Well written article full of humility and vulnerability? I love it. My reaction: you don't need to feel ashamed of not knowing something, there is far too much to know and I'm still learning new techniques and concepts 37 years in, so I would never judge you for it.
I would also not judge you for having your own preferences and opinions. I too prefer working in an office to remote work, but when I say this out loud other developers take it as advocating RTO or saying remote work is worse when it just doesn't suit my personality. I get that it's a touchy subject but there is no need to get up in my face about it.
You mention bullying and brigading and that seems to be an unfortunate reality of this industry. I suspect there is a lot of insecurity and imposter syndrome that causes people to write hyper-confident blog posts about why they are better without AI and how their tests have 100% coverage and how (unfashionable language which half the world uses) is garbage etc. Maybe if we all follow your example and be candid everyone could chill out a bit.
I'll go next: despite trying several times, I have never successfully written anything more complicated than Fibonnacci in Lisp or Haskell. I know it's clean and pure and all that, but my brain just won't work that way.
> I have never successfully written anything more complicated than Fibonnacci in Lisp
I hadn't until a joined a lisp based project. Learned a ton. My brain didn't work that way at first either, but working with it every day I eventually got it.
Every time that I read this about remote work, all I can think is how much I miss IRC and the culture that came from it.
We were doing remote work effectively decades ago. Don't have hallway conversations to fix bugs? Easy, just post your problems on the team chat and someone (often one of several people) would love to drop by to help.
I'm not sure exactly all of the forces that have led to this changing so much, but I'm certain that merely blaming "remote work" isn't it.
Somehow we were better at using remote tools while literally in the same office than some teams are at using them now while fully remote.
I have encountered people who are scared to post in large public channels. Part of growing up in chatrooms was an implicit bravery of saying something out loud in a room full of thousands of people. There seems to have been a shift, somewhat, in the comfort level of different generations about saying things "out loud" in large public rooms.
Chatrooms have evolved in a really interesting way. I think the first generation to have them didn't fully understand how "public" they were. Maybe there are more people in the more recent generations that have a more visceral understanding of online "publicness" as they have grown up with (and perhaps have been burned by) those concepts from the very beginning. Maybe they have a better understanding of the permanence of online utterances and therefore have a more conservative approach to interacting on what feels like the permanent public ledger.
Maybe it's because the concept of pseudonyms has devolved since the early days. Corporate social media has an interest in doxing its users to advertise to and control them but pre-corporate social media was filled with anonymous usernames. Posting in a large group under your permanent forever name is much scarier than posting under an anonymous, temporary identity. One of the things I advocate people do is post online anonymously, instead of with their real name. It alleviates a lot of the fear of speaking your truth, which we need more of!
There is something there. The ability to try on identities in a safe environment before you discover which one you really identify with. It's much harder to do this with your real name. Your past comes with a lot of baggage and people who know you don't want you to change because it makes them feel uncomfortable.
There's an issue of scale to it all, and interconnectivity. Back in '98 you could reasonably post something in one forum/BBS/IRC channel and it would only be viewed there. There was no way to look up who was on what website or room and where they regularly hung out. And even if there was, it's unlikely that more than ten people would ever see what you posted. There weren't enough people to care, and there wasn't any extrinsic incentive to look up what people did outside of your tiny island. Eventually the island would sink, and all traces of it except maybe an archived snapshot of the home page would remain.
Smartphones changed that with Youtube and Facebook. Youtube incentivized you to use a Google account, and Facebook wouldn't let you use it anonymously without an account. Because you could use one account to log into multiple places people could track you across websites. People could make archives, screenshots, and transcriptions of anything you had done with those linked accounts. With this change there was no safe corner to hide if you said something stupid. And because so many people were foolish enough to tie their real identities to these online accounts with their real names or pictures of themselves, it gave a way for particularly unruly people to track these individuals even offline. There was now a real danger if you said something stupid, because instead of just getting your post deleted or starting a derailment in the thread people could harass you at your home, get you fired, and even send the police to terrorize you in the middle of the night via SWAT raids. It's no longer just one person calling you out. It's now hundreds, maybe even thousands, all armed with information.
And this is why I say it's stupid to require phone numbers and real names to sign up for insignificant things like being able to view someone bake a duck shaped cake live over the internet.
> There seems to have been a shift, somewhat, in the comfort level of different generations about saying things "out loud" in large public rooms.
I think this is merely the shift from doing this as a hobby, to doing this for work. Random coding problems mixed with banter I posted or answered on IRC back in the day? Purely hobby stuff, things I done after school instead of doing my homework. No stakes beyond the community itself, I could disengage at any moment, nobody would care - there was no commitment of any kind involved.
Today? Even if we switched back from Slack/Teams/whatnot to IRC, the fact remains, the other people are my co-workers, and we're talking about work, and it's all made of commitments and I can't disengage, or else I starve.
I use pseudonyms and post weird shit online, but I still feel very reluctant to post anything on large public channels inside a company. Everything is tied to your real name and all of us are hyper aware now that every single fucking thing on the internet is monitored, and will be used against you if necessary. I am 99% confident a tool already exists that a manager can use to get all messages from employee X over the last N months and summarize the content and surface any "red flags", which in a corporate setting would be incredibly broad.
The problem in my opinion is that folks afraid of posting treat chat channels like email and official record instead of a conversation. I like to post ideas, brainstorm, engage if I have a minute to respond to someone with a thought - kinda like being in an office - whereas many others seem to use it to blast out information after a lot of polish and they form a culture of announcements and no engagement, and get stressed whenever someone asks a question or actually replies.
Use tools for what they are good for and create a culture that makes each tool work best for your organization.
Oh lord no. I’m not sure that’s true at all! I mean in the early cool era of the internet there was a concept of a “netizen” and a window of brief joy, but the internet has always had domineering trolls, bullies, spaces where clumsy newbies were brutally flamed etc.
The main difference is that more spaces were quasi-professional and non-pseudonymous, in that one largely got one’s internet access and identity (IP address, email address, invitation) from the institution of higher learning one attended or worked for. So there were direct, two or three degrees separation consequences (my boss knows someone at your institution) in those spaces. I suppose this is what you are referring to.
(In my early era of commercial internet work I can remember a colleague shutting down an accidentally abusive scraping bot by working out who was likely to be the boss of the person running it and phoning them up)
But away from those spaces were many places that were just as bad as they are now.
The internet has always (in my time of using it, which is all of my adult life as someone who is over half a century old) demonstrated that a good culture is a question of starting conditions and quick maintenance actions.
A non-trivial amount of the worst behaviour I have personally witnessed on the internet happened before the year 2000.
I got a minor reprimand in a performed review for having a slightly heated conversation in a public channel. In the past that company had been very open about communicating about negative stuff (our CEO emailed the entire company when he visited one of the hardware labs that was a mess), but the upper management started tamping down on anything negative, and one of the things that suffered was any sort of honest communication.
Kind of a ship of theseus situation culture wise - when the original leaders are all gone, did they pick good successors to fill their spots? Very often not.
This sometimes happens also when the original leaders are still present but they dont understand the effect the metrics are having on the entire company when it grows big.
Have a senior leadership team and want them to not tell you bad news when you are the CEO/Leader? Then link their salary/performance to metrics like number of production incidents their team has. Suddenly the number of incidents that you know of decreases.
If that does not work to isolate you as the leader from thr reality of your company then link their salaries to a metric like number of projects finished before or at deadline and watch how tech debt increases multiple folds and how everything is suddenly estimates are increasing all over the place.
Want people not to ask meaningful hard questions in All Hands? Just make sure anyone that seems critical be labeled as not culture fit and done. All questions are positive and nice. Make sure to always ask for name and disable any anonymous questions asked.
Not trying to say metrics are bad or they should not be used. But they are not pure functions :) they do have side effects and sometimes very large ones.
> Somehow we were better at using remote tools while literally in the same office than some teams are at using them now while fully remote.
I couldn't agree more. I pushed to get the place I worked for to use Slack when it first launched, moving us off AIM (ha!). Our use of Slack when we shared an office in the twenty-teens was so much better than the use I've seen of Slack/competitors on fully-remote teams.
I wonder if it's because the failure mode was, as you said, to "drop by." Now the failure mode is... just failure.
Perhaps the youngins are more cognisant that it's all monitored. Knowing your employer can read everything and it _will_ be used against you has a chilling effect and I'm pretty sure that's part of it.
We had such incredibly heinous group chats on our Slack that if an admin perused through the logs we'd be instantly fired and the company shut down right then and there lol. The paranoia drove everyone nuts which made it more fun.
They, for every team I’ve ever managed, have an off company owned systems chat on shit like slack or discord where they are roasting the fuck out of you.
I’ve managed to be invited or told of them after ingratiating myself to the teams, or more often, after quitting and getting invited as one of the “good ones”
They all know that every word on company shit is being monitored
Sure, but this ends up poisoning any sort of culture and creating all sorts of in-group nonsense which is almost impossible to undo.
It’d be like using Blind as your company chat - nobody goes on there to say how great their experience has been, and the tone infects everything else.
But maybe I’m just not very fun at parties…
This should be avoided at all costs by creating a culture that is receptive to people’s concerns and doesn’t do stupid things without explanation - but I get how difficult that is in reality and most orgs end up messing this up.
Maybe I'm a bit unfair to you but to me your comment basically reads as wishing employees would be good little cogs in your machinery rather than people. Like making friends is natural human behavior. Forming friend groups is natural human behavior. It's not nice to disrupt this except that of course everyone has to be able to work together when needed.
I’ve never worked _anywhere_ where reporting someone to HR was anything but negative impacting for your prospects at the company. And I’ve worked at lots of places in many dimensions (company size, industry, age, etc)
Is that a dynamic they have? I haven’t worked at Big Bank but I’ve worked in finance a few times and at those places and other industries I’ve worked in reporting anything to HR wouldn’t necessarily get direct consequences but you would permanently be on their radar and have to work to rule after that
I think it’s more a shift in the culture amongst most people now than an argument of remote/in office.
Notification fatigue is a thing and people are just used to ignoring notifications and messages nowadays which ends up with slow responses and poor communication all around.
It's much easier to get a feel for how urgent things may be when you can yell down the hall.
I'm currently in feeling things out phase with my current team, and people seem really laid back about responding to messages - but it also seems like we're getting stuff done. Hard to figure.
> how much I miss IRC and the culture that came from it.
IRC selects for people who like chatting and communicating via text.
I think the mistake made with remote work was assuming that everyone could easily work that way.
The best experiences I had with remote work were pre-COVID, when the teams working remote were carefully selected for having good remote work abilities and anyone who couldn’t handle it was kicked back to the office (or out of the company)
Then something changed during COVID and remote work was treated as something everyone could do equally well. The remote teams I worked with were now a mix of people who could work well remotely and people who wanted to work remote but tried to force communication to happen like we were back in the office: Meetings for everything. Demands to “jump on a quick call” when a few Slack messages would have done the job. Then there were the people who read “Four Hour Work Week” and thought they were going to do their jobs from their iPhone while traveling the world or at the ski resort.
I don’t know. Having seen the before and after it doesn’t feel so surprising that remote work faltered when applied indiscriminately to everyone. The best remote teams I work with to this day are still the ones who know how to communicate in that old school IRC style where communication flowed easily and everyone was on the same page, not trying to play office games through Slack.
It's a strange pattern I observe often, whenever an idea gets promoted from organic-natural-human-ritual to official-new-visible-main-idea, it becomes bloated and off point.
Yeah. Random example: I have better "ambient awareness" remotely because with slack I am in every hallway simultaneously, and can skim the conversations and set up highlight words
I wonder how much of that comes down to culture. Since going remote I have come to wonder if a direct-message-first chat culture is harmful to collaboration.
IMHO most companies encourage public-first conversation, but still end up with DM-first as their employees don't have enough trust in how their messages will be received.
It requires to be comfortable exposing lack of knowledge or saying weird things to peers, and be confident it will be taken in good faith. As you point out, that requires a whole level of culture building.
