Sounds shocking... but having worked as a contractor for a goodly portion of my career, I'd argue that something is seriously wrong if you can't get at least one employee to vouch for you.
Really, every place I've worked, you could say this happens every month; almost always I've needed an employee to sign off on my hours, either the person I report to or the person they report to. I always took that as a vote of confidence, personally.
(I mean, obviously, even if I get fired, I expect to get paid for hours worked, but point being, the guy just signed off on spending $15k-$20K, depending on the body shop cut, for a month of my work, I'd assume that would trigger at least a few minutes of "what did Luke do for me this month?" and a phone call letting me know not to come in Monday if the answer was "Not much".)
> I'd argue that something is seriously wrong if you can't get at least one employee to vouch for you.
The corollary to that is that if you get to the point where you need to check on the vouchsafety of all your contractors - en masse - something is seriously wrong with your company.
I think Elon was pretty clear that he considers the situation to be seriously wrong, hence the extreme approach. It's a refreshing attitude; this sort of problem is common in most large companies.
Yeah, a startup founder seeing a problem and having a clumsy, heavy handed, poorly thought out reaction is super refreshing. Hadn't been this refreshed since like, Friday.
Confinity (Levchin, Thiel, Nosek, no Musk) launched PayPal in late 1999. X.com (founded by Musk) merged with Confinity in March 2000. In no way can Musk be considered a "founder" of PayPal.
> His first company was acquired by Compaq for almost half a billion.
Zip2? $340M. Nowhere near half a billion and that was in 1999 at the height of the ridiculous tech bubble.
The man's done a lot of stuff - there's no need to embellish them.
I agree. Tesla is only worried about one thing right now: they are about to run out of money. If Elon has identified that contractors aren't providing the necessary ROI or at least traceability to the ROI, he's making a smart choice to let them go. Tesla is, in the view of some, on the brink of failure, where failure is defined as insolvency.
I dunno, GM and Ford build a fuck-ton more cars and have better build quality (which is shitty by int'l standards). Seems like Teslas's problem is bigger.
Having owned both a GM and a Tesla, I'm amused by the idea that GM's build quality is better. I actually liked my GM a lot, but quality was lacking to say the least.
In terms of driving experience, most Ford cars I've driven (Contour, Focus, and Taurus, probably all from early 2000s; they are probably better now) handle like boats; turning radius comparable to an oil tanker, unresponsive brakes and gas, etc. So there are tradeoffs among all makes and models within those makes.
Weeelll... given that a modern car is a complex dynamic system I would presume it's not to trivial to state where exactly design constraints concerning build quality end and dynamics and handling start.
For the end user they both affect the 'feel' of the product - how it behaves on the detail level when in use. "I have a door I have a hard time closing" and "the dynamics feel a bit clumsy" deal with totally different things on the engineering level, but end users (generally) are not buying engineering but holistic products.
Namely, concerning this comment: Are there compromises made on the design of the handling to facilitate the build quality? And if there are, then those are not two different things, but systems which interact - hence, it's relevant to offer commentary on one while discussing the other.
> it's not to trivial to state where exactly design constraints concerning build quality end and dynamics and handling start
Maybe I'm being overly simplistic here but I'd say if the finished product doesn't match the design then it's a build quality issue, whereas if it does match the design and it's still crap then that's a design issue.
Arguably if they end product doesn't match the design because the design isn't feasible to manufacture then that would also be a design issue.
Unless the design is limited by build quality issues. The AK-47 is perhaps the most famous example, but most designs are a back and forth around the manufacturing process.
Ford doesn't compete with Land Rover, Mercedes and BMWs but deliver a good value-for-money nonetheless. Compare that to a Cadillac Escalade from GM: Worse quality than a Ford, priced at Mercedes level.
I've never driven a modern for F Series but I have driven most of their Sedans and I can understand why they're withdrawing them from the US market. They're quite bad.
