"I now have six amazing clients, and I’m making an amount of money equivalent to my Google total compensation package,[1] which proves the thesis that it’s possible to be a professional maintainer earning rates competitive with the adjacent market for senior software engineers."
His experience is totally unique. Please don't let this make you think you can quit your job and earn the same you did at your high paying FANG job. He's an outlier in the industry. I was a nobody when I left my job to work on open source full time. It was a year before I found a corporate sponsor and that was only because my friend worked there and he understood the value of what I was working on. Patreon was laughable in terms of what came back. Just be fully aware as you read this. With existing brand and following you can do what he did, without it you'll struggle immensely like I did.
> I’m sharing details about my progress to hopefully popularize the model, and eventually help other maintainers adopt it,
Hopefully this won't inspire people who don't meet the right conditions and whatever luck contributed to this existence proof.
I've known a lot of poor people trying to make it as independents in open source. I once sent a laptop to a homeless kernel hacker (and, earlier, sent them food), and had to find a laptop specifically to be small and discreet, because they feared being stabbed for anything flashy-looking. Another, who has done talks on their novel work at major hacker-as-in-HN conference, as well as other accomplishments, I had to tell them about Medicaid, because they couldn't afford to go to the doctor when they really needed to. One who accomplished something major that most HNers have used or heard of, was living in a trailer, and died. I've also known plenty of people in open source who had modest day jobs and were pretty stressed and depressed from money problems, and the cascading effects of that, despite being at least as tech-skilled as people making FAANG money.
If you happen to find yourself as the official maintainer of multiple open source components that are recognized as key by numerous enterprises (and cryptobro ventures) that are flush with cash, and you have ins at some of those, and you have a safety net warchest from years of FAANG, and enough reputation you could probably go back if the whole indie open source consultant thing didn't work out... sure, consider a consultancy like this post describes, as a lifestyle move.
Otherwise, it's like the movie star child of a Hollywood producer evangelizing this great career success formula they've found, prompting a bunch of aspiring actors to buy one-way Greyhound bus tickets from Kansas to LA, where most of them will be lucky if the worst that happens is they end up waiting tables.
I'm making about €30/month from Github sponsors. Of course I had to get a day job, so my open source time is limited compared to the previous decades. Though my company is fine for me contributing to some GNU projects or openssl.
I quit my job a couple years ago and secured several small retainers over a few weeks using my network. My employer at the time generously agreed to be my first client to smooth the transition.
Today I have am managing 8 active retainer clients, and regular 1-2 week audit contracts, while rarely working more than 40 hours a week. Virtually all code I write is open source, or on track to be so soon, and I only work with clients okay with that.
I am making triple my previous salary, and am actually onboarding new team members as a "tier 1" to help me meet demand without overworking myself.
I am a full stack security engineer with 20 years of experience, and most companies can't have access to senior security engineers without paying GAFAM money in the range of $600k+ total comp, which they just can't afford.
Instead I offer most companies start a retainer with my team and I for as little as 10 hours a month and we can be there when they need help with security architecture, important code reviews, risk assessments, conducting interviews, or just to help unblock people in general.
This model is a win for companies that can't yet afford experienced full time security hires in house, and it is a win for me who can be in control of my time and life with higher income, and only have to focus on the most interesting problems of many different companies with minimal exposure to internal politics.
I can't express how much happier I am. Best career choice I ever made. YMMV.
> His experience is totally unique. Please don't let this make you think you can quit your job and earn the same you did at your high paying FANG job. He's an outlier in the industry.
…and yet…
I run a self-funded SaaS business. I regularly pay (sponsor) developers of libraries that my software depends on. These are not large amounts, but they slowly grow over time. Additionally, there are some libraries (Semantic-UI for example) that are critical for me, but have been unmaintained for a while, and I'd gladly pay significantly more, on a regular basis, to have them maintained.
I am pretty sure I am not the only one. The money is there. The problem is in gathering critical mass: both for any single developer, to make a living, and for the entire movement, so that we shift from a culture of "FREE FREE EVERYTHING IS FREE" to a more responsible and sustainable "it's free, but if you depend on it, you better contribute money every month".
Quite frankly you're extremely unusual. As much as I sound cynical here, the only reason we use open source stuff in our production SaaS is because we don't have to raise a purchase order to get it or go through the whole onboarding process which is a pain in the ass. The money isn't even the issue; it's there and available but it's a bureaucratic shit show trying to give it to people. And the same is true everywhere I've worked for the last 20 years. Yes I know this is wrong.