DM-first is an extremely frustrating culture. That kind of operation tells me that that folks are too risk averse and political to discuss things openly. Typically this is led by panicky managers that are worried about involving too many people or having to explain things to folks they don’t want to deal with, and it escalates from there and gums up ALL the things. It makes Slack basically useless.
The same people DMing however will also extol the virtues of posting in public and lament why there is not more conversation happening in the open.
It is. You need to be aware of it and have people that can set examples about chatting in public rooms or who can recognize when to stop a dm chat and move to be public
I started doing remote back in 2002, working with companies in the UK and US, something that was largely unheard of in my country. While our mode of communication was web messengers (AOL, Yahoo, MSN etc and the aggregators), I used IRC a lot to get tech help, and participate in online dicussions.
Right, because there's nothing special about remote vs in-office. It's just communication and collaboration.
Living beings do it all kinda ways. Bees waggle their butts, crickets rub their legs, geese honk, snakes hiss, some fish detect electrical signals. And to collaborate, the bees' dance indicates a flight path, birds singing indicates interest in mating, the snake's hiss and the geese's honk tells you to watch out. You use the tools you have and develop collaboration with them. There's clearly no right way, there's just ways.
But tomorrow morning, would you wanna learn to honk at people, or rub your legs, or waggle your butt, to order a latte at Starbucks? It'd be awkward, weird, painful, and unnecessary. So if you were asked to, you'd probably not try very hard to adapt to it. And if everybody you knew were in the same boat, all being forced to change with no real guidance, kinda not trying that hard to make it work? It would suck for everybody.
People just don't like changing what they're used to. They probably don't even mean to fight it. But we do like culture we're already familiar with. Change is hard, not changing is easy. We like easy. So people who grew up with remote work (on IRC, mailing lists, etc) find it easy, even more productive. But a company that's thrown into it without a healthy established culture are going to be swimming upstream indefinitely.
"Everyone is different" is my point. Face to face once in a while is nice, though I don't care so much about having it very often with my colleagues in a professional setting.
Of course it's not the same, that's the point. I personally prefer the async, chat based mediums I've used since I was a child. Some of my coworkers have disabilities that make conversational typing difficult and prefer video calls as a result.
Consider the effort to accommodate those preferences though. Accommodating a video call preference is easy. Same for chat. Accommodating a preference for face-to-face requires spending an hour (2x average US commute) traveling to meet you. That's quite a significant ask of the other person.
In electronic chat I can ask someone to explain their question and wait for it in writing. In person, I often have to listen to them stumble over the concept because they didn’t think about what they wanted to ask before asking it.
In a video call I can clearly see the other person’s screen and zoom in on what I’m trying to look at. In person I have to lean over their desk and squint at the right angle.
> Somehow we were better at using remote tools while literally in the same office than some teams are at using them now while fully remote.
This is sort of the point. Remote tools work great when you have spent a lot of time building relationships and rapport with the people involved. That's hard to do in professional settings, and extremely hard to do in remote professional settings.
Letting teams that know each other well work remotely works great. Building teams remotely is very hard.
I'm a diehard for remote work, but we have to be realistic abouts limitations.
IDK, some of the remote teams that I've worked on were only able to meet in person once per year, if that. They were very communicative on the tools that we had though.
No one is saying it’s impossible to build fully remote teams from scratch, it’s just very hard and requires strong leadership. Most companies have crappy management so they can’t pull it off.
I think it's largely that as this became more of a business, the "yappers" who want to talk things out got more leverage as PMs, etc. It sounds like a caricature, but they honestly seem to get super antsy only typing and sitting in one spot.
I've never seen it put this way but I think you've hit the nail on the head.
Text requires correctness to some extent; bullshitters will just yap away for hours and nobody can point to one piece of text and say "Here, this is where you are objectively wrong, and/or misrepresenting things".
The unfortunate reality of remote work is there's a lot of zoom meetings where yappers in high places will BS away -- a lot more "important" zoom meetings than "important chats", especially in public.
It is exactly that and it’s clear the author is one of these “yappers”. I like that term. These people are also absolutely obnoxious irl and completely fail to read the room. I once had a PM like this who went into a near mental health crisis that the team of engineers were not “engaging” enough with her (in her head) witty “engineer” banter.
Perhaps it’s useful to have these people in the office, in a room of mirrors, where they can listen to themselves talk all day. There’s a subset of people who have weaseled their way into tech coming from the world of hyper-anxious very public social media engagement that simply make life miserable for everyone else.
One of the author’s primary reasonings for why remote work sucks is apparently that they find it difficult to treat other people like human beings without close proximity to them.
That’s pretty weird and uncomfortable and I don’t know that I would want to work with someone like that in or out of office.
It's a big problem especially if you haven't met the people you're working with. It's easy so think one dimensionally about a person, I catch myself doing it all the time and I can't say I'm wise enough to always stop it dead in its tracks.
now we do the work of 7 projects in half a team paid 50% less and can't get to help anyone as we all drown in 7 tickets we should do in parallel with agents writing docu on the side and assist and some of the easier code on the side because management drank the koolaid of going full into AI and "the Team now can do 300% more right". I miss the old times where making 100k and still could have few minutes to help each other and now we're in this hypercapitalistic garbage AI age were we have to just output, output, output and fuck quality and else they lay you off and get the next guy from wherever.
Everything is FAST now, I 'member 10-15 years ago if someone came to you for help you actually had the time to pull up a (possibly virtual) chair and spend 3h helping them. If you do that now you'll get canned in 4 months.
Obviously the help also came with you bonding and chit chatting about other stuff, I miss it.
> You lose ambient awareness of coworkers’ problems, and asking for help is a bigger burden.
When I was in school, I discovered that I studied more effectively and efficiently when I'm surrounded by other students who's also studying.
Then at work, I found I worked much more productively if my coworkers are all doing their work.
It's not just simply peer pressure, it's an atmosphere effect, it tell you "hey, this place is for doing this thing, now you do it too", it makes you concentrate. Sometimes being concentrated is a good thing.
I feel much the same as the article author in that
"this place is for doing this thing, now you do it too"
Is somehow powerfully motivating. But at least for me it's about the place, not the other people in the place.
I had the same covid-related journey from an office worker to unexpectedly fully remote. But I'm also lucky/privileged enough to be able to dedicate a room in my house that's quite separate from the rest of the house, and for me that's "where I work". I had coworkers who started out having to work from their kitchen table, some with housemates or children around - pretty sure that would have completely killed my productivity.
I do sometimes resent losing that room, effectively subsidising my boss by relieving his office rent costs. It used to be my "workshop" where I used my 3D printers, built drones, tinkered with electronics, and repaired stuff that broke - and I just don't do those things much any more because going into that room now feels way to much like "work" not "hobby or play".
We have the Dlang conference once a year where we all meet in person. It's amazingly productive. And yes, we do video chatting frequently. It just isn't the same.
Doors are a necessity in the work place and I hate open offices. 1 other person is okay but I'd rather a small room than no room.
A door let's be close out the rest of the world when I'm in the zone. There's time for collaboration but there's time for isolation. In a physical place I can turn off all notifications and close my door. I can make a space where there's low physical distractions like noise or people walking in front of my desk (or talking 5 feet away...) A slack setting of "away" is interpreted as either "eh, they'll probably answer" or "they forgot to turn it back off" (or they don't notice/care)[0,1]. But a physical door, people are much more cautious about knocking on it when it's usually open. It's not the same thing as a busy sign.
But I also don't think a door should be usually closed. It should usually be open. Indie collaboration but also respect your coworkers. A door is a great communication tool that you just can't get online.
[0] and for the love of god, do not hit me up with "hey". It's an asynchronous messaging system. I'll read the notification as it comes across my screen. Don't try to become synchronous with me that way. Call me, physically find me, or ask when I'm free for a call.
[1] seriously, my time is just as valuable as yours. To me it's even more valuable.
I think the author is too hard on themselves for not knowing things, probably coming from shame. I can only speak for myself but learning to shrug that off just how to say "I don't know X" or "I forgot X" is very freeing. I say I don't know shit all the time at work, and so do people I work with. We aren't encyclopedias. Just chuckle about it and figure it out and learn.
I much prefer working with people who can just be honest about what they don't know, it's way better than pretending to know or trying to save face, and generally people in the former camp seem to have higher EQ.
My boss actually said that to me once. He said he really appreciated that I would say "I don't know" instead of bullshitting him. Of course I would also try to brainstorm with him or find the answer later on by searching.
For some reason I'm able to say "I don't know" to coworkers and superiors at work with no problem, but I have a much harder time online where it may affect future employers' opinions of me.
- I blog with my real name, which includes an uncommon first name. It's easy for hiring managers to search the web for.
- My blog is linked from the website I host on the domain name I use for my email address, including for job applications. Anybody I email is likely to follow that thread.
What does it say about me, that I was SURE his article was going to be admitting out loud that we are engineering ourselves into obsolescence, a lot of us are really enjoying it, and nobody is seriously discussing how afraid we should be for our families and future. I’m afraid to mention it professionally, given we have a literal policy around “AI doomers” (not the exact term) that has the word “separation” in it. Worse, I’m afraid to THINK it, like a cognitive dissonance while Claude writes module after module for me.
I am enjoying the hell out of it, I’ve done nothing else for dozens of months, and I feel that hence I am/developers are in a unique position to understand what type of hell - or heaven - our society might experience in the next five years. Shouldn’t we be openly discussing how we can leverage this foreknowledge?
> I’m afraid to mention it professionally, given we have a literal policy around “AI doomers” (not the exact term) that has the word “separation” in it.
Dude, your employer is toxic AF. Look for a new job starting today.
The joy of US "at-will" employment is that every company's Code of Conduct reserves the right to "separate" you for undermining mission alignment. The whole system is toxic.
I appreciate the author being vulnerable like this in a public setting. It's easy to see why it would be scary, especially since admitting being wrong or not knowing something can easily be turned into questioning one's overall competence.
I wish we'd be more open about our flaws and knowledge gaps in general. I think we'd all benefit.
I had an old colleague (at various points he was my boss, colleague, and subordinate at different places) that really opened my eyes up to the power of saying you don’t know how to do something.
I used to also fear appearing incompetent if I admitted to not knowing too many things, so I would avoid showing my knowledge gaps whenever possible.
However, this colleague was the exact opposite. He would gleefully tell people he had no idea how to do certain things, would be a ready listener when the person he was talking to explained how it worked, and would heap praise on the person for their knowledge and teaching skills. He would always defer to other people as experts when he didn’t know, and would make sure our bosses and coworkers knew who had helped him and how much they knew about the topic.
What I saw and experienced was that this did NOT, in any way shape or form, make people think less of him. It did the exact opposite. First, it made people REALLY happy to help him with stuff; he made you feel so smart and capable when you explained things and helped him, everyone jumped at the opportunity to show him things. He learned so much because he made everyone excited to teach him, and made his coworkers feel smart and appreciated for their knowledge.
And then, when he did speak with confidence on a subject, everyone knew he wasn’t bullshitting, because we knew he never faked it. Since he gave everyone else the chances to be the expert and deferred all the time, you didn’t get the one-upmanship you often get when tech people are trying to prove their bonafides. People were happy to listen to him because he first listened to them.
I have really tried to emulate him in my career. I go out of my way to praise and thank people who help me, always try to immediately admit where my skills and experience lack, and don’t try to prove myself in subjects I don’t really know that well. It has worked well for me in my career, as well.
I like the vulnerability displayed by the author. I'll share a moment myself:
A few years ago I was the TL on a FAANG Android project, where for a few months I was doing more spreadsheet/TPM work than usual, and didn't have much time for coding. Once we had a meeting where I ended up coding in Kotlin live in front of a dozen younger devs to discuss the implementation of some feature. My work background is Android and Java/Kotlin, but at the time I was mostly coding in C on the side, and in the moment my brain just forgot what the syntax in Kotlin is for a "switch-case" statement, so I wrote "switch", "match", etc, struggling like a first year student, while everyone watched me fumble, until I just gave up and said: "oh my god, I'm forgetting Kotlin. What the hell is the switch keyword in Kotlin called?". Then someone said: "it's when".