We took a road trip from Poughkeepsie to Niagra in a rented Ford Focus. It had a USB charge port in the center console that came out when we went to unplug the cable. This got me prodding about at the rest of the interior and most of it was too easily removable with my bare hands. It was scary to think who the whole passenger compartment might just explode in your lap with an accident.
I know many parts of the car are held together with retention clips and other tool-less mechanical fasteners but this Ford was exceptional. I don't know if it was down too the poor materials quality or the shoddy engineering of the fasteners but everything could be pulled apart with astonishingly little effort.
And good luck finding the proprietary parts. My friends motorcycle displays used encrypted handshakes to connect to the rest of the engine - ostensibly to prevent theft, but incidentally delivering lock-in.
I only ask, because honestly, I bought ( ordered and customized online really ) a new 2015 GMC Canyon and I love this truck like I've never loved another vehicle in my lifetime... and literally NOTHING has gone wrong with it, not even a loose screw... or burnt-out light bulb...
For a 3-4 year old car, having nothing break yet should be regarded as about par, not some outstanding accomplishment. These things should be lasting at least a decade with reasonable maintenance, and every component that can't last a decade should be on the maintenance schedule for preemptive replacement.
2005 Malibu. The side mirrors stripped their gears, the exhaust got a leak and needed complete replacement, the brake rotors got scored in a bizarre way that the shop couldn't understand, the tie rod ends went bad a couple of times, the instrument cluster stopped working spontaneously, then on my next drive started working equally spontaneously, the ABS system would occasionally go on strike for no apparent reason, from time to time it would somehow start badly and end up in a weird low idle state where it ran extremely rough, but the good ol' "turn it off and on again" would usually resolve it....
My Tesla has not been problem-free either, but it hasn't been that bad.
They have been building basically the same cars for far longer so that does not say much.
Tesla is attacking too many different problems at the same time IMO. Just building batteries at scale and low cost is on it's own a huge undertaking. Now add cars, self driving systems, charging network, home battery and solar etc.
It's over valued as a pure car company, but the other lines could also be huge companies on their own.
Yes, but startups are shitty and wavy by nature, it stabilizes with time. While culling the contractors, they might lose sight of something important and get into the next crisis. I don’t think there is any other way around. Sometimes it’s even external factors like wars and financial crises like the dinosaurs of the automobile industry had.
While it definitely depends once your definition of startup... I bet most people in the automotive industry consider it a “startup”. Especially considering the age of the competition, Tesla is a brand new startup company with hyper growth in its plans.
> Sounds shocking... but having worked as a contractor for a goodly portion of my career, I'd argue that something is seriously wrong if you can't get at least one employee to vouch for you.
That sounds reasonable, although I could see a need for some exceptions. For example, I could see a company using a contractor for its front end web development and another contractor for running the web server. The first contractor would be working with employees, such as sales and marketing that decide what is supposed to go on the site, and they could vouch for that contractor. The second contractor, the one running the web server, could easily end up only interacting with the front end web contractor.
1) I honestly don’t know what the org structure is at Tesla
2) assuming there are hiring managers, if they’re hiring people they wouldn’t vouch for (to boost their dept size (again not knowing the org chart)), then managers are gaming Tesla, and that’s another problem
Maybe the person who hired the web server admin, has since left to work with something else. And there's a new boss / hiring manager, who is a bit clueless.
I suppose this scenario is unlikely to happen, but still, in a really large company, then, low risk, times many many people = maybe shouldn't be ignored.
I'm a regular employee, and I don't see it being any different. Figuratively speaking, my boss has to sign my paycheck every two weeks. If they don't like me, they can fire me.
see, I think there is a big cultural difference. Like, I always had that attitude as an employee, but I think that as an employee, it holds me back.
I think the key is that as an employee, you are expected to act in the interest of the company in ways that are different from what you are expected to do as a contractor.
My perception and experience is that as a contractor, I'm supposed to do what you tell me to do. If you tell me to do something stupid, I'm supposed to tell you it's stupid, and to explain why, but if you say do it anyhow, well, that's my job. It has more to do with task-based work.