Business idea: If there was a single corporate intermediatory who would handle all this sitting somewhere we could create a supply agreement with and funnel the cash through to the right people we could probably deal with it. We currently do this via AWS marketplace regularly so we don't have to deal with the paperwork.
I feel like there is a business waiting to bloom here. Imagine a stripe like company that says “we are the unified B2B transaction company” who takes both sellers of software and buying enterprises as customers and create a easy to use purchase system where a software dev in the US could sell to a company in New Zealand without worrying about
1. Currency conversion
2. Local tax collection
3. Invoicing
4. Any other local formalities
That is totally worth day 10% of the value of the product!
These companies exist, at least domestically in varying regions of the world.
A lot of software/B2B sales are procured through a channel whose primary purpose is an existing business relationship with the company you're trying to sell to.
They take a % as a transaction fee. Anywhere from 10-30%, depending.
I think it's a logical next step for companies like Crossbeam[1]. You have a network of partners, some of which are fulfillment partners. They can be the middlemen to expand your network and take a piece.
I get your point. I'm not too keen on administrative overhead, either.
> Business idea: If there was a single corporate intermediatory who would handle all this sitting somewhere we could create a supply agreement with and funnel the cash through to the right people we could probably deal with it.
Isn't that exactly what Github does through its Sponsors program? I think I only handle two endpoints these days: Github Sponsors and Clojurists Together. Github works very well, and they will even fold/consolidate new sponsorships into your existing invoices as you add them over time.
I don't think "overhead" is a valid excuse anymore.
I don't think current platforms make it easy for developers to make money from their software. Especially libraries. There's a total unwillingness to pay for support because it's easier to just open a GitHub issue and complain. I think you're a rare case and that really means developers can't make a living off stuff like this. The few exceptions are something like sqlite maybe. Other stuff ends up needing to be heavily VC funded or backed by corporate sponsorships.
If GitHub actually helped developers make money this would be a different story. Sponsorships are a tipjar, it's not a sustainable path, it's not a form of employment. Grants, same thing, waste of time. We need the ability for developers to put a Pay but on their repositories. This is not about optional sponsorship. This is about paying to download the code, paying for use after a certain point. This is about putting a real number on the value of software. It only works when you define the economic model. If each Dev has yo setup their own website, integrate payments, do sales, etc its a struggle. GitHub is a big enough distribution channel where they could actually streamline this, App Store style. I know they have a marketplace but realistically who's using it?
> If GitHub actually helped developers make money this would be a different story. Sponsorships are a tipjar, it's not a sustainable path, it's not a form of employment.
I really don't understand. GitHub does help developers make money. Sponsorships are subscriptions, not one-time tips. If you can get 20 companies to pitch in with, say, $250/month, you begin to look at a sustainable living. From a company point of view, paying, say, $1000/month for four most-used pieces of software that the company depends on, is still many times less expensive than hiring even a single full-time developer.
I feel like rather than trying to change the mindset ("everything must be FREE FREE FREE"), we are trying very hard to find reasons not to use a perfectly good existing solution.
With GitHub, sponsorships are either one-off payments or regular subscriptions. FWIW I've only ever been paid once through GitHub sponsors and that was a one-off payment. This payment ($500) was actually from GitHub itself because they use the software I work on.
It's the issue with the concept of sponsorship. It's still associated with optional donation rather than payment for a service or tool you need. That mindset shift is huge. Until someone does it we'll continue in the way we're going.
The number one rule of transitioning to contractor is _do not start from zero_. This can't be stressed strongly enough!! The amount of time and energy it takes to get the ball rolling in that space before momentum kicks in is enormous. Hoping to rely on the goodwill of the anonymous masses, and not leaning _extremely_ hard on your existing direct network is absolutely a failure waiting to happen in 99.9% of cases. If you are working full time, and plan to transition, you absolutely should be moonlighting it first and/or have hard contracts in place with your first 'medium/long-term' client already (i.e. not a one-off engagement).
With six clients one could think, ok it's not so expensive for them, but still sums up nicely for the developer.
Still I think the situation is highly exceptional. Which employer/client would be happy with someone working for them only 1 day a week? And think about the adminstrative overhead for a single person to deal with 6 contracts all the time. With some clients the paperwork can be significant.
