a public official earning top wages (~$150k/year) in DC, you can just afford to buy the median home in the area. But if you want a nice house (single-family) in a good neighborhood? Forget it! Whereas if you took your skills to the private sector, finding a place to raise your family would not be a major concern.
Singapore is an extreme example of this (perhaps too extreme) - the Prime Minister earns over $1.5M a year.
Actually the first family does have to pay for many incidentals, like every day food, and it comes in the form of a bill.
I don't see who this applies to in your last election
$450k is a lot and rare, even for the bay area
Remember this. As someone who started their professional life in 2000, lived through 2009, and went through a multinational's "reorg" (mass-firing by geography, several 10s of thousands of people), let me tell you :
As soon as you notice there is general discontent, and people no longer care enough about their jobs to ask questions like this to the top brass, it is time to make sure you have another job lined up. You can get better jobs this way, and the hammer will come down.
Above all, know that this Rogers guy is fairly honest. And yet, he directly contradicts the Washington Post. He's an exec, and he has ONE job : do exactly what his superiors say without bother them.
He (once for me it was a she) will NOT give you any piece of information you do not already have. They are there to get people to work together.
By "motivating".
By lying.
By cheating.
This means that when it comes to the health of the company and your future job prospects, a sober chat with the cleaning staff will yield more useful information than an hour long "fireside chat" with a highly placed manager. Not because they know a lot, but simply because they don't lie, and they know a little. An executive is not talking to you, except in the way a vet is talking to stray dogs at the impound. They don't respect you, they can't do anything for you, they don't like you ... and they're very, very good at hiding this (just look at financial presentations on youtube, say of the IBM CEO, knowing that she has for years, every 2 months or so, fired 1000+ people, with the constant online reports of them getting cheated out of severance. Listen to her, then think about the kind of person who does that. You half expect to get free candy and a hug every 2 minutes or so when listening to her. Or listen to Clinton talking, knowing that he's the sort of man that threatens women to have sex with him. Think about what those people he talks to really mean to him)
Of course he "talks" with staff and takes questions. Why ? Because despite his wages, he is not all that good at correctly transmitting information, so he is doing this so he might do his job better by having his team interpret his ideas, and try to clarify them to him by asking questions. This is done mostly to avoid the worst of screwups. Only questions that follow this pattern get answered, or even seriously considered (check those financial videos again, trust me you'll see this happen).
He does not have an opinion (I mean he does, but none that you'll ever hear). Remember that. You cannot ask him for it, you will never get an answer. And when you do, because there are amounts of alcohol that will make him divulge opinions, please don't start destroying careers (mostly your own, but you may drag others down) by telling other people what those opinions are.
He won't care about attrition until the ability of the organisation to carry out his boss's orders is in danger, or when his boss cares. Nothing other than that can make him care. But he's paid a LOT of money to make you think he cares.
It's sad because I started out my adult life as a very trusting person...
There are a mixture of feelings about the agency. The same can be said of the CIA and FBI. And this is nothing new.
A lot more people are comfortable with, say, targeted foreign attacks than broad domestic dragnets or attacking American companies or standards.
Are you talking about metadata collection?
From what I have seen, there are a mix of issues that cause this: - It takes a long time to hire someone (usually 2-3 months), and if they need a clearance, the process is even longer (some clearances can take easily over a year now, and that person cannot do any work until they have them). - A lot of people want to go into these organizations to do cool things, but then find out they have a lot of additional duties they do not wish to do. - Government salaries are capped, and it is very difficult to fire a government civilian.
The issue of taking a long time to hire someone means that if someone leaves, their job does not go away. So their duties have to go somewhere, usually on someone else, and that will take a minimum of 3 months if you hire someone off the street. If the job requires a clearance, you now are waiting over a year if the person did not have the right clearance (while the person sits on that spot, so you can't hire anyone else if they come up).
From what I have seen, most people hired off of the streets are from college/grad school, and they are excited to do technical work. However, since the government caps civilian salaries, they will frequently contract out technical work (so they can hire the very expensive technical personnel). This means most of the technical work is done by contractors. Civilians more often than not oversee the contractors doing the technical work. If you have a technically inclined government civilian, then those contractors frequently snatch them up.
So now you have a compounding problem of it is difficult to hire/keep technical people, and the work they do is more and more non-technical, as most of the work to oversee contractors is what is called an "inherently governmental function", so those have to be done by a government person. This is a negative feedback loop that really hurts keeping these types of technical people.
To give you an example, there was a reorg a few years ago to outsource NSA's IT. (Makes a lot of sense that an agency that pretty much only does IT would outsource IT, right?)
In the reorg, 500+ people were identified as having no discernible skill that was relevant to the mission. They were segregated into their own org and given busy work. When the outsourcing occurred, the contractor was incentivized to employ and retain them for 2 years, including the 500+. To entice them, they paid them above their government salary.
The contractor could not figure out what to do with them either. It treated them as "casual employees". They were given a salary, but did not have to report to work and had no responsibilities. For 2 years.
Once the incentive ran out, these folks went back to NSA and got rehired as government employees. According to the hiring rules, they went right to the top of the stack.
Cool, because the demand is gone too.
I told this story to my Uncle who worked for 30 years at a national lab. He said they had a similar problem and solution for using a broom to sweep up a mess in their work area. These stories sound unbelievable but I believe they are common and that situations like this regularly cripple and demoralize government work.
Can’t imagine why anybody would want to leave.
Also, GCHQ has hired 494 contractors for 71m GBP[2].
[1]: https://sites.google.com/a/independent.gov.uk/isc/files/2016..., page 40
[2]: Ibid, page 70
Another friend worked at the big James Bond building in London, was regularly taken around the world at short notice to do security things, and was on less than 40k. He loved that job more than anything in the world but literally couldn't afford to live there so went contracting for 5x the money. Apparently in the government a large part of salary reviews is based on (age + seniority), so you're waiting a long time for a decent uplift that may never arrive.