I felt old and a little embarrassed, but mostly I was surprised at how quickly I could forget a programming language I used daily.
I don't know if I could tell you with confidence the proper way to get a string length in any language. Is it a global function or an object method or property? Is it length or count or size? I have to look it up or rely on intellisense every time. I do too much bouncing between languages.
No purpose in memorizing something that doesn’t meaningfully improve your understanding of the language, you only need it every now and then, and you can find it nearly instantly.
I’ve spent so much time in both the .NET and JS worlds that I can't even begin to count how many times I’ve typed `Where` when I meant `filter` or vice versa.
I can successfully type "def main()" and not "function main()", unless I'm switching between JavaScript and Python and forget which language I'm in :-). It's the if statement with all the underscores that I need to check every time.
I set up some .NET services years ago. Since then it was just adding new stuff. If I was asked how to set up another service, I would have no idea how to do it
Is that a bad thing? I also call man <libc function> several times per day, most times I think I already know the calling semantics, but there is always some nitpick about the calling semantics or a bug description, that I didn't had in mind.
LLMs are wonderful for this. I can't write hardly a line of shell script without looking something up. And then there are three different ways to do <thing> so I spend time beard-tugging as to which way to do it. Now I just tell the LLM what I want changed about this shell script and look at what it comes up with. 100% of the time it's fine.
It’s horrifying to learn that someone feels guilty for not fully obeying the commandments of Uncle Bob. It would be more appropriate to feel guilty for taking his advice.
Really appreciated this. It’s refreshing to see someone be this honest about their gaps and growth. A lot of us quietly deal with the same things, so thanks for putting it into words.
The majority of these hidden truths are due to senior engineering management in their 40s and 50s who have not coded in decades, and yet pick up the latest trend or fashion and impose that on their teams.
The monolith to microservices trend was one great example of this.
That blog post is a poor example. It replaced a simple straightforward if-else loop with a hard to understand abstraction spaghetti. A few unit tests and you'd be well on your way to shipping rather than messing around with beautiful patterns.
Obviously, because it was a fictional, simplified example for the post. My real use case was more complicated and involved multiple developers working on different parts of that flow.
The problem with if-else chains is it's easy for a programmer to forget to handle a case that another developer added in the called component. Unit tests can't help a spec miscommunication. But, visitor pattern can as it forces the handling logic to be complete.
Hence my example at the end using discriminated unions and exhaustive pattern matching in F#. Much, much simpler with the same benefits.
The section on cyberharassment is really troubling, although with the current vitriol on AI I'm not surprised. Do wish the author mentioned the name of the site though, if only so I can avoid it (and not in the Always Sunny "oh no terrible! where?" way)
I was as curious as you were. Turns out there are only so many popular threaded discussion sites in the vein of HN on the Internet, so an educated guess is all it takes.
Without making judgment on the actions of any involved party, I do wonder why the author would choose to bring up this incident and submit it as part of a story to a site where there is a significant overlap in readership.
That incident catalyzed the fear that suppressed my desire to participate online for months. I figured that if I couldn't talk about it now, I might never participate again.
Honestly, if there's any chance the content they posted on your profile before locking you out comes close to defamation, I'd consider talking to a lawyer about it. It could be that getting one to send them a cease-and-desist letter on your behalf could take care of the problem.
Well that sucks. It's exactly the site that comes to mind when I think "most popular alternative to HN".
I've generally found conversation there to be more respectful than HN, rather than less, when discussions get heated - so I had high hopes it would be a different site, but alas.
This leaves a really bad taste in my mouth.
Edit: you know what, screw it. In the spirit of "no more self censorship", here's the link: https://lobste.rs/~7u026ne9se
I don't find the conversation to be especially disrespectful. The people in the thread in question attempted to shame him, to some extent. Shame is a social measure to coerce people who are behaving contrary to society's expectations to change their behaviour. However, while shaming him, they did not especially resort to childish name-calling or ad hominem. They reasoned with him extensively as to why his behaviour was deeply undesirable. He went so far as confessing that he did not even know the language of the PR he submitted, yet intentionally withheld information about the provenance of the code. Sometimes the shame mechanism is misused for things that should not be shamed, but this seems like a clear case of shameful behaviour that deserves social repercussion.
Sadly, it seems like nothing was learned, since he settles only for diminishing his culpability in anti-social behaviour. He goes so far as to describe, in his blog post, his code as an "AI-assisted patch". When you profess that you don't even know the language of the code that the LLM generated, there is no "assistance" about it, you're at the deepest end of vibe coding. And in submitting it to an open-source project, you're making a maintainer spend more time and effort reviewing it than you did prompting it, which is not sustainable. Moreover, if the maintainer wanted a pure-LLM-generated solution, there was nothing stopping them from hopping over and typing in a prompt themselves.
In fact, most of the comments were purely a debate with no direct attacks at all. The extent of "not respectful comments" I see are something like...
So your original comment that you "didn't want to hype up AI" was a lie, you really just wanted to pretend it was your own work and didn't want the project to be able to make a choice about it. There are good reasons why projects may not want to accept code generated by AI. They may not care. But by consciously choosing not to disclose that, you took that choice away from the project. That's pretty lousy behavior if you ask me.
"Pretty lousy behaviour if you ask me" is incredibly tame. If that's what counts for toxicity, then you're advocating for a toxicly positive carebear forum where nobody is allowed to criticise anybody else's decisions.
I agree that the discussion doesn't seem to be toxic on the whole, though not superb, although I don't know what happened following in terms of harassment, so that's up in the air for me.
And I want to offer some contrast—not as a rebuttal, but just as a reminder that there’s lots of different ways to navigate this strange field.
The _majority_ of the paid code delivery I’ve done for a decade+ has been in Ruby. (The balance has been a mix of mostly devops and some TS/JS and Elixir.)
Remote work has been an utter boon. Admittedly, I do feel like it’s got worse since Covid. But I’ve been able to work with people all across the globe without uprooting my family and leaving my community, and conversely can travel without having to leave my job or clientele.
And I do find that some places benefit from thinking hard about their process. Small senior teams do great with Shape Up. Projects where you have a non-negotiable scope (replatforms) and work streams that are more reactive than planned do better with kanban than something involving estimates.
That’s not to say the author’s wrong! Again, just that the world is wide and experiences differ.
Some context here: I’ve consulted full time almost continuously since 2018, which certainly colors my experience.
Like the post overall, but the last section is a bit weird for "confessions" as it's all HIGHLY subjective. For example, I worked at a company where no one worked from home and we paired 100% of the time. When COVID hit, we started pairing over Tuple and I found it to be a superior experience to pairing in person (Tuple's drawing and attention drawing tools are far more accurate than my finger, I can use my own keyboard the odd time I want to control my pair's computer, and there are no office distractions of other pairs in the same room are benefits that come to mind). I continued to enjoy (and prefer) it for the 1.5 years I stayed after lockdown.
Taking a switch statement and spreading it out over 3x classes is not a general improvement, it is very context specific. It makes the code difficult to navigate because what used to all be in one spot and easy to read is now spread out who-knows-where and there might be a special case lurking somewhere.
It's not as subjective as it is more of a case by case decision. This example is quite misleading but polymorphic classes are sometimes useful when the domain grows, when you have have to update new behaviors all the time.. In that case then the switch becomes harder to maintain. Classes isolate behavior so new types don't modify existing code. I'd stick with switch statements in all the other cases. Sure, this could be abused and make simple things unnecessarily complicated but am just pointing out that there's a use-case for it.
I suppose I included the last section as a "confession" because I spent years pre-pandemic wishing I could go remote, talking loudly about how open office spaces were bad, etc. That, plus it's embarrassing to admit that I dislike remote work even though I am a remote worker. And not just a remote worker, but an enfranchised remote worker living 3 hours away from the nearest city with an international airport. I made the mistake of rebuilding my life around remote work before I had experienced it long-term.
Awfully nice, isn't it? You get all the privileges of being a remote worker, while "confessing" that remote work sucks to upper management who read your blog, giving them justification to deny other people the same opportunity you have. If you genuinely believe remote work sucks, you should own up to it, sell your property, and move to a city rather than reaping the benefits of it to the extent of your entire life being dependent on it while talking about how bad it is. The fact that you haven't done that indicates to me that you actually quite enjoy the benefits more than you suffer the drawbacks, but aren't acknowledging it, instead choosing to denigrate remote work for having any drawbacks at all.
> giving them justification to deny other people the same opportunity you have
I don't fear they'll deny others the opportunity for remote work. The company is "headquartered" in California, but I don't know if they even lease an office anymore. The CTO lives in the upper midwest, the architect lives on the east coast, my manager lives along the Mississippi River, and I live in the Ozarks.
> enjoy the benefits more than you suffer the drawbacks
Yes. I'm sorry, I thought I made that clear in the post. The benefits of remote work include, but are not limited to: no stress or time from commuting, an opportunity for geographic arbitrage, and the ability to build a better lifestyle around the lack of a commute. Beyond just the remote worker themself, a society that transitioned all office work to remote would also gain more benefits: more efficient use of real estate with entire office buildings rendered unnecessary, less chance of land value distortion due to centralization of workers, and less pollution due to fewer commutes.
I'm glad to also criticize in-office work for having other drawbacks. For example, I was rear-ended commuting to work more than once, the family needed the expense of two cars, we spent more on clothing, and the ambient level of noise being above 35 dBA was annoying.
While I do appreciate this joke (and I do hope this is a joke), I've recently had a project majorly held up because a lead dev didn't understand SQL. It's great to admit gaps but it's equally important to close those gaps.
> As a hiring manager I interviewed software engineers and tried to filter for object-oriented knowledge. Retroactively, it’s clear I was hypocritical.
As some one who has been on the other side of "rejected by an interviewer who didn't understand the thing they've interviewed you about" I, again, appreciate the transparency, but I'm not entirely feeling that the lesson has been learned in the case.
There was a time in my life where I felt ashamed that I didn't know calculus... so I learned calculus and my life has been better for it. While refusing to admit ignorance of a topic is particular problem in tech, confessing that you don't know something and gleefully stopping there is not much better. Holding people up to a standard you do not hold yourself to is a major problem in this field. The technical people I've learned the most from hold you to a high standard and hold themselves to an even higher one.
Of course not every engineer has to hold themselves to a high standard, but if you want to write a blog about a topic, then part of the requirements here is that you do hold yourself to a high standard. Yes, we all have gaps, and we shouldn't let shame get in the way of learning, but we shouldn't let shamelessness about what we don't know limit us either.
I am indeed learning, working to close those gaps.
For automated testing, I'm in the middle of reading Developer Testing by Alexander Tarlinder, with xUnit Test Patterns by Gerard Meszaros coming close behind. I'm also working through Test-Driven Development: By Example with my wife as we have time.
For SQL, I read Grokking Relational Database Design by Qiang Hao last winter, and I started SQL Queries for Mere Mortals by John Viescas this week. Sadly, my flub with "left inner join" was not a joke.
For OOP, I've been on a whirlwind tour: OOA&D With Applications by Booch et al., Object Thinking by David West, POODR and 99 Bottles of OOP by Sandi Metz, Domain-Driven Design by Eric Evans, IDDD and DDDD by Vaughn Vernon, Design Patterns in Ruby by Russ Olsen, Clean Architecture by Robert C. Martin, and Smalltalk Best Practice Patterns by Kent Beck. Still on the docket are Design Patterns by the Gang of Four, PoEAA by Martin Fowler, Smalltalk, Objects, and Design by Chamond Liu, and Object Design by Rebecca Wirfs-Brock.
> confessing that you don't know something and gleefully stopping there is not much better [...] we shouldn't let shamelessness about what we don't know limit us either
I promise you, this was not gleeful and this was not shameless. Shame and fear affected me for months on these issues. And I'm not stopping there... From the end of the article: "I’m going to continue to work on skill building, but now I feel free to write about it. If [...] you’d like to help me fill [my knowledge gaps], [...]"