The context switching is substantial with 6 clients, even if it's the same tech. As for 1 day per week: a very senior person can give good advice or point to the right direction. The company I work for isn't a tech company, but we do hire people to provide specialist knowledge and it doesn't require full time position to do so.
...even more than that. He's an outlier in the industry and has chosen an industry with access to pyramid scheme money.
Neither the investment bubble surrounding FAANG, nor the job market bubble surrounding the talent pool that FAANG recruits from (namely SE talent that happens to be localized in the Bay Area) were sustainable.
Now that the market seems to finally be correcting away from that unsustainable local equilibrium, he's hopping right into the next one by becoming an open source crypto bro.
> "proves the thesis that it’s possible to be a professional maintainer earning rates competitive with the adjacent market for senior software engineers."
"Prove" is a strong word but "possible" weakens the statement.
"It's possible to earn 95th-percentile compensation." True, by definition, for 5% of all people. Nothing to see here. "There's more than one way of getting there." True. "Honest pay for honest work will get you there." Probably not. "Just seek out the bubbles and jump right in is a reproducible way of getting there." Probably not, you'd have to get the timing right, and that's mostly luck.
For anyone reading this, don't be misled by "open source crypto bro." into thinking the author is a web3 "crypto" developer, he is the maintainer of the go crypto library. Also what do you mean by 'Bro'?, it sounds demeaning. I have met filippo and he is far from being how you're insinuating him to be.
As are geologists who do “proper geology” research that aids the identification of underground oil wells. Yet it’s still relevant to point out where their funding comes from when it’s an oil company.
Pointing out, maybe. But calling OP an "open source crypto bro" and saying he earn "pyramid scheme money" is too much.
Effectively what should be pointed out is "this guy makes some extremely fundamental crypto libraries that are used by millions of projects out there, including cryptocurrencies". But that's hardly relevant.
Does it matter if "the guy is doing proper crypto" if he is getting paid by "pyramid scheme money"? Arguably it is worse since his presence is ostensibly legitimizing the "pyramid schemes". It feels like the techie version of celebrity endorsement
I just use the term "crypto bro" broadly to refer to anyone who benefits, directly or indirectly, from cryptocurrencies, NFTs, and things like that. I do consciously choose a term that expresses the fact that I have a negative attitude towards those things as I believe that they're not "honest money".
Taking Google-money means taking money that's earned through surveillance capitalism and anticompetitive tactics deployed by a monopolist that erode our free markets. Taking crypto-money means taking money earned through "greater fool theory" of valuations of investable assets.
People, on the whole, are never all-good or all-bad. When I see somebody showing off their good sides, I instinctively start looking for the bad. When I see somebody owning up to their bad side, I instinctively start looking for the good.
The good in this person is that he does open source. But that doesn't make him an angel. The bad in this person is that he's a top earner in part because he takes money that causes bad things to happen in the economy. As to his personality, I simply have no information on that and have never met him.
> I just use the term "crypto bro" broadly to refer to anyone who benefits, directly or indirectly, from cryptocurrencies, NFTs, and things like that
Which he does not do. He develops cryptographic libraries (used to encrypt files, network connections and the like). Nothing to do with cryptocurrency at all, save some cryptocurrencies might use the library he writes, but most of the use will be for TLS connections, file encryption etc.
It mentions it as one of the well-known outputs of a company he consults for. I am confident that he is not working on a cryptocurrency at all.
If your definition of "crypto bro" is so broad to include "receives money from any person or company that has ever incidentally done anything with cryptocurrency" you've basically painted the entire industry that way.
Just because it mentions "Filecoin, whatever that is" doesn't imply that he's working in cryptocurrency.
I use "crypto bro" to describe people who actively work/invest in cryptocurrency directly and/or evangelize it. This usage does not intersect with Filippo at all.
Look again at the logos prominently displayed in the blog post. There's nothing incidental here and you don't get to make that kind of money otherwise...
That still doesn't make him a "crypto bro", any more than those companies using cloud providers makes the cloud providers cryptocurrency specific. They require stuff that's pretty universally applicable.
Or change my mind and show me which of his projects is cryptocurrency-specific.
> to refer to anyone who benefits, directly or indirectly, from cryptocurrencies, NFTs, and things like that
You don't?
There's a massive difference between "hey this signature scheme is safe, trust me I'm a cryptographer, all my colleagues agree" and "hey this signature scheme is safe because if anyone can break it they can steal over $1,000,000,000 USD anonymously"
> I do consciously choose a term that expresses the fact that I have a negative attitude towards those things as I believe that they're not "honest money".