I'm particularly interested in China's equivalent(s?) of the NSA. Do they manage to retain talent? Is it because the PRC's elite believes more in its mission? Is it higher status than the NSA? Do they offer higher relative salaries?
For example, for successful US-trained Chinese researchers, the Chinese government offers generous funding, publication bonuses [2], and high pay (comparable to, or better than, US institutions) for those who are willing to return to China and run research labs there.
Additionally, due to the relative lack of commercial opportunities for talent, creative, and ambitious home-grown talent (vs the opportunities in the US available to similar US-trained talent), the bar the government must exceed is commensurately lower.
0. https://en.wikipedia.org/wiki/PLA_Unit_61398
1. https://en.wikipedia.org/wiki/Imperial_examination
2. http://www.sciencemag.org/news/2017/08/cash-bonuses-peer-rev...
I’m really curious if they’ll have the power to get more people killed at google or at the NSA.
To be used by not-top-talent.
Interesting times.
This actually points to one of my major career mistakes in that when I left my government job I was burned out on the standard I had just spent some years seeing through to law, and instead of working consulting on it I took a totally unrelated job (based on the enticements of a friend who promised we would get to work on something I really wanted to work on)
I left millions on the table just by that one stupid decision. but then again I was really burned out on it.
on edit: improved formatting
Some even register their own company - which makes sense - as freelancing full time, end of the day, is possible. Make your own schedule, decline to go to work, take a month off - it's good fun. We got some people booked for over 110 days out of 250 for 2018 already.
As for pensions and savings - folks go to wealth managers.
Hit me up if you want some really good training courses for your company or your buddy is looking for extra gigs (GCHQ folks are welcome).
Think productschool.com for cyber security, compliance, cloud and data.
Cheerio
Reorganization of dept. has created red tape. Ok, that seems like one of many opinions a source at the NSA could have.
Snowden hurt our agency and made our jobs more difficult. I'm not sure that an NSA source could have any other opinions on this matter, especially given that a) I've never read anyone who currently works at the NSA call Snowden anything but a traitor and b) the NSA certainly knows the identity of the sources for this story. I think we have to reject such complaints as facile.
That was circulated with the comment "the risk is not so much people running to the Russians as people running to civil society - the NSA is not as popular as they think."
The NSA, like much of the US system, demands absolute unconditional loyalty. Internal dissent is not possible. Taking dissent public will get you sent to prison where you're not allowed to speak to journalists.
17.
WTF.
Each specializes in collecting a different kind of intelligence or in a different focus for analysis.
[1]: https://en.wikipedia.org/wiki/United_States_Intelligence_Com...
I'd rather those perverts not peep on us; this isn't an unreasonable request.
if you're going to invade privacy on a massive scale, you better be ready for the consequences.
You're getting down voted, but I've seen this first hand. I don't completely agree with it. But the employer's justification was that he would never be able to trust that the candidate wasn't a mole. (I pointed out that the NSA is probably capable of better hiding its moles.)
Thankfully resume sorting is effectively already an entirely random process, so I hope this comment doesn't bother anyone too much.
Consequences such as a $500 billion market cap[1]
[1] https://www.cnbc.com/2017/07/28/facebook-touches-500-billion...
too late, Facebook and Google already do that
Like a job at Google/Facebook, where you'll continue invading privacy on a massive scale?
I've seen an uptick of negativity on HN lately, especially targeted at the big SV companies. Not intelligent criticism mind you, just basic trolling. For example your account appears to be 99% dedicated to shitting on SV companies.
VS
> Facebook is attempting to kill all open communication protocols and force everyone to communicate through their platform with powerful network effects.
There's no contradiction here. Both can be true at the same time. Large organisations (government & commercial alike) have even less of a fixed & permanent nature than do individuals, and can carry out huge numbers of activities in parallel, some harmful, some beneficial.
You can try some sort of rough consequentialist calculus on an organisation's overall benefits / harms to society, which is complex and subjective to an extent that gives almost completely free play to prejudice.
Alternatively you can look at the systems within which organisations grow and assess the functioning of the whole thing. I think this is the more powerful approach, having more potential for sidelining empty moralising. Personally I believe that for fundamental reasons, hypergrowth-oriented technological society inevitably leads to the destruction of everything of value (in truth a subject for a book, not a forum comment). Facebook & Google are natural & unwitting agents of these fundamental forces. At the same time, decent & capable humans work for them in large numbers, so inevitably make positive countervailing contributions.
The same can be said of most monopolies, like AT&T in the 70's , Bell Labs did top notch research, and the highly reliable phone network provided huge value to people and businesses alike. Although it did have high value, it also had a high cost to the consumer and few people want to go back to the days of paying $30/month for local phone service (plus $3.99 if you want to use touch tone), and spending $7 for a 10 minute cross country call.
It is more difficult to run open source on your handset than it is on your desktop/laptop/server and Google wants it that way to ensure you are sucked into their ecosystem of data collection. And handsets outnumber PCs. Google is directly participating in the destruction of open source computing at the consumer level while distracting us with contributions to a few projects.
Facebook is attempting to kill all open communication protocols and force everyone to communicate through their platform with powerful network effects. Anything they have done for open source is dwarfed by the decades they have set us back in open communication and privacy.
Your website can't be reached btw.
When you have talent and options why stay with a creaky ship that's sailing straight into a hurricane with a mad captain at the helm?
> and analyzing the intelligence that goes into the
> president’s daily briefing.
I seem to recall that the president's daily briefing is now a single page with pictures. How many people does it take to produce that?
Faker News