> if you want to write a blog about a topic, then part of the requirements here is that you do hold yourself to a high standard
A high standard of writing, maybe. But plenty of great stories come from those who are striving for a high standard, not just those already in the upper echelon. It's what makes this place different from academic journals.
This is great to hear and I do appreciate the clarification. Having put lots of content out in public myself (though this account is intentionally pseudonymous) I know it's also equally difficult to comment on content like this without being on either end of the "asshole" <-> "awesome!" spectrum, and sincerely hope my comment to not fall to close the the first part.
Not exactly your main point, but where’d you go to learn calculus? I did the usual classes in high school but none since, and I’d love to develop a better appreciation for it.
Check out Mosaic Calculus and see what you think: https://www.mosaic-web.org/MOSAIC-Calculus/. It's a free resource and it takes an interesting perspective on calculus pedagogy.
I find 3blue1brown to be a great resource to build up good intuition about math topics, his videos about calculus and linear algebra are wonderful in particular. https://www.youtube.com/playlist?list=PLZHQObOWTQDMsr9K-rj53... is his essence of calculus series, I found the visualizations made it a lot easier to grok.
I can second 3blue1brown as intro, but anything practical needs practise. Get your local 12th grade or undergrad math book which has problems to solve will bring more focus and faster learning. Have a list of problems to solve in your head while learning theory has been faster for me than just learning theory.
My suggestion is using Khan Academy if you want to better your math knowledge. It's really quiet good for that sort of thing. It was just starting to take off when I finished my degree. I wish it was available before then.
Learning stuff that doesn’t help in work(calculus is not helpful for 99% of software engineering) is really hard if you don’t base it in reality I find. Maybe it’s just me but I would never read a text book for fun so suggesting learning by reading a text book seems crazy. Calculus can be fun and interesting but the teacher has to actively try to make it so. The learning will take longer but you’re more likely to see it through and I think it’s more likely to stick long term too.
I learned way more reading crafting interpreters than I did in my compiler class for example.
That’s interesting because I can’t imagine learning a subject without a textbook. I have a hard time believing another medium would have the depth and density to get all the points across. Although it does depend on the subject matter and one’s learning goals.
But I also do read textbooks for fun… Now that I have a few decades of experience in a lot of these subjects I get way more out of the books. And I can start to understand more of the meta information. Like, of all the things the author could’ve used as an example, why did they pick that. Also, it’s hugely interesting for me to look at the homework problems and theorize why this particular problem was picked. Especially fun for electrical engineering books. But ya, I’m weird like that.
> And yet, my lack of awareness of polymorphism showed me I’ve been writing little more than structured programs. That I could replace conditionals and case staments with specialized classes had never crossed my mind.
> Polymorphism is covered in every college OO course.
Consider yourself blessed then because you're in for one hell of a ride if you pursue that path to its extreme. For me, it was the opposite: been taught OOP, I had to unlearn most of it to be able to better structure my mind and how I think about programs.
You should know what polymorphism is (also, there's static, dynamic, ad hoc, single dispatch, multiple dispatch), but I don't think it's a weakness if you have not been using it that much (the real weakness is over-using it and making a clusterfuck out of your code.)
Which is a long-winded way of saying that you could be doing much worse, if that makes you feel any better, lol.
Refreshing to read, I bet it was cathartic to write. I hope your fears don't come true. I think they won't. Many people do genuinely appreciate this kind of honesty, even when directed against them, but it is a gamble.
A good reminder that everything we say/hear/write/read exists in the unseen context of all the things we believe we should not say.
Remote work is great (for the reasons you gave and more) and saying it "sucks" made me roll my eyes, and it's reductive in the same way as saying office work "sucks." I wouldn't have had a job if in-office was the only option. It certainly didn't suck for me.
Being bad at problem solving with people far away is just another problem you can solve with practice. Same as being bad at problem solving even when help is right next to you.
> made me roll my eyes, and it's reductive in the same way as saying office work "sucks."
Yes, "remote work sucks" is reductive, but I elaborated beyond the heading. Also, I wouldn't disagree with "office work sucks." Remote work simply has its warts, too.
> just another problem you can solve with practice
Perhaps, but practice alone clearly isn't enough. I've been working remotely since 2020 and it hasn't gotten more enjoyable. I would love to solve that problem, though. I read Remote: Office Not Required by Jason Fried in the past, but that was written a long time ago. I've added more recent works (Effective Remote Work by James Stanier and The Async-First Playbook by Sumeet Gayathri Moghe) to my reading list.
I’ve worked remote for at least 16 of the past 22 years including my first job out college. It’s always been friggin awesome. The only downside was when I was contracting and I’d get calls in the middle of dinner and I didn’t have the self-discipline to ignore the call. A few times a year I have to travel to work, it’s nice to see folks, but it’s not required to get the work done, I put my big boy pants on and figure it out, or ask for help when I can’t.
> it’s easy to form an enemy image of somebody at the end of video call, but difficult to keep that image when you share a room with them and sense their pain.
I'm honestly so confused by this. Has the author never worked in an office before? Building a grudge for someone that you are forced to work with and sit next to all day is one of the classic office dilemmas. Being forced to be around them all day can really build resentment to people
Remote work isn’t for everyone. Their point of view is just as valid as your point of view.
And this is my biggest complaint about arguments about remote working. People turn it into something that’s evidence-based when actually it’s a deeply subjective topic and thus different personality types thrive in different working environments.
My point is you cannot have any fact based discussions because statistics are generalised whereas people’s ability to thrive in office or remote work depends deeply on both the company culture and the individual.
Thus it is always going to be an empirical argument rather than a fact-based discussion.
Or work with a random programmer at a random company for a bit. We had to do some audit/estimate for a company with huge tech issues (race conditions causing data corruption, huge slowdowns with just small usage spikes etc) costing them clients/money. The company runs around $50m/year or so selling software to enterprises. Anyway; software backend written with Java/Spring, deployed/updated on EC2 manually, no automated tests (zilch). Frontends with Vaadin. Almost no processes are used, just Jira + tasks and then 'start at the top every morning'. No one knows sql anymore (Hibernate), no one knows html/js (Vaadin) and, even though most people are senior and there since the beginning, no-one has done anything high level on the job in the past 20 years or so. They have just been inside this 'ecosystem' writing code and it works. Old Java with some modern updates just to satisfy the compiler/linter (but not fully understanding why that nonsense is needed). None of the core seniors I interviewed touches computers outside of work, they had 0 tech courses since working there etc. They are all 9-5 code producing robots. I want to bet they can mostly/all be replaced today by Claude Code, of which existence, of course, they are not aware (they did chatgpt but not Codex or Copilot). We have since found so many issues in the code. Yeah, I do feel very much uplifted about my own skills after encounters like this, and these are by no means rare, i would rather say; extremely common.
Unlike OP though, I cannot be as open about these companies as we would definitely not have any clients left after.
That sounds amazing, what you can get away with while still shipping a product and getting paid. In some way probably the engineers there are unwilling to do any automation in fear of becoming redundant, and the company is still fine with that.
Also, I am not sure how not touching computers after work is a bad thing; people can have families and other hobbies?
> computers after work is a bad thing; people can have families and other hobbies?
No not bad per se, but it did clearly show that, without on the job courses, why all of them are stuck in the early 2000s tech wise.
Some people start with a company and get lucky with early success and then get restricted because of that success: get new clients via existing, everyone likes it and asks for new features and without noticing it you might find yourself 15 years down the road with ancient tech and no one understanding anything current. Then you can still thrive if your clients like it... we have similar clients: a 1980s factory, another 1980s factory, a logistics app from the 1990s etc. Things deeply ingrained in some vertical, expensive but better priced than the SAPs etc of this world so it keeps going and going.
> Remote work eliminates a lot of problems with office work: commutes, inefficient use of real estate, and land value distortion. But software development is better when you breathe the same air as the folks you work with. Even with a camera-on policy, video calls are a low-bandwidth medium. You lose ambient awareness of coworkers’ problems, and asking for help is a bigger burden. Pair programming is less fruitful. Attempts to represent ideas spatially get mutilated by online whiteboard and sticky note software. Even conflict gets worse: it’s easy to form an enemy image of somebody at the end of video call, but difficult to keep that image when you share a room with them and sense their pain.
Every ounce of data proves this statement wrong. If you feel like you work better non-remote then do it. Don’t shill it as a panacea. I’ve been remote for 11 years now and if I wasn’t I wouldn’t have been able to take care of my family, go back to school part time, work on my health with better meals and reasonable gym hours, etc. even IF in office was better for the employer (even though all data says it’s not in terms of productivity) it is unequivocally better for the employees life to work remote as much as humanly possible.
This hot take is just simply insane. Humanity had no problem coordinating massive projects over IRC and mailing lists. It’s clear the author is a “nu-coder”.
> software development is better when you breathe the same air as the folks you work with
Ever heard of flu season? What if you have a family and don't want to bring diseases home?
> Attempts to represent ideas spatially get mutilated by online whiteboard and sticky note software.
Right... like, the Linux kernel team? Or any of the major open source key pieces of technology you use? Built by large teams that worked remotely for decades even when tools where orders of magnitude worse than the current state of the art? Some of them never meeting each other in person?
---
Remote work DOESN'T suck. YOU make it suck.
Remote work is great if you care about shipping.
Want to go for coffee or want to talk about our weekends? No thanks.
Did you see the distracting thing outside the building? No, I didn't because I don't have to go there anymore.
Is the heat too high or too low? It's your own home, just adjust it to YOUR convenience.
Worried about your pets being alone? Just be next to them. I care more about my pets than some stranger from work.
Want to be loud and flex about random stuff? Log into LinkedIn and talk to the other geniuses like you while I focus on doing my job.
Most people SUCK at drawing, suck at calligraphy and their whiteboard diagrams SUCK. Therefore, whiteboards SUCK. Unless you have great calligraphy and drawing skills, whiteboards are not helpful. You are just sad because you are not getting attention by being in front of other people looking at you.
I don't know that it is necessarily sick ... I suppose it depends what you do, how the people respond, and if you are playing a useful role in the interview process for your organization.
Tell us more?
On the positive side... Seeing how people respond to pressure (done within some bounds) might be useful information. Watching people dig deeper into their problem-solving toolkit (and even their determination and resolve) might be informative. But I have not recently reviewed interviewing research, so I don't know how strong the causal connections are between observing these things in an interview and on-the-job performance.
On the negative side: If one is primarily interested in finding flaws as some sort of intrinsically valuable thing or to boost one's own ego, these are probably hints to look closely at oneself and reflect and make sense of what is going on inside.
Well written article full of humility and vulnerability? I love it. My reaction: you don't need to feel ashamed of not knowing something, there is far too much to know and I'm still learning new techniques and concepts 37 years in, so I would never judge you for it.
I would also not judge you for having your own preferences and opinions. I too prefer working in an office to remote work, but when I say this out loud other developers take it as advocating RTO or saying remote work is worse when it just doesn't suit my personality. I get that it's a touchy subject but there is no need to get up in my face about it.
You mention bullying and brigading and that seems to be an unfortunate reality of this industry. I suspect there is a lot of insecurity and imposter syndrome that causes people to write hyper-confident blog posts about why they are better without AI and how their tests have 100% coverage and how (unfashionable language which half the world uses) is garbage etc. Maybe if we all follow your example and be candid everyone could chill out a bit.
I'll go next: despite trying several times, I have never successfully written anything more complicated than Fibonnacci in Lisp or Haskell. I know it's clean and pure and all that, but my brain just won't work that way.
> I have never successfully written anything more complicated than Fibonnacci in Lisp
I hadn't until a joined a lisp based project. Learned a ton. My brain didn't work that way at first either, but working with it every day I eventually got it.
Every time that I read this about remote work, all I can think is how much I miss IRC and the culture that came from it.