The fact that "bro" is a derogatory term for you is also not great.
> has chosen an industry with access to pyramid scheme money.
Seems you can confusing cryptography with cryptocurrency, this guys is a cryptographer, that's a proper expert level security guy, nothing to do with pyramid scheme money.
I’m pretty sure that’s fully accurate. Filippo mentioned one of his backers is the Interchain Foundation [1], and several others of his backers are at the very least cryptocurrency/web3 adjacent. Note, the GP didn’t say that Filippo is working directly on cryptocurrency - but that the funding is likely (at least in part) coming from cryptocurrency profits.
Cryptocurrency is one way of applying cryptography, and the article mentions "Filecoin", whatever that is.
Even aside from cryptocurrency, blockchain, NFTs and that kind of stuff, there's a lot to question when it comes to the ethics of the computer security industry. A lot of it is snake oil, like Firewalls that basically whitelist everything so as not to become annoying. A lot of it is a racket (e.g. you can't get insurance for your company if it doesn't have antivirus software). VPNs basically make money by helping people break the law by circumventing geoblocking. I could go on, but I won't.
I read that comment as referring to the Bay Area startup bubble.
I myself don't refer to anything that isn't paying new customers with old customers money as a pyramid or Ponzi scheme, because I think that trivializes actual pyramid schemes.
But a lot of people do, apparently, and it's completely understandable that a self perpetuating scheme where startups losing money at their core business at a varying rate are constantly sold at higher and higher valuations to see who holds the last hand, is regarded with the same skepticism.
By "pyramid scheme" I meant crypto, not Bay-area startups.
> I myself don't refer to anything that isn't paying new customers with old customers money as a pyramid or Ponzi scheme
In Wikipedia's definition, that aspect doesn't seem to be strictly necessary [1]. They define it as "a business model that recruits members via a promise of payments or services for enrolling others into the scheme".
In my mind it also plays a bit of a role whether you're doing that with retail investors vs. high-net-worth or institutional players. A retail investor generally can't invest in startups, but might invest in crypto if their neighbor recently bought some and then talked them into it.
A person doesn't get any direct reward for convincing their neighbour to buy crypto, though. Compare to a multi-level marketing scheme where the person would directly sell to the neighbour.
Yes, it's absolutely mathematically correct, while being entirely uninteresting when taken in its strict mathematical meaning.
When a motivational speaker says something like "Billionaire X proves that it's possible to be a billionaire" that's mathematically correct, yet totally uninteresting. What people go there to hear about is methods for reproducibly becoming a billionaire or even just slightly increasing your odds of becoming a billionaire, and this article is just as lacking in that department as most motivational speeches.
Don't get me wrong. I think open source is a good thing. It seems like the author is working hard, doing good work, sharing it, and making a solid livelihood may be well-deserved for him. There's just nothing here that suggests a reproducible method.
"... I spend most of my time on maintenance, and I offer retainers to companies that benefit from my work and from access to my planning and my expertise. I now have six amazing clients, and I’m making an amount of money equivalent to my Google total compensation package"
Well done and keep it up! I highly recommend that you find say two other individuals like yourself and form a triumvirate. An individual can have trouble taking time off, dealing with life's inevitable adversity etc.
That can work well for a fledgling org because two can out vote the one who gets it "wrong". This model does require a certain amount of trust and the ability to accept being wrong.
A few people seem to agree with me, so to flesh this out a bit:
Not all of us are going to fly a unicorn to a glorious multi billionaire finish. Most of us would like to be able to sleep at night, have a reasonably comfortable income and basically "get by". As this is HN, I will require a certain amount of "get up and go" but not too much!
I fell into the triumvirate thing 23 years ago. I became an IT contractor at a helicopter factory in the UK around 1996. Helpdesk n that. I learned quite a lot quite quickly and by 1999 ("party like it's") I was in the NETS team - sysadmin land. 2000 - we were Facility Managed off to ... ourselves and our company was started. I was recruited by two other NETS chaps to be the MD of their idea. It wasn't my idea.
So we have three blokes with equal shares of a nascent company. It works really well out of the box. Two can always gang up on one. We would have screaming arguments about direction or whatever but in the end there was always resolution.