My perception is that as an employee you are expected more to act in the best interest of the company.
I personally have a hard time seeing the difference between the two, because I'm not omniscient, I am sometimes wrong, and if the boss has heard my arguments for why the way they asked me to do a thing is stupid and still wants me to do it that way, well, maybe the boss knows something I don't know, right? The company, by making them the boss, has said that the company thinks the boss knows more than I do, so maybe the right thing to do is to go along with it?
(I'm mostly talking of "the right way to do it" in the sense of "the most efficient or correct way to do it" - I think of ethical dilemmas in a different sort of way.)
> My perception is that as an employee you are expected more to act in the best interest of the company.
As an employee, if you object to something your boss is really on board with, you can expect to be fired.
It is, in general, a career limiting move to tell anyone above you that what they're asking you to do is stupid. Maybe if you're a God-tier management consultant, you can propose changes and expect to have them implemented. As a regular employee or contractor schlub, no. If you want to keep your job, adopt their perspective.
(Illegal actions are another issue. While it's illegal in many jurisdictions to fire a whistleblower, in practice you're still risking your job to blow the whistle. Not that you shouldn't do it.)
>It is, in general, a career limiting move to tell anyone above you that what they're asking you to do is stupid.
My experience? unless you have like a bottom quintile boss, you can tell him/her almost anything, if you do it verbally, with the door closed.
I mean, in most cases, yeah, I'm technically more skilled than the boss... and the boss knows it, that's why I was hired. The fact that the boss hires people who are more technically skilled than they are means that the boss is good at their job.
Now, of course, there might be other issues related to the business or to other systems that I don't know about... the boss might not take my advice, but only a terrible boss won't listen to the advice of a technical individual contributor that they hired.
the places where I've seen people get in trouble for arguing with the boss is when they do it angrily and publicly, and when they don't get over it after the boss explains why it has to be another way.
I tend to think of myself as a professional first, and employee second. The distinction between employee and contractor becomes a matter of accounting and tax law.
>>> The company, by making them the boss, has said that the company thinks the boss knows more than I do, so maybe the right thing to do is to go along with it?
I don't assume that for one minute. I work in a job where there is literally nobody else in the company with my expertise.
>I don't assume that for one minute. I work in a job where there is literally nobody else in the company with my expertise.
So, I've been there, and... the thing that makes me chuckle is that a few years later, in part because I thought the boss was such an idiot, I started my own company with my own employees.
The funny thing was, I found myself, at times, getting into the same sort of arguments with my employee, only on the other side. There are... other factors in play, when you are looking at things from a organizational or business perspective.
Indeed, and I run my own business on the side. I've also been in management. I have a very close relationship with many of the managers, so they are pretty open with me about their thought processes and the issues that they face.
And they face the same arguments with the managers above them, and so forth... "So, naturalists observe a flea..."
In the race for attention, is this article interesting because it's a new model for implementing and delegating work, or more the fact that a company has decided to rapidly cut down on contractors?
Because there's lots of companies that rapidly cut down on contractors, and I doubt they'd all have an article written about them at the top of Hacker News.
I'm neither here nor there about the company, but I note that it competes for a lot of my attention. I feel like if they decided to switch to all Macs, or abandoned Slack for IRC, I'd see a story about it here on Hacker News.
As a contractor myself, this sounds like a great idea. If there aren't multiple people at an organization that can vouch for you, there's a good chance you're dead weight - or at any rate, there are no checks in place to prevent you from becoming so.
My Dad liked to tell a story. The parts of the body were arguing about which was most important. The heart pumps the blood, the brain keeps everything working together, the lungs bring in air etc... Then the anus closed up!
Even if no one vouches for the butt hole in department C they can still be necessary. Something to keep in mind before making sweeping changes without regards to the details.