We were doing remote work effectively decades ago. Don't have hallway conversations to fix bugs? Easy, just post your problems on the team chat and someone (often one of several people) would love to drop by to help.
I'm not sure exactly all of the forces that have led to this changing so much, but I'm certain that merely blaming "remote work" isn't it.
Somehow we were better at using remote tools while literally in the same office than some teams are at using them now while fully remote.
I have encountered people who are scared to post in large public channels. Part of growing up in chatrooms was an implicit bravery of saying something out loud in a room full of thousands of people. There seems to have been a shift, somewhat, in the comfort level of different generations about saying things "out loud" in large public rooms.
Chatrooms have evolved in a really interesting way. I think the first generation to have them didn't fully understand how "public" they were. Maybe there are more people in the more recent generations that have a more visceral understanding of online "publicness" as they have grown up with (and perhaps have been burned by) those concepts from the very beginning. Maybe they have a better understanding of the permanence of online utterances and therefore have a more conservative approach to interacting on what feels like the permanent public ledger.
Maybe it's because the concept of pseudonyms has devolved since the early days. Corporate social media has an interest in doxing its users to advertise to and control them but pre-corporate social media was filled with anonymous usernames. Posting in a large group under your permanent forever name is much scarier than posting under an anonymous, temporary identity. One of the things I advocate people do is post online anonymously, instead of with their real name. It alleviates a lot of the fear of speaking your truth, which we need more of!
There is something there. The ability to try on identities in a safe environment before you discover which one you really identify with. It's much harder to do this with your real name. Your past comes with a lot of baggage and people who know you don't want you to change because it makes them feel uncomfortable.
There's an issue of scale to it all, and interconnectivity. Back in '98 you could reasonably post something in one forum/BBS/IRC channel and it would only be viewed there. There was no way to look up who was on what website or room and where they regularly hung out. And even if there was, it's unlikely that more than ten people would ever see what you posted. There weren't enough people to care, and there wasn't any extrinsic incentive to look up what people did outside of your tiny island. Eventually the island would sink, and all traces of it except maybe an archived snapshot of the home page would remain.
Smartphones changed that with Youtube and Facebook. Youtube incentivized you to use a Google account, and Facebook wouldn't let you use it anonymously without an account. Because you could use one account to log into multiple places people could track you across websites. People could make archives, screenshots, and transcriptions of anything you had done with those linked accounts. With this change there was no safe corner to hide if you said something stupid. And because so many people were foolish enough to tie their real identities to these online accounts with their real names or pictures of themselves, it gave a way for particularly unruly people to track these individuals even offline. There was now a real danger if you said something stupid, because instead of just getting your post deleted or starting a derailment in the thread people could harass you at your home, get you fired, and even send the police to terrorize you in the middle of the night via SWAT raids. It's no longer just one person calling you out. It's now hundreds, maybe even thousands, all armed with information.
And this is why I say it's stupid to require phone numbers and real names to sign up for insignificant things like being able to view someone bake a duck shaped cake live over the internet.
> There seems to have been a shift, somewhat, in the comfort level of different generations about saying things "out loud" in large public rooms.
I think this is merely the shift from doing this as a hobby, to doing this for work. Random coding problems mixed with banter I posted or answered on IRC back in the day? Purely hobby stuff, things I done after school instead of doing my homework. No stakes beyond the community itself, I could disengage at any moment, nobody would care - there was no commitment of any kind involved.
Today? Even if we switched back from Slack/Teams/whatnot to IRC, the fact remains, the other people are my co-workers, and we're talking about work, and it's all made of commitments and I can't disengage, or else I starve.
That changes the dynamic quite a bit.
I use pseudonyms and post weird shit online, but I still feel very reluctant to post anything on large public channels inside a company. Everything is tied to your real name and all of us are hyper aware now that every single fucking thing on the internet is monitored, and will be used against you if necessary. I am 99% confident a tool already exists that a manager can use to get all messages from employee X over the last N months and summarize the content and surface any "red flags", which in a corporate setting would be incredibly broad.
We have slack integrated with glean and anyone in the company can do this, not just managers.
It also has access to our internal wikis, GitHub, and other internal tools.
The problem in my opinion is that folks afraid of posting treat chat channels like email and official record instead of a conversation. I like to post ideas, brainstorm, engage if I have a minute to respond to someone with a thought - kinda like being in an office - whereas many others seem to use it to blast out information after a lot of polish and they form a culture of announcements and no engagement, and get stressed whenever someone asks a question or actually replies.
Use tools for what they are good for and create a culture that makes each tool work best for your organization.
> I have encountered people who are scared to post in large public channels.
People weren’t assholes and/or snowflakes in those days. Implicit in being on the net was that you were fairly well behaved.
Oh lord no. I’m not sure that’s true at all! I mean in the early cool era of the internet there was a concept of a “netizen” and a window of brief joy, but the internet has always had domineering trolls, bullies, spaces where clumsy newbies were brutally flamed etc.
The main difference is that more spaces were quasi-professional and non-pseudonymous, in that one largely got one’s internet access and identity (IP address, email address, invitation) from the institution of higher learning one attended or worked for. So there were direct, two or three degrees separation consequences (my boss knows someone at your institution) in those spaces. I suppose this is what you are referring to.
(In my early era of commercial internet work I can remember a colleague shutting down an accidentally abusive scraping bot by working out who was likely to be the boss of the person running it and phoning them up)
But away from those spaces were many places that were just as bad as they are now.
The internet has always (in my time of using it, which is all of my adult life as someone who is over half a century old) demonstrated that a good culture is a question of starting conditions and quick maintenance actions.
A non-trivial amount of the worst behaviour I have personally witnessed on the internet happened before the year 2000.
I got a minor reprimand in a performed review for having a slightly heated conversation in a public channel. In the past that company had been very open about communicating about negative stuff (our CEO emailed the entire company when he visited one of the hardware labs that was a mess), but the upper management started tamping down on anything negative, and one of the things that suffered was any sort of honest communication.
Kind of a ship of theseus situation culture wise - when the original leaders are all gone, did they pick good successors to fill their spots? Very often not.
This sometimes happens also when the original leaders are still present but they dont understand the effect the metrics are having on the entire company when it grows big.
Have a senior leadership team and want them to not tell you bad news when you are the CEO/Leader? Then link their salary/performance to metrics like number of production incidents their team has. Suddenly the number of incidents that you know of decreases.
If that does not work to isolate you as the leader from thr reality of your company then link their salaries to a metric like number of projects finished before or at deadline and watch how tech debt increases multiple folds and how everything is suddenly estimates are increasing all over the place.
Want people not to ask meaningful hard questions in All Hands? Just make sure anyone that seems critical be labeled as not culture fit and done. All questions are positive and nice. Make sure to always ask for name and disable any anonymous questions asked.
Not trying to say metrics are bad or they should not be used. But they are not pure functions :) they do have side effects and sometimes very large ones.
> Somehow we were better at using remote tools while literally in the same office than some teams are at using them now while fully remote.
I couldn't agree more. I pushed to get the place I worked for to use Slack when it first launched, moving us off AIM (ha!). Our use of Slack when we shared an office in the twenty-teens was so much better than the use I've seen of Slack/competitors on fully-remote teams.
I wonder if it's because the failure mode was, as you said, to "drop by." Now the failure mode is... just failure.
I can't speak for everyone, but I'd say that I've noticed that younger devs simply do not chat.
My team rooms are pretty dead. I'll send stuff there but by and large the team simply doesn't use chat functions.
Perhaps the youngins are more cognisant that it's all monitored. Knowing your employer can read everything and it _will_ be used against you has a chilling effect and I'm pretty sure that's part of it.
We had such incredibly heinous group chats on our Slack that if an admin perused through the logs we'd be instantly fired and the company shut down right then and there lol. The paranoia drove everyone nuts which made it more fun.
Every healthy person I know does that. It’s as if human communication required to have secrets before people would relax about opening up.
Partially thats about teams and how most corps use it, which is built primarily around information siloing and management visibility.
They, for every team I’ve ever managed, have an off company owned systems chat on shit like slack or discord where they are roasting the fuck out of you.
I’ve managed to be invited or told of them after ingratiating myself to the teams, or more often, after quitting and getting invited as one of the “good ones”
They all know that every word on company shit is being monitored
Sure, but this ends up poisoning any sort of culture and creating all sorts of in-group nonsense which is almost impossible to undo.
It’d be like using Blind as your company chat - nobody goes on there to say how great their experience has been, and the tone infects everything else.
But maybe I’m just not very fun at parties…
This should be avoided at all costs by creating a culture that is receptive to people’s concerns and doesn’t do stupid things without explanation - but I get how difficult that is in reality and most orgs end up messing this up.
Maybe I'm a bit unfair to you but to me your comment basically reads as wishing employees would be good little cogs in your machinery rather than people. Like making friends is natural human behavior. Forming friend groups is natural human behavior. It's not nice to disrupt this except that of course everyone has to be able to work together when needed.
It doesn't help that reporting people to HR is a way to career advancement.
I’ve never worked _anywhere_ where reporting someone to HR was anything but negative impacting for your prospects at the company. And I’ve worked at lots of places in many dimensions (company size, industry, age, etc)
This usually applies to Big Bank.
The karmic cost / benefit is all worked out then.
I worked at a big bank and it definitely did not.
Is that a dynamic they have? I haven’t worked at Big Bank but I’ve worked in finance a few times and at those places and other industries I’ve worked in reporting anything to HR wouldn’t necessarily get direct consequences but you would permanently be on their radar and have to work to rule after that
I think it’s more a shift in the culture amongst most people now than an argument of remote/in office.
Notification fatigue is a thing and people are just used to ignoring notifications and messages nowadays which ends up with slow responses and poor communication all around.
It's much easier to get a feel for how urgent things may be when you can yell down the hall.
I'm currently in feeling things out phase with my current team, and people seem really laid back about responding to messages - but it also seems like we're getting stuff done. Hard to figure.
> how much I miss IRC and the culture that came from it.
IRC selects for people who like chatting and communicating via text.
I think the mistake made with remote work was assuming that everyone could easily work that way.
The best experiences I had with remote work were pre-COVID, when the teams working remote were carefully selected for having good remote work abilities and anyone who couldn’t handle it was kicked back to the office (or out of the company)
Then something changed during COVID and remote work was treated as something everyone could do equally well. The remote teams I worked with were now a mix of people who could work well remotely and people who wanted to work remote but tried to force communication to happen like we were back in the office: Meetings for everything. Demands to “jump on a quick call” when a few Slack messages would have done the job. Then there were the people who read “Four Hour Work Week” and thought they were going to do their jobs from their iPhone while traveling the world or at the ski resort.
I don’t know. Having seen the before and after it doesn’t feel so surprising that remote work faltered when applied indiscriminately to everyone. The best remote teams I work with to this day are still the ones who know how to communicate in that old school IRC style where communication flowed easily and everyone was on the same page, not trying to play office games through Slack.
It's a strange pattern I observe often, whenever an idea gets promoted from organic-natural-human-ritual to official-new-visible-main-idea, it becomes bloated and off point.
Agile. Case in point.
Did you read the article? :)
Yeah. Random example: I have better "ambient awareness" remotely because with slack I am in every hallway simultaneously, and can skim the conversations and set up highlight words
I wonder how much of that comes down to culture. Since going remote I have come to wonder if a direct-message-first chat culture is harmful to collaboration.
IMHO most companies encourage public-first conversation, but still end up with DM-first as their employees don't have enough trust in how their messages will be received.
It requires to be comfortable exposing lack of knowledge or saying weird things to peers, and be confident it will be taken in good faith. As you point out, that requires a whole level of culture building.
DM-first is an extremely frustrating culture. That kind of operation tells me that that folks are too risk averse and political to discuss things openly. Typically this is led by panicky managers that are worried about involving too many people or having to explain things to folks they don’t want to deal with, and it escalates from there and gums up ALL the things. It makes Slack basically useless.
The same people DMing however will also extol the virtues of posting in public and lament why there is not more conversation happening in the open.