23 years later, it still works. The three of us are quite different but we do completely trust each other. The trust thing is where my advice might become unhinged but the model is still very decent 8)
There's a saying, I don't know if its actually old or not, along the lines of: go to sea with one watch or three
the point being that with one watch you just accept whatever it says, with three you pick the closest two, but if you have two... fuck it man who knows: coin flip
If it's old, could it be about watches in the sense of a period of time that one person keeps watch on the ship. So the meaning is go by yourself or as three people. Not about teling the time.
It's about using clocks to compare local noon to Greenwich noon in order to calculate longitude. Thing is that about 50% of all English expressions relate to the British navy and the other 50% are falsely attributed to the British navy so it's hard to say if 1-or-3 was actually real advice.
Navy sailor here. Trained in celestial navigation. The 3 clocks thing is for real. On my first ship we still had mechanical clocks, on the theory that an EMP wouldn't bother them.
Captains have navigated oceans in rowboats to fetch rescue and survive mutiny.
It all starts with knowing where you are and where you intend to go.
Also: Ships aren't neccesarily killed by EMP - mechanical engines still work, they can be tuned by hand, rudders are operated by levers and hydraulics, these can be manually moved, etc.
Yes, but the scenario where you're relying on a mechanical clock because it survived an EMP attack is the scenario where you're getting attacked with nuclear weapons. Which... I dunno, the idea of rowing to safety seems a bit implausible in that scenario?
As for mechanical engines still working -- I would assume there's electronics in any modern hydrocarbon engines, for efficiency reasons (adjusting engine timing etc), never mind nuclear powered ships.
Well, EMP bursts going off overhead is not the same as being attacked by such weapons - the scenario planned for would be maximal operation after an EMP burst.
Dunno about you but I still have my working early model Sun workstation (pizza box years) rated to survive EMP with shielded casing, monitor, etc.
> I would assume there's electronics in any modern hydrocarbon engines
assume .. so, you've never worked on a container ship as a mech engineer babysitting a Wärtsilä RT-flex96C and you think the navy has a lot of nuclear powered ships then?
Have a deep think on this - do you think the US military designs ships to be useless when the electronics go?
No capability for manual weapons aiming, no ability to operate the engines or steer?
We've got a navy person commenting upthread here about having three mechanical clocks for longitude estimation in the event of no GPS .. what do you think that's all about?
early model Sun workstation (pizza box years) rated to survive EMP with shielded casing, monitor, etc.
Obviously, while it might survive some EMP, there's a limit to the efficacy of the shielding. Anything your Sun workstation would survive is trivially survivable by a quartz watch sitting in a shielded box.
you think the navy has a lot of nuclear powered ships then?
My understanding is that the entire submarine and aircraft carrier fleets are nuclear powered, yes.
Have a deep think on this - do you think the US military designs ships to be useless when the electronics go?
I think in a nuclear war scenario, surface ships are already useless (and most likely vaporized) so maximizing the efficacy of their navigation systems in such a scenario is probably not a priority.
We've got a navy person commenting upthread here about having three mechanical clocks for longitude estimation in the event of no GPS .. what do you think that's all about?
My guess is that the three mechanical clocks was more a matter of tradition than efficacy in modern warfare.
> Obviously, while it might survive some EMP, there's a limit to the efficacy of the shielding.
It was EMP rated by US nuclear agencies and used for nuclear test monitoring and radiometric surveys.
> the entire submarine and aircraft carrier fleets are nuclear powered
which leaves a lot of other ships .. and avoids the puzzle of why they wouldn't be able to function after an EMP burst - do you think you're the first to think of such a thing and no one has modelled about such an event?
> surface ships are already useless (and most likely vaporized)
You're assuming that an EMP burst high in the skay also vaporises all ships (or other weapons do), the design of military ships is to assume that they still need to function when damaged to a degree.
> My guess is that the three mechanical clocks was more a matter of tradition than efficacy in modern warfare.
You're hung up on warfare .. navigation may fail for all manner of reasons and three clocks for reference in determining latitude (after allowing for mechanical error) is a matter of sound numerics rather than crusty tradition.
Longitude: The True Story of a Lone Genius Who Solved the Greatest Scientific Problem of His Time by Dava Sobel is decent introductory read if you'd care to guess less and learn more.
Digging old quotes is actually one of the perfect ChatGPT use-cases:
me: Quote along the lines of "go to sea with one watch or three"
bot: "Take one watch or three, but always be on time."
me: Origin?
bot: This quote is often attributed to British admiral and explorer, Horatio Nelson. However, there is no definitive evidence to support this claim. It is possible that this phrase has evolved over time and has been adopted as a quote attributed to Nelson due to his association with naval and maritime history.
