I sort of feel like an out-of-context Twitter comment from several years ago is not the optimal way to start an HN conversation about recruiting math, but I would like to emphasise that the tech industry heavily underinvests in funnel development for recruiting and that outsized returns are quite possible with reasonable amounts of effort.
If you want to hop in the time machine and steal what was working wonderfully: back in 2018/2019 we'd routinely run Remote Coffees, which were just "get a few engineers and EMs on a Zoom call to chat in a semi-directed fashion about what they work on daily", and I'd drum up a few hundred attendees for those coffees every month, and then observe how the funnel performed over time.
Important to scroll up and see that this statement is made in the context of Dan Luu discussing how expressing an interest in writing can be damaging for your job prospects, regardless of your other merits/skills.
Patrick is chiming in on how his public writing may have greatly benefited Stripe.
Specifically, the context seems to be 7% of people joined the team due to his writing, not that he is worth 7% of the team. Others in the thread seem to be reacting negatively based on the second interpretation.
It took me a while to work out that he was saying "7% of recruitment it due to my work" not "I am so amazing I am worth 7% of Stripe's massive engineering team".
Linking to what tweet out of context is probably going to get Patrick a lot of undeserved hate.
Twitter's market share has only shrunk, and their user base has been flat since about 2015. It is "popular" with a very specific group of people but I don't think they've ever topped 350 million human users at any point in their existence, which while not small, but they're not growing, either.
It took me about 30 seconds. It was obvious to me that the tweet was missing context so I had to read the thread. Which wasn't very long to begin with.
The UI does not clearly present the prior tweet in a conversation when linking to an individual tweet. This makes it easy for people to ignore the context since they have to work to see it. Combine this with the preference for using real names and it is easier to make the "look at what an asshole this person is" posts.
Early in my career I did some bouncing between tech writing and programming gigs (my undergrad degree, after all, was in English and I have close to no formal education in c.s.), and a big part of why I ended up staying on the programming side was that it paid way better than the tech writing did.
The general context is hiring managers under-/dis-valuing writing ability for technical people, and Patrick countering that a significant part of the value of their team is him , who is known for his writing.
This is shocking to me, as I highly value good communication skills, and I assumed my industry did as well. For example, I think the main reason Rust is as popular today is thanks to Carol Nichols' & Steve Klabnik's excellent writing, and Steve's unrelenting advocacy (be it talks, blog posts, or online comments). The impact of a technology is nothing if it can't be communicated well.
I've been working in improving my technical writing, contributing significantly to our team's docs, and I have more respect for my colleagues who do the same.
Have other people experienced this scorn for writing that Dan Luu refers to?
> Patrick countering that a significant part of the value of their team is him , who is known for his writing.
Just to clarify, I don't think he's saying that 7% of their value is him specifically, I think he's saying that 7% of Stripe's engineers are there because of his blog. He's replying to Dan Luu's note about recruiting fees and how valuable a popular blog can be.
Have other people experienced this scorn for writing that Dan Luu refers to?
Sort of. At a point >10 years into my career I had a junior dev ask me to remove my method comments during code review because they might become out of date some day.
I personally downplay any writing/organizational skills I have, as it always seems to count against or be used against me. It means you get stuck with non-technical work/documentation (which everyone wants but nobody values), are given errands to review copy for the product manager.
With no leadership aspirations, that is career incoherent work for me.
Maybe, though it can also be a way of learning. You crystalize & organize your thoughts with words. Regardless, the aphorism is old, tired and just plain wrong. Both writing and teaching are amazingly valuable skills.
thanks for morally correcting me (and to the others who downvoted), however as the comment I replied to asked if anyone had ever observed that kind of contempt for learning in other contexts it would be a reasonable interpretation that I was providing an example I had observed. There was nothing in the comment that should have caused anyone to think I was a true believer in the aphorism.
> There was nothing in the comment that should have caused anyone to think I was a true believer in the aphorism.
There was nothing in the comment to indicate that you were not, either. It wasn't in quotes. Especially here, people take what's written at face value. I guess that's a communication lesson.
> There was nothing in the comment that should have caused anyone to think I was a true believer in the aphorism.
Even so, saying "Writing is a form of teaching." needs more justification, especially in this context. Leaving that half-connected and confusing invites a downvote or two.
Alright, if you'd rather insult me than put any elaboration onto how "writing is a form of teaching", that's your prerogative.
To be clear, it's easy to see how your post connects to the post you're replying to. The problem is that internally, your post is not fleshed out at all, and it's not clear how you came to the conclusion that writing and teaching should be interchanged in this context.
I could see this. Patrick has a really strong brand among engineers. So if he’s willing to lend his personal credibility to Stripe, that’s a strong signal that it’s a good place to work.
I formerly worked at Stripe -- last day was less than a month ago.
In summary, culture is downstream of hiring decisions. If a company is doubling headcount each year, then after two years 75% of the population will be new. If the newcomers' culture is different from what old-timers prefer, then attrition increases, which further skews the demographics.
This effect is strengthened when the hiring pipeline is focused on alumni from specific companies. For example, if you hire three layers of management from Amazon, and they replace tenured engineering leadership with newly hired ICs from Amazon, your team's culture will become that of Amazon.
Some blog posts I've found to be relevant to current circumstances:
I’m also quite interested in hearing about this. I used to only hear good things about Stripe, but over the last year, there’s been quite a few negative posts. I’d be interested in more details.
It doesn't intrinsically mean it's a great place to work.