> Since going remote I have come to wonder if a direct-message-first chat culture is harmful to collaboration.
There's no question! It absolutely is.
It is. You need to be aware of it and have people that can set examples about chatting in public rooms or who can recognize when to stop a dm chat and move to be public
I hate direct messages. They were normally considered rude in IRC.
Yeah, I joined the Mozilla community in ~2003, and that was all IRC, and distributed all over the world, and it worked very, very well.
Note that these days, the Mozilla community has moved to Matrix, which also works very well for these things.
I started doing remote back in 2002, working with companies in the UK and US, something that was largely unheard of in my country. While our mode of communication was web messengers (AOL, Yahoo, MSN etc and the aggregators), I used IRC a lot to get tech help, and participate in online dicussions.
Right, because there's nothing special about remote vs in-office. It's just communication and collaboration.
Living beings do it all kinda ways. Bees waggle their butts, crickets rub their legs, geese honk, snakes hiss, some fish detect electrical signals. And to collaborate, the bees' dance indicates a flight path, birds singing indicates interest in mating, the snake's hiss and the geese's honk tells you to watch out. You use the tools you have and develop collaboration with them. There's clearly no right way, there's just ways.
But tomorrow morning, would you wanna learn to honk at people, or rub your legs, or waggle your butt, to order a latte at Starbucks? It'd be awkward, weird, painful, and unnecessary. So if you were asked to, you'd probably not try very hard to adapt to it. And if everybody you knew were in the same boat, all being forced to change with no real guidance, kinda not trying that hard to make it work? It would suck for everybody.
People just don't like changing what they're used to. They probably don't even mean to fight it. But we do like culture we're already familiar with. Change is hard, not changing is easy. We like easy. So people who grew up with remote work (on IRC, mailing lists, etc) find it easy, even more productive. But a company that's thrown into it without a healthy established culture are going to be swimming upstream indefinitely.
Electronic chat is really not the same as face-to-face communication. Neither are video calls.
For me, electronic chat is better most of the time.
Everyone is different. I vastly prefer email over chat, but also wouldn't want to live without the occasional face-to-face.
"Everyone is different" is my point. Face to face once in a while is nice, though I don't care so much about having it very often with my colleagues in a professional setting.
Of course it's not the same, that's the point. I personally prefer the async, chat based mediums I've used since I was a child. Some of my coworkers have disabilities that make conversational typing difficult and prefer video calls as a result.
Consider the effort to accommodate those preferences though. Accommodating a video call preference is easy. Same for chat. Accommodating a preference for face-to-face requires spending an hour (2x average US commute) traveling to meet you. That's quite a significant ask of the other person.
Yes, it’s not the same at all.
In electronic chat I can ask someone to explain their question and wait for it in writing. In person, I often have to listen to them stumble over the concept because they didn’t think about what they wanted to ask before asking it.
In a video call I can clearly see the other person’s screen and zoom in on what I’m trying to look at. In person I have to lean over their desk and squint at the right angle.
> Somehow we were better at using remote tools while literally in the same office than some teams are at using them now while fully remote.
This is sort of the point. Remote tools work great when you have spent a lot of time building relationships and rapport with the people involved. That's hard to do in professional settings, and extremely hard to do in remote professional settings.
Letting teams that know each other well work remotely works great. Building teams remotely is very hard.
I'm a diehard for remote work, but we have to be realistic abouts limitations.
IDK, some of the remote teams that I've worked on were only able to meet in person once per year, if that. They were very communicative on the tools that we had though.
No one is saying it’s impossible to build fully remote teams from scratch, it’s just very hard and requires strong leadership. Most companies have crappy management so they can’t pull it off.
Debian works like that but the startups can't manage… interesting.
Open-Source Project that doesn't have Project Managers and MBA telling you to work faster is a lot different then working at Startup that does.
I think it's largely that as this became more of a business, the "yappers" who want to talk things out got more leverage as PMs, etc. It sounds like a caricature, but they honestly seem to get super antsy only typing and sitting in one spot.
I've never seen it put this way but I think you've hit the nail on the head.
Text requires correctness to some extent; bullshitters will just yap away for hours and nobody can point to one piece of text and say "Here, this is where you are objectively wrong, and/or misrepresenting things".
The unfortunate reality of remote work is there's a lot of zoom meetings where yappers in high places will BS away -- a lot more "important" zoom meetings than "important chats", especially in public.
It is exactly that and it’s clear the author is one of these “yappers”. I like that term. These people are also absolutely obnoxious irl and completely fail to read the room. I once had a PM like this who went into a near mental health crisis that the team of engineers were not “engaging” enough with her (in her head) witty “engineer” banter.
Perhaps it’s useful to have these people in the office, in a room of mirrors, where they can listen to themselves talk all day. There’s a subset of people who have weaseled their way into tech coming from the world of hyper-anxious very public social media engagement that simply make life miserable for everyone else.
Yeah text based beats video a lot of the time.
One of the author’s primary reasonings for why remote work sucks is apparently that they find it difficult to treat other people like human beings without close proximity to them.
That’s pretty weird and uncomfortable and I don’t know that I would want to work with someone like that in or out of office.
It's a big problem especially if you haven't met the people you're working with. It's easy so think one dimensionally about a person, I catch myself doing it all the time and I can't say I'm wise enough to always stop it dead in its tracks.
I can’t say I have ever experienced this problem after conversing with someone more than a handful of times.
Sure it applies to things like random people on social media and such, but after a mutual exchange or two you should be over it.
now we do the work of 7 projects in half a team paid 50% less and can't get to help anyone as we all drown in 7 tickets we should do in parallel with agents writing docu on the side and assist and some of the easier code on the side because management drank the koolaid of going full into AI and "the Team now can do 300% more right". I miss the old times where making 100k and still could have few minutes to help each other and now we're in this hypercapitalistic garbage AI age were we have to just output, output, output and fuck quality and else they lay you off and get the next guy from wherever.
Everything is FAST now, I 'member 10-15 years ago if someone came to you for help you actually had the time to pull up a (possibly virtual) chair and spend 3h helping them. If you do that now you'll get canned in 4 months.
Obviously the help also came with you bonding and chit chatting about other stuff, I miss it.
> You lose ambient awareness of coworkers’ problems, and asking for help is a bigger burden.
When I was in school, I discovered that I studied more effectively and efficiently when I'm surrounded by other students who's also studying.
Then at work, I found I worked much more productively if my coworkers are all doing their work.
It's not just simply peer pressure, it's an atmosphere effect, it tell you "hey, this place is for doing this thing, now you do it too", it makes you concentrate. Sometimes being concentrated is a good thing.
That sounds akin to like body doubling, a method used by those who have ADHD.
Is there also an opposite effect? I concentrate best in solitude, preferably somewhere like an abandoned lighthouse.
I'm not sure that's necessarily "opposite"?
I feel much the same as the article author in that
"this place is for doing this thing, now you do it too"
Is somehow powerfully motivating. But at least for me it's about the place, not the other people in the place.
I had the same covid-related journey from an office worker to unexpectedly fully remote. But I'm also lucky/privileged enough to be able to dedicate a room in my house that's quite separate from the rest of the house, and for me that's "where I work". I had coworkers who started out having to work from their kitchen table, some with housemates or children around - pretty sure that would have completely killed my productivity.
I do sometimes resent losing that room, effectively subsidising my boss by relieving his office rent costs. It used to be my "workshop" where I used my 3D printers, built drones, tinkered with electronics, and repaired stuff that broke - and I just don't do those things much any more because going into that room now feels way to much like "work" not "hobby or play".
We have the Dlang conference once a year where we all meet in person. It's amazingly productive. And yes, we do video chatting frequently. It just isn't the same.
If only it could be daily, all year round! Or maybe the part about it happening once a year is a vital ingredient.
I'm still trying to replicate my favourite working condition: being on a laptop in the middle of a concert. Pure focus.
Some folks just socialize internally, or have enough little self loops to approach a task in multiple ways
I prefer working in person iff I have a door.
Doors are a necessity in the work place and I hate open offices. 1 other person is okay but I'd rather a small room than no room.
A door let's be close out the rest of the world when I'm in the zone. There's time for collaboration but there's time for isolation. In a physical place I can turn off all notifications and close my door. I can make a space where there's low physical distractions like noise or people walking in front of my desk (or talking 5 feet away...) A slack setting of "away" is interpreted as either "eh, they'll probably answer" or "they forgot to turn it back off" (or they don't notice/care)[0,1]. But a physical door, people are much more cautious about knocking on it when it's usually open. It's not the same thing as a busy sign.
But I also don't think a door should be usually closed. It should usually be open. Indie collaboration but also respect your coworkers. A door is a great communication tool that you just can't get online.
[0] and for the love of god, do not hit me up with "hey". It's an asynchronous messaging system. I'll read the notification as it comes across my screen. Don't try to become synchronous with me that way. Call me, physically find me, or ask when I'm free for a call.
[1] seriously, my time is just as valuable as yours. To me it's even more valuable.
I think the author is too hard on themselves for not knowing things, probably coming from shame. I can only speak for myself but learning to shrug that off just how to say "I don't know X" or "I forgot X" is very freeing. I say I don't know shit all the time at work, and so do people I work with. We aren't encyclopedias. Just chuckle about it and figure it out and learn.
I much prefer working with people who can just be honest about what they don't know, it's way better than pretending to know or trying to save face, and generally people in the former camp seem to have higher EQ.
My boss actually said that to me once. He said he really appreciated that I would say "I don't know" instead of bullshitting him. Of course I would also try to brainstorm with him or find the answer later on by searching.
For some reason I'm able to say "I don't know" to coworkers and superiors at work with no problem, but I have a much harder time online where it may affect future employers' opinions of me.
Where are you conversing online that this is a concern?
My blog.
- I blog with my real name, which includes an uncommon first name. It's easy for hiring managers to search the web for.
- My blog is linked from the website I host on the domain name I use for my email address, including for job applications. Anybody I email is likely to follow that thread.
What does it say about me, that I was SURE his article was going to be admitting out loud that we are engineering ourselves into obsolescence, a lot of us are really enjoying it, and nobody is seriously discussing how afraid we should be for our families and future. I’m afraid to mention it professionally, given we have a literal policy around “AI doomers” (not the exact term) that has the word “separation” in it. Worse, I’m afraid to THINK it, like a cognitive dissonance while Claude writes module after module for me. I am enjoying the hell out of it, I’ve done nothing else for dozens of months, and I feel that hence I am/developers are in a unique position to understand what type of hell - or heaven - our society might experience in the next five years. Shouldn’t we be openly discussing how we can leverage this foreknowledge?
We’ve got a few years left where we function as tech leads for an army of AI agents.
The inflection point will be when business hires an AI to fill a managerial role. AI will discriminate against hiring human developers.
Since you’re so familiar with the process, I suggest you start positioning yourself as an AI development efficiency management consultant or similar.
> I’m afraid to mention it professionally, given we have a literal policy around “AI doomers” (not the exact term) that has the word “separation” in it.
Dude, your employer is toxic AF. Look for a new job starting today.
The joy of US "at-will" employment is that every company's Code of Conduct reserves the right to "separate" you for undermining mission alignment. The whole system is toxic.
I appreciate the author being vulnerable like this in a public setting. It's easy to see why it would be scary, especially since admitting being wrong or not knowing something can easily be turned into questioning one's overall competence.
I wish we'd be more open about our flaws and knowledge gaps in general. I think we'd all benefit.
I had an old colleague (at various points he was my boss, colleague, and subordinate at different places) that really opened my eyes up to the power of saying you don’t know how to do something.
I used to also fear appearing incompetent if I admitted to not knowing too many things, so I would avoid showing my knowledge gaps whenever possible.
However, this colleague was the exact opposite. He would gleefully tell people he had no idea how to do certain things, would be a ready listener when the person he was talking to explained how it worked, and would heap praise on the person for their knowledge and teaching skills. He would always defer to other people as experts when he didn’t know, and would make sure our bosses and coworkers knew who had helped him and how much they knew about the topic.