Yeah, I was wondering how his kind of arrangement could work if one needed or wanted to take weeks/months off at a time. The traditional model based on features or projects might be a better fit, unless one partners with others as you describe.
I'm so isolated. I never got a degree in this field and I don't socialize much. So I feel permanently trapped working for a company. It sucks, but not enough apparently to get me to change.
> Long term, I want this model to grow beyond me and become a known professional path. This experiment is both easier and harder for me than it will be for those after me: easier because I have an extensive personal network and the financial means to safely take risks; harder because it’s uncharted territory for both me and the clients and because there’s a lack of legal, administrative, and marketing tools. I hope that as things progress the barriers will lower, making the model accessible to more and more people.
I feel this is attitude is honourable and should be commended for its wholesome goodness. I really like your attitude in trying to raise opportunities for software engineers and normalise paying for software maintenance.
Do people really enjoy paying for software?
Do people actually just pay for bespoke development when they pay for software engineers with the silent industry wide acceptance that software is custom created and not for sharing outside that company. I'm thinking of your custom development for a wordpress blog for a small business or an ERP installation for a particular large organisation.
I feel the independent software vendor market for desktop software has stalled. Antivirus is supplanted by Windows Defender, except for Photoshop and some audiocreative software that everyone uses, I don't see the popularity of download websites that there was in the late 90s early 2000s when I was growing up.
I feel, as a software engineer I would like to love a codebase more, to do the things how I truly want to do them, but am held back by financial obligations and for my employer to be rewarded for shipping.
In my observations of internet comments, even software engineers and people don't enjoy buying software packages unless it is an application on a mobile device.
Find a problem people have, especially less engineering-type people, which you would love to help them solve. (Not "solve for them"; it's always a mutual process.)
Wherever people use Excel as the centerpiece of their work, there is an opportunity to help improve things, for fun and profit :)
In general, most opportunities lie on the seams between well-understood areas. Knowing more than one area helps. Many of these opportunities are too small for hockey-stick growth which VCs crave, or for huge contracts which large corporations desire. They are perfectly sufficient for a mid-size sustainable business though. Specialized things, like the bespoke work on and around open-source software from the post.
> which proves the thesis that it’s possible to be a professional maintainer earning rates competitive with the adjacent market for senior software engineers
Yeah, if you're Filippo Valsorda. Not sure that "proves the thesis" broadly whatsoever, though.
I think the main caveat is “Senior” and he means fairly senior at that anyone less will definitely have trouble, but if you’ve been in the industry 10-15 years and are looking to be your own boss and are very competent, I don’t think it’s unreasonable to do this and make comparable compensation.
As someone with more than 15 years of experience and who has leveraged my open source work to improve my own career, my two cents is that GP has a fairly sober view on OSS as a job: Filippo is an outlier among outliers. Most people simply don't have the opportunity to build a reputation from multi-year, full time, corporate sponsored open source work on a high visibility project to leverage when leaving a cushy job to pursue higher rungs in Maslow's pyramid.
Yes. This is a variant of building an OSS project using company funds then leaving the company to commercialize it. It's a real path, but requires taking advantage of years of corporate largess. Not impossible, but some big "if"s involved.
Hahaha. Not every engineer is a salesman who can close a deal. Contracting is a cutthroat business as well where the competition is fierce. It’s global competition too!
I wouldn’t recommend this route to any random senior engineer.
Wow, the footnote about dentists surprised me because I just found out about this company [0] (and apparently it’s one of many[1]) that just stamps out pre-built dentist offices all over the US with all the equipment and staff included.
All they need is the dentist. It’s honestly super creepy how generic they all are and how many middlemen end up involved in your dental care.
Wow, TIL about the Association of Dental Support Organizations.
Now I'm imagining a future where independent open source maintainers are common and successful enough that there's a conference for the Association of Open Source Maintainer Support Organizations.
This article is fantastic. Really great to see how Filippo is getting this to work, and in so much detail.
As someone who frequently complains about all of the administrative overhead this kind of lifestyle requires I appreciated the footnote pointing out that dentists have the same problem and just get on with it.
That said... as a self-employed mostly full time open source maintainer myself I do often think about how much I would appreciate some kind of agency or talent management relationship that would take a bunch of that off my hands. It works for Hollywood, why can't we have that in tech too?
I think it means who is charged. Top is the customer and bottom is the supplier.