The bigger concern I have is that it encourages extreme nepotism — people vouch for people from their schools and prior jobs, and so risks introducing systemic bias against people from different backgrounds. Obviously gender and race are heavily represented in this kind of thing, but even basic stuff like west coast vs east university nonsense can start happening.
Your own reputation is put on the line by vouching for a contractor, according to the article. So if something bad happens down the line, your continued engagement with Tesla will be questioned too.
Nepotism and bribes are going to happen, but I think there will be some fast lessons when Tesla axes employees for this behavior.
It sounds to me like you are mixing the concepts of trust, accountability, tribalism and pathological cronyism.
Personal vouching for other people generally is the best guarantee for performance in absence of better criteria.
Trust is not a bad thing.
Trust does not alone degrade accountability. But you need to have a second accountability mechanism in place. Trust, but observe is the classical zen guidance into holding accountability within a group.
Then the group dynamics kick in - tribalism and pathological cronyism.
Exclusive clubs are a typical organization principle in the human species and they increase efficiency within the group by increasing trust and improving the quality of communication through larger shared background. Tribalism alone is not a problem to members within the group.
To agents external to the specific group tribalism may manifest in many negative ways, of course, but it is totally context sensitive if it's a problem or not in the general sense. Many elite organizations, for example, exhibit highly tribal behavior.
Then we get to the black sheep of the lot, pathological cronyism. I think we all can agree without further explanations why it's bad, but, the prerequisites for the pathological cronyism are many of the same kind as needed for the positive patterns exhibited in human socities, including trust and tribalism.
The thing that keeps everything from collapsing into a self-satisfied old boys network is accountability.
But using trust - instead of, say, bureaucracy - as one of the key ingredients of a company culture sounds to me totally great.
No you’re missing the point. It isn’t “trust is bad” it is that trust necessarily excludes people from different backgrounds.
If a bunch of your engineers (or whatever) went to Stanford, they people they trust are also much more likely to be from
Stanford, because those are the people that they know. The same holds for any university.
Then as time goes by the organisation becomes increasingly dominated by Stanford(or whichever) because the follow on hires also recommend people that they know — eg Stanford and the likes.
Now you have ensured that if someone cannot - for any reason - get into one of the feeder supplies, typically because going to those colleges requires a huge amount of money, or they’re super prestigious (the fancy UCs say) which have strong correlation to wealth (if you’re poor you can’t support the costs involved in getting a successful application).
So now you end up where large groups of people, for reasons outside of their control, and irrespective of talent, cannot get a job at your company.
At no point does anyone say “I don’t trust X”, it is simply the requirement that you have to /know/ someone to get a job there. Even if people are not trying to be assholes it is required nepotism. Even if you’re trying not to be biased, your hiring pool has already shrunk to match the backgrounds of your existing engineers.
If you want you could probably make a dumbass simulation of this, and I suspect if you start with X colours it will devolve to a small number fairly quickly.
I don't really understand the argument that employers should recruit absolutely fairly. Humans are always biased. If there was some AI based solution that would deal jobs absolutely fairly it would negate this, but there isn't, so the only alternative to smart human based hiring is dumb bureaucracy based one.
The large scale discriminatory criteria of gender, race and age should be accounted for but beyond those constraining who you should vouch for sound extremely constraining.
I understand the argument, and I think the effect you suggest is plausible - I'm not just convinced it's such an effect that could or should be accounted for (mainly because I have no idea what process beyond blind bureaucracy could do this).
You are basing this off of...what, exactly? Employees vouching for contractors are able to do so along any metric, but given that they are sticking their neck out professionally on their behalf, I assume that the criteria they care about the most is the exact criteria that will later be used to determine if the contractor was actually a good egg.
He's probably talking about consultants. I'm sure Accenture, Deloitte, and McKinsey (among others) infections have run rampant in just about all departments. Consultants are great when they're great but almost unimaginably inept and expensive when they're not.
And the ones that are inept, are experts at sounding exactly the same or better than the ones that are great, making the difference difficult to understand without equal expertise.