The idea that Patrick works there and thinks it's a great place to work at is "lending Patrick's personal brand to stripe". If you have a lot of trust in his judgment (and thus his brand), you're more likely to think that "hey stripe is probably a great place to work" and then apply there, because if it Wasn't then why would he still be working there? Or why would he be endorsing it? It also puts his personal brand/reputation at risk; if Stripe Isn't a good place to work, that hurts his brand
Chiming in to say I am also shocked that Steve Klabnik was so dramatically underpaid at Mozilla. He did amazing work with an incredibly difficult job (essentially sell a new systems programming language to the most cantankerous engineers around, and do it happily with deep technical knowledge). Great move by Oxide to hire him.
I've been following Patrick for years (from the "have coffee with me in tokyo" days) and can attest that him joining Stripe makes it very attractive as a place to work.
Not that you shouldn't write blogs, but bear in mind for folks trying to split SWE & blogging, your manager will see your split priorities as a risk. Every slipped team deadline paints a bullseye on the hobbist blogger who has to spend time with legal getting approvals. Every public mistake you make will end up on their desk, with the expectation that they somehow fix it. Even if you do well, if the rest of the company finds out they'll ask you to do more writing and less of the thing your manager was hired to lead a team to achieve, without any formal headcount change.
Generally speaking, blogging increases your personal profile more than it increases your company's profile, if any. I can name several HN famous bloggers, but at this point many have job hopped frequently enough IDK who they even work for anymore. Even when you hire someone like Patrick who's already a celeb, the recruiting pull only lasts as long as his employment. (I imagine Stripe is fine since there aren't many companies willing to hire remote workers in Japan while paying SV wages. YMM very much V).
I really expect most tech bloggers to fall into one of three camps:
1. Consultants, who need to build a name brand around their niche expertise. The typical consultant formula is spending 30 percent of your year working for clients, 30 percent finding new clients, and 30 percent skill building.
2. Product Evangelists who need to get people excited about a tech platform their company profits from.
3. New grads whose career hasn't taken off and consumed their free time yet.
I maintain a personal “thought journal”. Writing, primarily, is a tool for the mind. To explore lines of thought, brainstorm with oneself, explore lines of argument and so on.
Without writing thought process is single threaded and linear. With writing it becomes multithreaded and non-linear.
He thinks 7% of the engineering team works there because of him? I wonder how important he thinks compensation, work satisfaction, and the rest of the team is to engineers’ decisions to work there.
Speaking for myself, his blog is responsible for everything I know about fintech, and he makes it sound really interesting. It would not be at all surprising to me if 7% of Stripe's engineers only decided fintech might be worth pursuing after reading his blog.
Having briefly worked with him, he struck me as very much the opposite. And based on how well known and respected he is among engineers, I could believe what he's saying here.
I think it helps to take him at face value, serving the sharing of information he sees as valuable. Maybe it's easy to misread the phrasing as grandiose, but this hasn't changed about him for years. People recognise his name not because his side projects were insanely successful (they were quite pedestrian!) but because he wrote about them.
When he was writing about the bingo cards, I had side projects making more money for far less effort, but people knew about his while I didn't write about mine. I like to imagine the tone was (and still is) less about "Look at this amazing stuff I'm doing" and more "I'm amazed at doing this stuff."
The takeaway from his tweet likely isn't intended to be "Look how great I am" but "Here's an example of the power of writing, might be useful to consider". Presented with minimal tone, maybe people misread the intent.
Where to begin? First, let's start with the most specific. Patrick claims to have written almost 3,000,000 words, which is almost 3 times the entire Harry Potter series. I'm not passing judgement on him. It's great that Patrick can write so prolifically, but such an undertaking can only be described as Herculean. Success in writing for the public typically follows a power distribution. Using words written as a proxy for popularity, it can likely be assumed that impact will be exponentially lower for someone who writes less.
Next, let's get a little more general by talking about opportunity cost. Assuming Patrick's writing really is responsible for 7% of Stripe's engineering team (which is a huge assumption), what would happen if Patrick had never written anything? Maybe the company would flounder and die, or, more likely, engineers would still be attracted to a high growth startup and Stripe would largely be in the same position.
Finally, I'd like to make a statement about writing in general. The goal of writing is to convey an idea. What constitutes an acceptable conveyance of an idea is context-dependent, but generally you want to convey an idea for a reason, e.g. you may want to convey the design of a system you're building. In these cases, creating the idea to convey (e.g. designing the system) is so much more important than the actual conveyance of the idea (e.g. explaining the system), unless you're so bad at explaining that people are hopelessly lost after reading your writing. Most people know when they understand something and know when they need clarification. Humans are consensus builders by default, which makes ok writing and great writing equally good in many contexts. The person who started this discussion may have been offended at the insinuation that they can't program, but I largely agree with such an assumption. Writing is a tool ultimately used in our profession to aid in the building of things. To focus only on writing means you're not focusing on building.
If he is at a team of 20-30 people, that's an easy figure to defend. But when stripe has over 800 engineers (maybe 400 at the time he made that tweet), that seems some kind of major exaggeration.
He's suggesting that his writing (HN, etc) is likely to have encouraged a certain number of people to join Stripe. That's the only context; he's not suggesting anything about effort beyond that, just talking about the value of writing.
He might be going by interacting with team members and a couple dozen saying things like "I got interested in Stripe because of you talking about it" or "I was reading your blog for years, and tipped me towards joining Stripe instead of [other offer]."
The Twitter UI is terrible, but if you scroll up, the context is writing and recruiting. Patrick's claiming that his writing has led to a lot of people applying to Stripe and getting hired, and that seems true given how popular his posts have been.
If you want to hop in the time machine and steal what was working wonderfully: back in 2018/2019 we'd routinely run Remote Coffees, which were just "get a few engineers and EMs on a Zoom call to chat in a semi-directed fashion about what they work on daily", and I'd drum up a few hundred attendees for those coffees every month, and then observe how the funnel performed over time.