If you charge the "bottom" then that might be passed up to the "top" or not. It depends what the "middle" does. The middle might absorb the cost or pass it up to the end customer.
They provide hardware, mailing lists and legal support. They don't help match-make maintainers with sources of consulting income (which is the kind of admin I'd most like help with!)
So it's kinda like a music tribute group but in the software world? Substituting for something that may be gone (the original FOSS author(s)) but for the die-hard fans who don't want to let go?
2 out of 5 of his clients are blockchain companies/orgs. I feel this path would be quite easy in blockchain, where almost all of the projects are open source
From the number of logos, it's 6 clients, and I would put 4 of them into the blockchain bucket, with smallstep and tailscale being the two non-blockchain exceptions.
Also note that for blockchain projects, it's important that they associate themselves with famous people to give their projects credibility. So they are willing to pay huge amounts of money to get celebrities like him on board.
On the 2nd statement, quite curious if that's true... if so, each of the companies will have at least a blog post, or tweet thread about the affiliation. If not true, you're probably reducing who is an efficient engineer who's worth their salt for their output.
You're thinking of it the wrong way, think of it as a banner ad.
If your open source project has a homepage, a fair amount of visitors, and a public list of sponsors, blockchain companies will pay for the highest tier to be on top of that list instead of going down the usual AdSense route.
From their perspective it's a link from a respectable source that reaches their target audience (those into tech) on a permanent basis that even adblocks don't block. And it only costs them up to a couple of hundreds of bucks per month, way cheaper then traditional banner ads. Doesn't matter if what you're actually building has anything to with cryptocurrencies, but of course having some touching ground works even better.
Whether this has happened for him in particular, I don't know. He at least has made a blog post with the links of the companies in it, but of course this is only indication, not proof. The trend is certainly a thing. Blockchain companies / NFT projects / etc live from attention. They need it for their growth.
That's because they must be open source, what other sense of accountability could you possibly give if you're asking money to a crowd upfront based on a piece of paper?
Honestly if you have to be a reasonably-well-known public figure to pull this off then I’m not sure how accessible it really is. I’m reasonably well known in my own city, and I could maaaybe pull off this level of comp working on strictly for-profit initiatives? Very feast or famine though, to the point of not being worth the stress.
That's great but his experience is totally unique. My side project is used by at least 10 companies. Some of them are tech giants. My total earnings are -300 USD and 5 years of development time.
It's Olric: https://github.com/buraksezer/olric. Publicly speaking about the companies may not be a good idea but you can dig into the issues, pull requests, and Discord channel if you are curious.
This is awesome for Filippo. I am curious what the average contract length will be (or is). 1/3/5/10yr? Essentially will there be enough time to maintain and not be searching for new contracts to cycle in? If it is indeed like enterprise sales, it takes a long time to source, negotiate, and close a contract. Hopefully we can get an update a few years down the road.
I suspect there's a sideline wink and handshake agreement to feature associated companies and indirectly promote them as case studies, etc. as the maintainer maintains. Not that that changes anything this is still a great arrangement and hopefully a positive example for our industry, but Smallstep and Tailscale are both trying to gain an authoritative hold in the production identity space and a little name dropping here and there by <famous dev> certainly doesn't hurt (:
There are a handful of these companies out there. Collabora, KDAB, D. Richard Hipp's SQLite company, and if you're lenient enough you could also include companies that productize their open source package such as NextCloud or The Qt Company. Heck, Red Hat (sorry, IBM) regularly gets pulled in to improve software across their stack, kernel or otherwise.
You could also include non-profits like the Linux Foundation, Blender Foundation or (new!) Godot Foundation, which do the same thing but without having to disguise as consulting, because development of the software itself is important enough to the industry that can pool its resources by donating to the respective foundation.
Still only works for important enough packages. I don't think there's a way around that. An open source project generally has to provide massively outsized value so that a handful of developers can capture a fraction of that value for paid maintenance.
Speaking for KDAB, only a relatively small fraction of our overall revenue is paid work on open source codebases. We mostly get paid to work _with_ Free Software and to teach it to people. Living off of maintaining or improving Free Software is only sustainable for freelancers or small boutique companies. Even the big foundations don't fund all that much development, they mostly focus on governance, infrastructure and promotion. The Linux kernel is possibly an exception, in terms of feeding an appreciable number of people working on it professionally outside of a company directly selling it.