Being a hard ass CEO isn't sexy, but it's what so many successful CEO's are.
> Often, it is like a Russian nesting doll of contractor, subcontractor, sub-subcontractor, etc. before you finally find someone doing actual work.
I'd also argue that these multi-level contractor setups are a major security concern. Ideally they should make it policy to have no more than one additional level between Tesla and the contractor performing the work.
Aggregate contracting will make assigning liability a difficult task and it's likely the person working for you is being underpaid as each layer takes their cut. Meanwhile you get 100% of the shit-storm from any media backlash about worker salary/conditions.
> Why do companies, when they could do it in house, use them?
A few basic reasons:
* You can't do it in-house as well as a contractor who specializes in this field, or as well as this particular contractor. Very broadly speaking, the best people in field X want to work for themselves, do things their way, and keep the revenue from their work. They don't want to be several layers deep in management at your company, to have f-ng morons (i.e., anyone else) telling them how to do their job, and to be paid a salary for the breathtaking, save-the-day miracles they routinely output.
* It's not worth the investment: You would have to find and hire talent, ramp up, provide resources to them (equipment, training, etc.), be able to manage the talent in a field that isn't your expertise, and keep them busy and interested. Maybe you don't have enough scale in the things they do to support quality resources; maybe you just don't have enough work or enough interesting work for them.
* Every additional employee or capability is a distraction for the organization, including management. Keep focused on the things you do best and that are strategically important; you don't even have time for that. If you're small restaurant chain, IT isn't strategically important - it just has to be reliable and good enough. Outsource it and focus on world-beating food, service, and marketing.
* You don't want to pay benefits, taxes, or have the other responsibilities of being an employer. See: Uber.
* Sometimes it's just politics: I once worked in an outsourced capacity for a company where the IT manager simply didn't want to invest in a technology that would threaten their fiefdom, so somebody brought in a contractor. It was not a fun place to be.
There were a few good reasons I saw over my years as a contractor:
1. You can get a functioning team much more quickly.
2. It's a lot easier for someone unfamiliar with the field in question to evaluate the quality of a consulting company than to evaluate a prospective hire.
3. It's a lot easier to get rid of them once you don't need them anymore.
some jobs are really temporary, the electrician installing your painting booth has no future at your company, the guy making the tools for the press either. Some consultants are here to solve a painting adhesion problem, and they have no reason to be here when it is solved.
It's not inefficient if your metric is quickly getting workers in the door as contracting firms are built for this purpose. But you could argue that it will result in a poorer selection of high-quality candidates. I'd say this is because when hiring non-technical workers they're incentivised for numbers.
This is in contrast to employment consultants getting fees in intervals over a period of a person's time with an employer, usually done for executive positions. So I'd say optimise your agreements with contracting firms for quality rather than quantity by paying them bonuses in intervals, with a low initial payment for the finding the candidate.
It can be helpful to view in-house employees as workstations and contractors as AWS. Employees can take more time to find, train, and manage while contractors can be hired (and fired) rather quickly.
The government has penalized companies who hire full time employees with lots of expensive regulatory requirements and litigation risks. Hiring a person is one of the riskiest things any company does. It's foolish to do so unless absolutely necessary.
Is this meant to be shocking? When I was a consultant at a big European telecom company the department got a new manager who started by saying that there should be no multi-year consulting contracts - within a few months of hiring a consultant the position should be either inhoused or outsourced. I remember thinking “about f-ing time” even though my job would probably be one of the first on the chopping block. The financial incentives are so screwed up I can't believe consulting is still such a big business.
Without even factoring in work culture, the cost of a FT employee is really really high in a place like Australia where you basically can’t fire people. Want a 6 month project done, makes more sense to pay for a consultant team than try spin up new employees. They come in for a set period, fixed deliverables and accountability, then they go. Effect multiplies when it’s something like IT and the business is completely unrelated to IT.
