I understand that highly compensated engineers get to solve unique challenges on grand scale, enjoy their jobs and are
generally content with their professional positions.
But I just can't stop wishing for those engineers using their immense knowledge, authority and financial resources to bootstrap own tech companies and improve the diversity of the tech ecosystem. They don't need venture backing, they don't have pressing need to compromise and have the means to experiment with new business models, whether it is open source, open-* or social enterprises.
Pushing back a little bit, not everyone is cut out for the startup lifestyle and the problems they're solving may not be solvable without the scale of a place like Google. Rather than encourage them to get out, I'd love to see them publish more about what they're doing, what challenges they're tackling, and how they've gone about solving them. That information gets pushed into the general ecosystem and other people then can run with it and iterate on it quickly. I.E. Google publishing the big table papers that became HDFS/Hadoop and the entire open source "big data" ecosystem that spawned from that.
To add to your point, people who are trying for H-1Bs aren't necessarily the same people who can apply to YC on a whim and get a company off the ground in no time.
I would not wish startup lifestyle on anyone and I don't advocate for that. I may have been unclear, but the engineers in question are exactly the kind of people who don't need to take outside investments and are able to bootstrap (as in use their own funds and rely on immediate/eventual cashflow).
They would have the independence, resources and authority to create self-sustaining social enterprises ( maximizing social impact rather than profits for external shareholders.), rallying supporters around their mission.
Seeding acquired knowledge back into the ecosystem is an orthogonal/complementary activity and depends more on their employer. I am not sure we can encourage the engineers to do that. Also, while sharing innovation "blueprints" is definitely a net positive for the ecosystem, I think it disproportionally benefits the megacorps, while indies and middle-sized tech businesses are being slowly squeezed out. The web is becoming more centralized and I am wary of that trend.
> but the engineers in question are exactly the kind of people who don't need to take outside investments and are able to bootstrap (as in use their own funds and rely on immediate/eventual cashflow).
no, they're exactly the kind of person who wouldn't do a startup. you're thinking about this from the wrong end. risk averse people are generally the ones who make $1M/year at salaried jobs, exceptions like financial traders notwithstanding.
quite frankly if i could make $1M/year working for a megacorp i wouldn't have started my own company. these two things are mutually exclusive for me. why would i bust my ass if i could take 2-week vacations whenever i wanted and get paid $1M/year due to my elite skills?
in many respects, i started a company because i wasn't good enough to really 'make it' as an industry leading developer. i capped myself out as a consultant at about 150k/year. but i wanted more money.
many of my associates who also own businesses (tech or otherwise) also started because they were basically unemployable due to a variety of factors. issues with authority, massive egos, chip on their shoulder due to a previous fuck-over or failure, inability to kiss ass, did terribly in school, have a felony conviction from their misspent youth, etc.
these kinds of things aren't talked about in mainstream media but if you dig deep into an entrepreneur's past, you'll find this stuff a lot. normal, employable, salary-seeking people generally do not start companies of any kind.
> I would not wish startup lifestyle on anyone and I don't advocate for that. I may have been unclear, but the engineers in question are exactly the kind of people who don't need to take outside investments and are able to bootstrap (as in use their own funds and rely on immediate/eventual cashflow).
The provide data is gross income, not net. Subtract federal and state taxes and you're at a more realistic figure for net income. Then subtract cost of living (i.e. property taxes, utilities) etc., and you're at a much lower figure for net expendable income than the numbers here. Your argument also assumes they're in a position personally to take on the associated risks of leaving stable employment and pursue a "moonshot" relying on savings and hopeful future cashflow. They might not be there. Also, running a business/organization is a far cry from slinging code and architecting and requires a different temperament, personality, and skill set. The individuals we're speaking of might be lacking in those or the desire to pursue such an activity. They might rather solve problems than deal with GAAP vs non-GAAP compliance issues, HR policies, marketing, or revenue generation. All things a corporate leader must address that an engineer solving problems might only ever have to address tangentially.
> They would have the independence, resources and authority to create self-sustaining social enterprises ( maximizing social impact rather than profits for external shareholders.), rallying supporters around their mission.
There are a lot of assumptions rolled into that sentence that I want to unpack.
First, what you're describing sounds incredibly similar to a non-profit and we're swimming in them. Why would engineers do a better job at building organizations that maximize social impact than people who spend their entire lives doing nothing but? The only comparative advantages I see are (possible) money and engineering acumen.
Second, how are we defining "social impact"? That's an incredibly nebulous term and one could argue that being part of a business generates social impact. Think of profit and revenue as an indicator of value provided. The more people buying and using my services generally implies I'm providing something of value. Ergo, pursuit of a profit and providing services people will expend money to pay for might be indicative of activities having a "social impact". People are willing to expend money for your solution to $PROBLEM over the possible alternatives indicating you're providing some value. As an engineer if I help the company solve a problem that lets them scale services to increase the number of people who can use it, offer new features, etc. I'm increasing the corporation's value proposition and providing a service to people to help solve their problems and meet their needs, I would argue that is a positive social impact.