What I saw and experienced was that this did NOT, in any way shape or form, make people think less of him. It did the exact opposite. First, it made people REALLY happy to help him with stuff; he made you feel so smart and capable when you explained things and helped him, everyone jumped at the opportunity to show him things. He learned so much because he made everyone excited to teach him, and made his coworkers feel smart and appreciated for their knowledge.
And then, when he did speak with confidence on a subject, everyone knew he wasn’t bullshitting, because we knew he never faked it. Since he gave everyone else the chances to be the expert and deferred all the time, you didn’t get the one-upmanship you often get when tech people are trying to prove their bonafides. People were happy to listen to him because he first listened to them.
I have really tried to emulate him in my career. I go out of my way to praise and thank people who help me, always try to immediately admit where my skills and experience lack, and don’t try to prove myself in subjects I don’t really know that well. It has worked well for me in my career, as well.
The "why I was afraid to admit it" is by far the best part of this. It gives a lot more insight and ideas to combat it.
I like the vulnerability displayed by the author. I'll share a moment myself:
A few years ago I was the TL on a FAANG Android project, where for a few months I was doing more spreadsheet/TPM work than usual, and didn't have much time for coding. Once we had a meeting where I ended up coding in Kotlin live in front of a dozen younger devs to discuss the implementation of some feature. My work background is Android and Java/Kotlin, but at the time I was mostly coding in C on the side, and in the moment my brain just forgot what the syntax in Kotlin is for a "switch-case" statement, so I wrote "switch", "match", etc, struggling like a first year student, while everyone watched me fumble, until I just gave up and said: "oh my god, I'm forgetting Kotlin. What the hell is the switch keyword in Kotlin called?". Then someone said: "it's when".
I felt old and a little embarrassed, but mostly I was surprised at how quickly I could forget a programming language I used daily.
My knowledge-gap confession: even after many years with the languages, I can't write a main() in Python or Java without looking up the format.
So many things like that!
I don't know if I could tell you with confidence the proper way to get a string length in any language. Is it a global function or an object method or property? Is it length or count or size? I have to look it up or rely on intellisense every time. I do too much bouncing between languages.
Well, I know it in BASIC. Len().
No purpose in memorizing something that doesn’t meaningfully improve your understanding of the language, you only need it every now and then, and you can find it nearly instantly.
With Java 24 it’s now:
Not if you want to handle arguments?
IntelliJ completes the format for me. I never write it by hand. Also I have never coded in something other than an IDE.
I’ve spent so much time in both the .NET and JS worlds that I can't even begin to count how many times I’ve typed `Where` when I meant `filter` or vice versa.
What, `def main():`? Or do you mean the __name__ == "__main__" thing for distinguishing whether the code was imported?
I can successfully type "def main()" and not "function main()", unless I'm switching between JavaScript and Python and forget which language I'm in :-). It's the if statement with all the underscores that I need to check every time.
I actually made that a keyboard macro because, even though I remember the syntax, I can't be bothered with all the special characters all the time.
I set up some .NET services years ago. Since then it was just adding new stuff. If I was asked how to set up another service, I would have no idea how to do it
That's what `dotnet new webapi` (and all the other dotnet new templates) is for.
They don't even have a main() any more, it's great
Continuing the confessions: I do php.net/<function_name> at least twice a day
Is that a bad thing? I also call man <libc function> several times per day, most times I think I already know the calling semantics, but there is always some nitpick about the calling semantics or a bug description, that I didn't had in mind.
In python, it's trivial:
def main(): # code
The dunder syntax you see around isn't required.
It's the main benefit of Rust ;)
(obviously it's not but it is super nice that main in Rust is just:)
LLMs are wonderful for this. I can't write hardly a line of shell script without looking something up. And then there are three different ways to do <thing> so I spend time beard-tugging as to which way to do it. Now I just tell the LLM what I want changed about this shell script and look at what it comes up with. 100% of the time it's fine.
"Knowledge means knowing where it is written down."
It’s horrifying to learn that someone feels guilty for not fully obeying the commandments of Uncle Bob. It would be more appropriate to feel guilty for taking his advice.
Really appreciated this. It’s refreshing to see someone be this honest about their gaps and growth. A lot of us quietly deal with the same things, so thanks for putting it into words.
The majority of these hidden truths are due to senior engineering management in their 40s and 50s who have not coded in decades, and yet pick up the latest trend or fashion and impose that on their teams.
The monolith to microservices trend was one great example of this.
I can relate about OP’s comment about OOP. On a similar note, I had written about re-discovering visitor pattern:
https://ssg.dev/are-interfaces-code-smell-bd19abc266d3/
That blog post is a poor example. It replaced a simple straightforward if-else loop with a hard to understand abstraction spaghetti. A few unit tests and you'd be well on your way to shipping rather than messing around with beautiful patterns.
Obviously, because it was a fictional, simplified example for the post. My real use case was more complicated and involved multiple developers working on different parts of that flow.
The problem with if-else chains is it's easy for a programmer to forget to handle a case that another developer added in the called component. Unit tests can't help a spec miscommunication. But, visitor pattern can as it forces the handling logic to be complete.
Hence my example at the end using discriminated unions and exhaustive pattern matching in F#. Much, much simpler with the same benefits.
The section on cyberharassment is really troubling, although with the current vitriol on AI I'm not surprised. Do wish the author mentioned the name of the site though, if only so I can avoid it (and not in the Always Sunny "oh no terrible! where?" way)
I was as curious as you were. Turns out there are only so many popular threaded discussion sites in the vein of HN on the Internet, so an educated guess is all it takes.
Without making judgment on the actions of any involved party, I do wonder why the author would choose to bring up this incident and submit it as part of a story to a site where there is a significant overlap in readership.
That incident catalyzed the fear that suppressed my desire to participate online for months. I figured that if I couldn't talk about it now, I might never participate again.
Good on you.
Honestly, if there's any chance the content they posted on your profile before locking you out comes close to defamation, I'd consider talking to a lawyer about it. It could be that getting one to send them a cease-and-desist letter on your behalf could take care of the problem.
Well that sucks. It's exactly the site that comes to mind when I think "most popular alternative to HN".
I've generally found conversation there to be more respectful than HN, rather than less, when discussions get heated - so I had high hopes it would be a different site, but alas.
This leaves a really bad taste in my mouth.
Edit: you know what, screw it. In the spirit of "no more self censorship", here's the link: https://lobste.rs/~7u026ne9se
I don't find the conversation to be especially disrespectful. The people in the thread in question attempted to shame him, to some extent. Shame is a social measure to coerce people who are behaving contrary to society's expectations to change their behaviour. However, while shaming him, they did not especially resort to childish name-calling or ad hominem. They reasoned with him extensively as to why his behaviour was deeply undesirable. He went so far as confessing that he did not even know the language of the PR he submitted, yet intentionally withheld information about the provenance of the code. Sometimes the shame mechanism is misused for things that should not be shamed, but this seems like a clear case of shameful behaviour that deserves social repercussion.
Sadly, it seems like nothing was learned, since he settles only for diminishing his culpability in anti-social behaviour. He goes so far as to describe, in his blog post, his code as an "AI-assisted patch". When you profess that you don't even know the language of the code that the LLM generated, there is no "assistance" about it, you're at the deepest end of vibe coding. And in submitting it to an open-source project, you're making a maintainer spend more time and effort reviewing it than you did prompting it, which is not sustainable. Moreover, if the maintainer wanted a pure-LLM-generated solution, there was nothing stopping them from hopping over and typing in a prompt themselves.
In fact, most of the comments were purely a debate with no direct attacks at all. The extent of "not respectful comments" I see are something like...
"Pretty lousy behaviour if you ask me" is incredibly tame. If that's what counts for toxicity, then you're advocating for a toxicly positive carebear forum where nobody is allowed to criticise anybody else's decisions.I agree that the discussion doesn't seem to be toxic on the whole, though not superb, although I don't know what happened following in terms of harassment, so that's up in the air for me.
I don't know, I'd kind of like to see their responses before passing that much judgement on them.
> you're advocating for a toxicly positive carebear forum
Please stop putting words in my mouth.
I need to figure out how to be as open as the author is - it comes across as fricking amazing!
I really appreciate public vulnerability.
And I want to offer some contrast—not as a rebuttal, but just as a reminder that there’s lots of different ways to navigate this strange field.
The _majority_ of the paid code delivery I’ve done for a decade+ has been in Ruby. (The balance has been a mix of mostly devops and some TS/JS and Elixir.)
Remote work has been an utter boon. Admittedly, I do feel like it’s got worse since Covid. But I’ve been able to work with people all across the globe without uprooting my family and leaving my community, and conversely can travel without having to leave my job or clientele.
And I do find that some places benefit from thinking hard about their process. Small senior teams do great with Shape Up. Projects where you have a non-negotiable scope (replatforms) and work streams that are more reactive than planned do better with kanban than something involving estimates.
That’s not to say the author’s wrong! Again, just that the world is wide and experiences differ.
Some context here: I’ve consulted full time almost continuously since 2018, which certainly colors my experience.
Like the post overall, but the last section is a bit weird for "confessions" as it's all HIGHLY subjective. For example, I worked at a company where no one worked from home and we paired 100% of the time. When COVID hit, we started pairing over Tuple and I found it to be a superior experience to pairing in person (Tuple's drawing and attention drawing tools are far more accurate than my finger, I can use my own keyboard the odd time I want to control my pair's computer, and there are no office distractions of other pairs in the same room are benefits that come to mind). I continued to enjoy (and prefer) it for the 1.5 years I stayed after lockdown.
All the confessions are highly subjective. If someone tried a refactor like the one at https://refactoring.com/catalog/replaceConditionalWithPolymo... there is a decent chance it should get picked up and reverted on code review.
Taking a switch statement and spreading it out over 3x classes is not a general improvement, it is very context specific. It makes the code difficult to navigate because what used to all be in one spot and easy to read is now spread out who-knows-where and there might be a special case lurking somewhere.
It's not as subjective as it is more of a case by case decision. This example is quite misleading but polymorphic classes are sometimes useful when the domain grows, when you have have to update new behaviors all the time.. In that case then the switch becomes harder to maintain. Classes isolate behavior so new types don't modify existing code. I'd stick with switch statements in all the other cases. Sure, this could be abused and make simple things unnecessarily complicated but am just pointing out that there's a use-case for it.
I suppose I included the last section as a "confession" because I spent years pre-pandemic wishing I could go remote, talking loudly about how open office spaces were bad, etc. That, plus it's embarrassing to admit that I dislike remote work even though I am a remote worker. And not just a remote worker, but an enfranchised remote worker living 3 hours away from the nearest city with an international airport. I made the mistake of rebuilding my life around remote work before I had experienced it long-term.
Awfully nice, isn't it? You get all the privileges of being a remote worker, while "confessing" that remote work sucks to upper management who read your blog, giving them justification to deny other people the same opportunity you have. If you genuinely believe remote work sucks, you should own up to it, sell your property, and move to a city rather than reaping the benefits of it to the extent of your entire life being dependent on it while talking about how bad it is. The fact that you haven't done that indicates to me that you actually quite enjoy the benefits more than you suffer the drawbacks, but aren't acknowledging it, instead choosing to denigrate remote work for having any drawbacks at all.
> giving them justification to deny other people the same opportunity you have
I don't fear they'll deny others the opportunity for remote work. The company is "headquartered" in California, but I don't know if they even lease an office anymore. The CTO lives in the upper midwest, the architect lives on the east coast, my manager lives along the Mississippi River, and I live in the Ozarks.
> enjoy the benefits more than you suffer the drawbacks
Yes. I'm sorry, I thought I made that clear in the post. The benefits of remote work include, but are not limited to: no stress or time from commuting, an opportunity for geographic arbitrage, and the ability to build a better lifestyle around the lack of a commute. Beyond just the remote worker themself, a society that transitioned all office work to remote would also gain more benefits: more efficient use of real estate with entire office buildings rendered unnecessary, less chance of land value distortion due to centralization of workers, and less pollution due to fewer commutes.