You're being downvoted but my work with a state government frightened me. It was amazing how many contractors are building castles of crap and then protecting them so that they can make an exorbitant living off of the teat of the taxpayer - and I'm about as liberal as one gets - AND this was at the state level.
While I largely agree with your sentiment, quite a bit of the problems with contracting that the US government has is because they don't do contracting well. Specifically, they are understaffing many offices, and they are putting not-so-qualified people in lead roles. Writing a good contract is hard. It's harder when the right staff are not in place to do it -- this is the current state of most US government contracting.
Grant programs always have multiple consulting firms that either evaluate the program or help manage the program. It was always so fun navigating the government/ contractor directives.
This is what happens when you don't fill out your TPS reports correctly.
Sarcasm aside it seems like Tesla has operational issues. This coupled with the story a few weeks back about somewhat "slapdash" power module assembly is a bit concerning.
Musk is a visionary... but few visionaries are excellent operations people.
This is one of those things that’s going to be painful in the short term and pay off in the long term, but given Tesla’s precarious financial situation, who knows if they’ll even have a long term.
It’s possible that musk is just trying to cut their burn rate and is going to blame the inevitable decline in production after this on ‘restructuring’ or whatever.
Having an employee vouch for you... does this not imply they don't even know where their money is going as a company and who is valuable and who is not? I agree, that having the line worker / employee using the part's blessing is important, but surely there are more holistic top-down ways of evaluating this?
What's the real message. Employees prepared to be fired as soon as we reduce the outsourcing budget. Your jobs are not safe so if a better opportunity comes up, take it.
Sometimes when employee learns how much given contractor make, they get very bitter. I wonder how it is going to play out in terms of vouching... edit: typo
Tesla seems like a company where people would place a premium on being a direct employee, vs. another organization. I'd certainly feel more 'pride' in saying "I work for Tesla" than "I work for X" for almost all non-SpaceX or tiny-startup values of X, and that's probably generally held by people who would contract for Tesla.
Tesla is doing more innovation than the median tech company, and probably more than the 95th percentile. Plus, it's a consumer-understandable product, so you could "explain what you do at a dinner party" to someone outside of the domain.
"Engineer at Tesla" beats "engineer at IBM" but probably is socially inferior to "radiation oncologist".
Obviously status is to some extent subjective and culturally/individually determined, but there's broad agreement that certain jobs are higher status than others by various groups of people.
I believe that in this particular case a contractor could be a company of couple hundred (or 1000s) people or so, producing a sub-part of a car. Car makers mostly just put the car together - they do not manufacture parts.
Which begs the question: can Tesla afford to fire contractors? R&D on complex parts takes years, even at Tesla ...
I wonder if this is to protect ip. I know of one self-driving startup in which the founder recently had his wife get an internship with Tesla's self-driving team. I'm sure there is a big risk of ip theft or just learning other things that can give a competitive advantage.
> Often, it is like a Russian nesting doll of contractor, subcontractor, sub-subcontractor, etc. before you finally find someone doing actual work. This means a lot of middle-managers adding cost but not doing anything obviously useful.
Employees have a lot of expenses that contractors don't have. Payroll taxes, health insurance, other benefits, and generally there are more HR policies governing termination for an employee than a contractor, even in an at-will employment situation.
Yes, in the major tech companies very few people will elect to stay W2 contractors if given the chance to convert. Most would prefer full time positions.
Not that sort of singleton contractor - and technically having a company employee manage your day to day tasks will be take by the tax man as disguised employment - not a good idea.
Really, every place I've worked, you could say this happens every month; almost always I've needed an employee to sign off on my hours, either the person I report to or the person they report to. I always took that as a vote of confidence, personally.
(I mean, obviously, even if I get fired, I expect to get paid for hours worked, but point being, the guy just signed off on spending $15k-$20K, depending on the body shop cut, for a month of my work, I'd assume that would trigger at least a few minutes of "what did Luke do for me this month?" and a phone call letting me know not to come in Monday if the answer was "Not much".)