How is this exact model different from regular freelancing? Taking on 6 clients at once is a bit much for my taste, there’s a lot of overhead involved usually. Now you not only need to manage various OSS communities, but also multiple clients and their expectations towards your contributions.
They're not paying him to do custom freelance work for him. They're paying him to continue contributing to the projects that he is already maintaining.
The higher tier plans also get what sounds like a few hours of custom consulting time (think calls with their team) per month.
> it boils down to this: I go in, meet the engineers, and learn what parts of my projects they use and how; then, I keep those use cases in mind in my own planning and I reach out and involve them for feedback when there are relevant changes on the roadmap. This improves outcomes for everyone: I want my projects to work well for users (regardless of whether they are paying me) and no one wants to find out something’s wrong after the release.
In the very first paragraph of the linked article, there is this clause which may provide some clarity:
> I’m making an amount of money equivalent to my Google total compensation package
He goes on to write,
> My Google compensation was dominated by stock grants, so it varied wildly. What I’m earning now per year is: slightly less than my first year at Google (which included a significant signing bonus), drastically less than I earned in 2021 (when the signing stock grant overlapped with three stock refreshes, and $GOOG was at record highs), and more than I would have made in 2022 had I stayed (even accounting for all benefits on one side and the salary of my assistant on the other side).
> My Google compensation was dominated by stock grants, so it varied wildly. What I’m earning now per year is: slightly less than my first year at Google (which included a significant signing bonus), drastically less than I earned in 2021 (when the signing stock grant overlapped with three stock refreshes, and $GOOG was at record highs), and more than I would have made in 2022 had I stayed (even accounting for all benefits on one side and the salary of my assistant on the other side).
That is vague. A specific number would have been much better.
According to levels.fyi [1] as of Feb 2023, a Senior Software Engineer at Google (L5) could negotiate a compensation package around USD $193k (base) + $119k (RSU) + (one-time) $26k signing bonus.
I have talked to Filippo in person and I have the impression that he was a Staff Software Engineer (aka. Google L6), which are supposedly able to negotiate around USD $244k (base) + $187k (RSU) + (one-time) $35k signing bonus.
These numbers are averages based on verified offer letters and RSU grant documents that Google employees have submitted to that website during the last couple of years, but I think the numbers do not account for annual RSU refreshers, performance bonus, or stock performance. That is what Filippo is referring to as “it varied wildly”. I still would have liked to see a comparison with real numbers, at least the base salary.
So is he making "slightly" less than $244k + $35k as a fulltime open source maintainer? That's awesome, wow, not everyone can pull it off so this is pretty impressive!
> These numbers are averages based on verified offer letters and RSU grant documents that Google employees have submitted to that website during the last couple of years, but I think the numbers do not account for annual RSU refreshers, performance bonus, or stock performance.
It's complicated, but I think they account for it in a way that gives a reasonably accurate picture in the steady state. levels.fyi for L6 says salary $244k, stock $187k/year, bonus $35k. Every year as a Google L6, I got a compensation letter with roughly that salary, equity refresh, and annual bonus. Each equity refresh vested monthly over the next four years. That means 4+ years in, you receive over the course of a year stock which was valued (at various times in the last four years) at about that equity refresh in total. Actual value when you receive it varies much more because that's what stocks do. And letters can vary more from year to year based on performance multiplier, you can get spot bonuses in addition to the annual bonus described in the letter, you may get promoted to level n+1 (or leave) before reaching level n's steady state, etc. On the flip side, it seems likely that along with the layoffs this year they didn't give anyone big annual bonuses or equity refreshes. I moved on a while ago so haven't asked.
Anyway, it's good money, and I'm really happy to see someone able to match it as a full-time open source maintainer. Filippo seems pretty exceptional though; I hope good open source compensation becomes normal.
I don't think that's the message here. You don't /need/ that kind of money to work full-time on open source if you're happy to keep to a lower cost of living.
The biggest cost involved in full time open source work is opportunity cost.
If you have the skills to maintain a popular open source project, you could almost certainly be earning $250,000/year or more at one of the big tech companies.
His experience is totally unique. Please don't let this make you think you can quit your job and earn the same you did at your high paying FANG job. He's an outlier in the industry. I was a nobody when I left my job to work on open source full time. It was a year before I found a corporate sponsor and that was only because my friend worked there and he understood the value of what I was working on. Patreon was laughable in terms of what came back. Just be fully aware as you read this. With existing brand and following you can do what he did, without it you'll struggle immensely like I did.