Third, what is a "self-sustaining social enterprise"? Generally, I've heard those called corporations and they resemble normal every day businesses. What is different about your notion other than they're privately held? They'll still need to generate a profit in order to grow and increase revenues to meet growing payroll and expenses. I'm having a hard time understanding how you're seeing social impact as different from bringing in $$$ in exchange for providing a solution to problems.
The other assumption here I think we need to address is the notion that the individuals possess the charisma to rally people to the cause. You're describing someone with an incredibly charismatic personality that draws people in, that's a significant minority of the population. Why is that skill set attributed to engineers making big figures? Generally, I've found the opposite to be true, the charismatics and visionaries tend to not be engineers -- engineering is generally too rigid for the charismatics and the people stuff can be too squishy and shades of grey for the engineering types.
> Seeding acquired knowledge back into the ecosystem is an orthogonal/complementary activity and depends more on their employer. I am not sure we can encourage the engineers to do that. Also, while sharing innovation "blueprints" is definitely a net positive for the ecosystem, I think it disproportionally benefits the megacorps, while indies and middle-sized tech businesses are being slowly squeezed out. The web is becoming more centralized and I am wary of that trend.
The open source culture is booming compared to what it was a few years ago. More people are involved in OSS and more companies are pulling in OSS software to solve their problems. Is the system perfect? No. Will it ever be perfect? No. Is it better than what it has been or could have been? Heck yes.
How does it overly benefit megacorps? Yes, they might have the resources to pursue implementations faster, but a ton of the innovation on those systems is happening at the small to medium sized corporate level where the engineers have the freedom to innovate. I'm missing the link between the web being centralized and small to medium sized businesses getting squeezed. Generally, if there's a "squeeze" it's the small to mid sized businesses getting bought out by megacorps to fill needs, which, is usually an encouraged outcome for some of them because they aren't going to "get big" otherwise and scale up to the point where they have more than minute impact. Being inside megacorp gives them the opportunity to tap resources they would lack otherwise and get their tech into way more people's hands. I'll raise the case of the Storm processor. It started as a side project at Backtype and become Nathan Martz's full time gig when Twitter bought them out. It's one of the "big" open source data processors on the market and they're still innovating on it.
Ideally there's a balanced ecosystem. Big corporations have resources and manpower, but are prone to inertia and lack responsiveness in the face of new technology. Smaller corps lack resources and manpower but try to get an edge through creating new technology or adopting it, proving it out, and extending it. Big corps finally come around and either adopt the new tech along with the rest of the people in the ecosystem or buy out small corp to get the people who built or are extremely experienced with the tech/platform/framework. These people then get the resources + manpower to work on their projects. Eventually some of them get bored or hampered by the inertia, drop out of megacorp, and start or return another small corp and repeat the process. That's the trend I see playing out over and over again.
Yes, there ARE exceptions to the rule. I.E. LinkedIn + Kafka etc.
Thank you for a detailed response and an opportunity to debate this issue.
>you're at a much lower figure for net expendable income than the numbers here.
I am from a country where most people earn less than $500/month. Most Americans earn much less than the numbers here. This argumentation always seemed a bit absurd to me.
>and pursue a "moonshot" relying on savings and hopeful future cashflow
While bootstrapping, a top engineer/s is/are always able to do consulting while simultaneously building a product in the same/associated domain. EmberJS/Tilde.io is a great example.
>They might rather solve problems than deal with GAAP vs non-GAAP compliance issues, HR policies, marketing, or revenue generation
Sure. The preferred strategy is to devise solutions for making all of that easier. One solution is to be a CTO/founder while delegating administrative and some business matters to the CEO/CFO/CMO cofounder.
>..similar to a non-profit and we're swimming in them.
The number of tech non-profits is miniscule.
Let's look at the organizations who build popular database software. This is a quasi-random choice without any specific reason.
PG(non-profit), MySQL(for-profit, Oracle), MSSQL(for-profit, Microsoft), OracleDB(for-profit), MariaDB (non-profit), MongoDB(for-profit), RethinkDB(for-profit), Neo4j(for-profit), Complexible(for-profit), Apache Jena(non-profit), Apache Cassandra(non-profit), Couchbase(for-profit), Apache CouchDB(non-profit), OrientDB(for-profit). Those are major SQL, NoSQL, graph and semantic database products.
There are 10 for-profits and 5 non-profits (3 of them are under the stewarship of one entity -- Apache Foundation). The list can be certainly expanded and weighted by userbase for accuracy.
Another interesting domain would be collaboration tools for developers. The list of popular products is much shorter here: Github(for-profit), Bitbucket(Pivotal, for-profit), Gitlab(for-profit).
3 for-profits, 0 non-profits.
My argument is opposite of yours: the number of non-profits is too damn small.
>Why would engineers do a better job at building organizations that maximize social impact...
Technology is a great enabler. Companies with [1-5]0 employees have valuations measured in billions. Obviously, most of them had VC backing (money & network effects) but same principles can be applied to social impact.
>Second, how are we defining "social impact"?
Depends on the domain. Number of people who are able to afford expensive high-tech medical procedures, for example. This is an easy part.
>Think of profit and revenue as an indicator of value provided.... I would argue that is a positive social impact.
Not at all. Think BP, early days Microsoft, Oracle, Linkedin, MongoDB, etc, etc.
>..the charismatics and visionaries tend to not be engineers..
One doesn't have to be perceived as a charismatic leader by general population, respect and authority among engineers/programmers/IT professionals/geeks is quite enough. There are millions of these and they possess powerful technological skills to change the world for the better.
> How does it overly benefit megacorps?...
This discussion in the HN format would require too much citations, data and tangents to be efficient.
Income disparity, unemployment, lack of social mobility is increasing locally and globally. Technology, robotics and automation is one of the main reasons for these trends. Corporations are beneficiaries of these trends and won't do much to stop them. I would much prefer non-profits and social enterprises to constitute a bulk of tech businesses while for-profits would be a rare exception.
The pg essay quote seems to have gone out of fashion here a bit lately, but there's a perfect one for this occasion:
> So one way to build great software is to start your own startup. There are two problems with this, though. One is that in a startup you have to do so much besides write software. At Viaweb I considered myself lucky if I got to hack a quarter of the time. And the things I had to do the other three quarters of the time ranged from tedious to terrifying. I have a benchmark for this, because I once had to leave a board meeting to have some cavities filled. I remember sitting back in the dentist's chair, waiting for the drill, and feeling like I was on vacation.
I quite disagree with the premise of this quote for the following reason. When commercial product is being built, external requirements and interests are imposed on founders and their aspirations to create interesting/innovative software will be subverted by the need to compromise. However, the essay addresses this problem.
The engineers have the opportunity not to take outside investments and run their companies as they see fit. I believe that those bootstrapped companies would bring more social value than startups who are restricted by the boards, investment agreements, valuations and competition in high risk, high yield markets.
You're missing the point: They still would have to run the business. Meaning, not do what they set out in life to do, namely engineering. Where they are, they can pretty much just focus on the problems at hand and do research. All of the other crap is done for them.
There's a whole class of problems that it only makes sense to devote resources to in the context of a large organization. Compiler, kernel, library work that makes everything a fraction of a percent more efficient is nearly unjustifiable for most small organizations, but of immense value to large companies because of the scale at which they operate.
If you happen to be interested in solving those problems, a large company can be a wonderful place to work.
You are absolutely correct and I would love to see a scalable, sustainable community-based solution where languages would be designed for the social value first and megacorps second. I would prefer this solution to offer financial incentives comparable to the corporate jobs.
I think the problem is even though it sounds like a lot, $1.5m/year is not really some kind of "Tony Stark"-like riches. Yes, they might be able to temporarily stop working and live off savings, but probably not for that long. And furthermore, $1.5m, which is really $750k after taxes, doesnt really pay a lot of salaries. Hard to have an engineering division on $750k.
Also, I know plenty of people who are doing exactly what you are saying. The reality though is bootstrapping a startup is unlikely to make it more of a success anyways. The VC game picks winners in both senses - that they fund 'winners' and that they create winners by their own influence.
So a bootstrapped startup cannot ignore the VC realm regardless, especially once it is time to come to market.
The era of a silent genius sitting in a garage building durable software for the ages is a huge myth alas.
The VC game picks winners in both senses - that they fund 'winners' and that they create winners by their own influence.
It's about 5% the first and 95% the second, sadly. The well-connected VCs pretty much know what the outcomes and the valuations are going to be in advance. The less connected ones get burned by adverse selection and lose money.
So a bootstrapped startup cannot ignore the VC realm regardless, especially once it is time to come to market.
That I'm not as sure about. You just have to bootstrap into something that VCs aren't likely to be interested in, and get enough of an early lead and enough good will that they can't just make a competitor out of nothing should their interests change. It's quite possible, as far as I've seen. Bootstrapping is far harder than it should be (in part, because the American middle class is dying and no one has any fucking money) but I don't think that the VCs are the cause of that.
Speaking as an above average engineer, but not one of the super spectacular, I envy the work-life balance that many of these get to achieve, avoiding a lot of the slavish work tech startups often put them through. Even if they funded out of pocket, it would be a long hours, market pressured situation, where enough VC or big company money can still bury you despite brilliance.
I'm honestly not surprised some choose the good life at a big company - the world often doesn't treat such skilled people well.
If youre well paid to solve exactly the kind of problems youre interested in working on, its easy to see why some would prefer working at Google / Facebook. Most technical people I know who have left to start their own business have shared that they now spend most of their time building the organization rather than creating technology. Some appear to enjoy this organizational hacking, while others have a tinge of regret about it.
Once you pay off your house, adjusting for cost of living doesn't make as much sense, especially for areas where rent or mortgage payments makes up a huge percentage of expenses.