I'm glad to also criticize in-office work for having other drawbacks. For example, I was rear-ended commuting to work more than once, the family needed the expense of two cars, we spent more on clothing, and the ambient level of noise being above 35 dBA was annoying.
I think that eating meat is immoral, I still do it most days. Remote work sucks mostly for the employer, not the employee.
Loved your post! I've been feeling the same way (currently feeling crushed by work+master's)... hope to work the courage to break the dam as well.
> left inner join
While I do appreciate this joke (and I do hope this is a joke), I've recently had a project majorly held up because a lead dev didn't understand SQL. It's great to admit gaps but it's equally important to close those gaps.
> As a hiring manager I interviewed software engineers and tried to filter for object-oriented knowledge. Retroactively, it’s clear I was hypocritical.
As some one who has been on the other side of "rejected by an interviewer who didn't understand the thing they've interviewed you about" I, again, appreciate the transparency, but I'm not entirely feeling that the lesson has been learned in the case.
There was a time in my life where I felt ashamed that I didn't know calculus... so I learned calculus and my life has been better for it. While refusing to admit ignorance of a topic is particular problem in tech, confessing that you don't know something and gleefully stopping there is not much better. Holding people up to a standard you do not hold yourself to is a major problem in this field. The technical people I've learned the most from hold you to a high standard and hold themselves to an even higher one.
Of course not every engineer has to hold themselves to a high standard, but if you want to write a blog about a topic, then part of the requirements here is that you do hold yourself to a high standard. Yes, we all have gaps, and we shouldn't let shame get in the way of learning, but we shouldn't let shamelessness about what we don't know limit us either.
I am indeed learning, working to close those gaps.
For automated testing, I'm in the middle of reading Developer Testing by Alexander Tarlinder, with xUnit Test Patterns by Gerard Meszaros coming close behind. I'm also working through Test-Driven Development: By Example with my wife as we have time.
For SQL, I read Grokking Relational Database Design by Qiang Hao last winter, and I started SQL Queries for Mere Mortals by John Viescas this week. Sadly, my flub with "left inner join" was not a joke.
For OOP, I've been on a whirlwind tour: OOA&D With Applications by Booch et al., Object Thinking by David West, POODR and 99 Bottles of OOP by Sandi Metz, Domain-Driven Design by Eric Evans, IDDD and DDDD by Vaughn Vernon, Design Patterns in Ruby by Russ Olsen, Clean Architecture by Robert C. Martin, and Smalltalk Best Practice Patterns by Kent Beck. Still on the docket are Design Patterns by the Gang of Four, PoEAA by Martin Fowler, Smalltalk, Objects, and Design by Chamond Liu, and Object Design by Rebecca Wirfs-Brock.
> confessing that you don't know something and gleefully stopping there is not much better [...] we shouldn't let shamelessness about what we don't know limit us either
I promise you, this was not gleeful and this was not shameless. Shame and fear affected me for months on these issues. And I'm not stopping there... From the end of the article: "I’m going to continue to work on skill building, but now I feel free to write about it. If [...] you’d like to help me fill [my knowledge gaps], [...]"
> if you want to write a blog about a topic, then part of the requirements here is that you do hold yourself to a high standard
A high standard of writing, maybe. But plenty of great stories come from those who are striving for a high standard, not just those already in the upper echelon. It's what makes this place different from academic journals.
This is great to hear and I do appreciate the clarification. Having put lots of content out in public myself (though this account is intentionally pseudonymous) I know it's also equally difficult to comment on content like this without being on either end of the "asshole" <-> "awesome!" spectrum, and sincerely hope my comment to not fall to close the the first part.
Not exactly your main point, but where’d you go to learn calculus? I did the usual classes in high school but none since, and I’d love to develop a better appreciation for it.
Check out Mosaic Calculus and see what you think: https://www.mosaic-web.org/MOSAIC-Calculus/. It's a free resource and it takes an interesting perspective on calculus pedagogy.
I find 3blue1brown to be a great resource to build up good intuition about math topics, his videos about calculus and linear algebra are wonderful in particular. https://www.youtube.com/playlist?list=PLZHQObOWTQDMsr9K-rj53... is his essence of calculus series, I found the visualizations made it a lot easier to grok.
I can second 3blue1brown as intro, but anything practical needs practise. Get your local 12th grade or undergrad math book which has problems to solve will bring more focus and faster learning. Have a list of problems to solve in your head while learning theory has been faster for me than just learning theory.
I got mine doing a computer engineering degree.
My suggestion is using Khan Academy if you want to better your math knowledge. It's really quiet good for that sort of thing. It was just starting to take off when I finished my degree. I wish it was available before then.
Calculus for Dummies is surprisingly not bad.
No offence to everyone else in this thread but the holy grail is truly The Art Of Problem Solving textbooks + mathacademy [0].
[0] https://www.mathacademy.com/
Learning stuff that doesn’t help in work(calculus is not helpful for 99% of software engineering) is really hard if you don’t base it in reality I find. Maybe it’s just me but I would never read a text book for fun so suggesting learning by reading a text book seems crazy. Calculus can be fun and interesting but the teacher has to actively try to make it so. The learning will take longer but you’re more likely to see it through and I think it’s more likely to stick long term too.
I learned way more reading crafting interpreters than I did in my compiler class for example.
That’s interesting because I can’t imagine learning a subject without a textbook. I have a hard time believing another medium would have the depth and density to get all the points across. Although it does depend on the subject matter and one’s learning goals.
But I also do read textbooks for fun… Now that I have a few decades of experience in a lot of these subjects I get way more out of the books. And I can start to understand more of the meta information. Like, of all the things the author could’ve used as an example, why did they pick that. Also, it’s hugely interesting for me to look at the homework problems and theorize why this particular problem was picked. Especially fun for electrical engineering books. But ya, I’m weird like that.
Youtube
About as helpful an answer as “book”
https://www.youtube.com/playlist?list=PLF797E961509B4EB5
https://www.youtube.com/playlist?list=PLDesaqWTN6EQ2J4vgsN1H...
https://www.youtube.com/playlist?list=PLDesaqWTN6ESk16YRmzuJ...
I do unironically think the simplest method to learn calculus is to go through a college level textbook. They are usually pretty accessible.
> I Didn’t Understand Polymorphism For a Decade
> And yet, my lack of awareness of polymorphism showed me I’ve been writing little more than structured programs. That I could replace conditionals and case staments with specialized classes had never crossed my mind.
> Polymorphism is covered in every college OO course.
Consider yourself blessed then because you're in for one hell of a ride if you pursue that path to its extreme. For me, it was the opposite: been taught OOP, I had to unlearn most of it to be able to better structure my mind and how I think about programs.
You should know what polymorphism is (also, there's static, dynamic, ad hoc, single dispatch, multiple dispatch), but I don't think it's a weakness if you have not been using it that much (the real weakness is over-using it and making a clusterfuck out of your code.)
Which is a long-winded way of saying that you could be doing much worse, if that makes you feel any better, lol.
Here’s a great idea for a good opportunist:
A “Confessions of a Software Developer” website where devs can come in and make anonymous confessions.
https://thedailywtf.com/
It would attract the humble-braggers. ;)
Probably already a subreddit for it like r/programminghorror.
I think Uncle Bobs advice is mostly bad and am afraid to admit it because it’s like a (cargo) cult now.
Agreed. The quote from him in the article reads like a caricature of a comically overzealous SWE.
Add most of what Martin Fowler said to that list.
Refreshing to read, I bet it was cathartic to write. I hope your fears don't come true. I think they won't. Many people do genuinely appreciate this kind of honesty, even when directed against them, but it is a gamble.
A good reminder that everything we say/hear/write/read exists in the unseen context of all the things we believe we should not say.
Remote work is great (for the reasons you gave and more) and saying it "sucks" made me roll my eyes, and it's reductive in the same way as saying office work "sucks." I wouldn't have had a job if in-office was the only option. It certainly didn't suck for me.
Being bad at problem solving with people far away is just another problem you can solve with practice. Same as being bad at problem solving even when help is right next to you.
> made me roll my eyes, and it's reductive in the same way as saying office work "sucks."
Yes, "remote work sucks" is reductive, but I elaborated beyond the heading. Also, I wouldn't disagree with "office work sucks." Remote work simply has its warts, too.
> just another problem you can solve with practice
Perhaps, but practice alone clearly isn't enough. I've been working remotely since 2020 and it hasn't gotten more enjoyable. I would love to solve that problem, though. I read Remote: Office Not Required by Jason Fried in the past, but that was written a long time ago. I've added more recent works (Effective Remote Work by James Stanier and The Async-First Playbook by Sumeet Gayathri Moghe) to my reading list.
I’ve worked remote for at least 16 of the past 22 years including my first job out college. It’s always been friggin awesome. The only downside was when I was contracting and I’d get calls in the middle of dinner and I didn’t have the self-discipline to ignore the call. A few times a year I have to travel to work, it’s nice to see folks, but it’s not required to get the work done, I put my big boy pants on and figure it out, or ask for help when I can’t.
> it’s easy to form an enemy image of somebody at the end of video call, but difficult to keep that image when you share a room with them and sense their pain.
I'm honestly so confused by this. Has the author never worked in an office before? Building a grudge for someone that you are forced to work with and sit next to all day is one of the classic office dilemmas. Being forced to be around them all day can really build resentment to people
> Remote work sucks
Work sucks in general. Remote work is of course not perfect, but its problems need to be compared against non-remote work problems..
Remote work isn’t for everyone. Their point of view is just as valid as your point of view.
And this is my biggest complaint about arguments about remote working. People turn it into something that’s evidence-based when actually it’s a deeply subjective topic and thus different personality types thrive in different working environments.
People have painted themselves in a corner re: remote work and get wacky in discussions about it. Lots of emo, not much fact.
My point is you cannot have any fact based discussions because statistics are generalised whereas people’s ability to thrive in office or remote work depends deeply on both the company culture and the individual.
Thus it is always going to be an empirical argument rather than a fact-based discussion.
I am dumber for having tried to understand what I just read.
If you ever feel bad about yourself as a programmer you can go read some Rasmus Lerdorf quotes to cheer up :)
Or work with a random programmer at a random company for a bit. We had to do some audit/estimate for a company with huge tech issues (race conditions causing data corruption, huge slowdowns with just small usage spikes etc) costing them clients/money. The company runs around $50m/year or so selling software to enterprises. Anyway; software backend written with Java/Spring, deployed/updated on EC2 manually, no automated tests (zilch). Frontends with Vaadin. Almost no processes are used, just Jira + tasks and then 'start at the top every morning'. No one knows sql anymore (Hibernate), no one knows html/js (Vaadin) and, even though most people are senior and there since the beginning, no-one has done anything high level on the job in the past 20 years or so. They have just been inside this 'ecosystem' writing code and it works. Old Java with some modern updates just to satisfy the compiler/linter (but not fully understanding why that nonsense is needed). None of the core seniors I interviewed touches computers outside of work, they had 0 tech courses since working there etc. They are all 9-5 code producing robots. I want to bet they can mostly/all be replaced today by Claude Code, of which existence, of course, they are not aware (they did chatgpt but not Codex or Copilot). We have since found so many issues in the code. Yeah, I do feel very much uplifted about my own skills after encounters like this, and these are by no means rare, i would rather say; extremely common.
Unlike OP though, I cannot be as open about these companies as we would definitely not have any clients left after.
That sounds amazing, what you can get away with while still shipping a product and getting paid. In some way probably the engineers there are unwilling to do any automation in fear of becoming redundant, and the company is still fine with that.
Also, I am not sure how not touching computers after work is a bad thing; people can have families and other hobbies?
> computers after work is a bad thing; people can have families and other hobbies?
No not bad per se, but it did clearly show that, without on the job courses, why all of them are stuck in the early 2000s tech wise.
Some people start with a company and get lucky with early success and then get restricted because of that success: get new clients via existing, everyone likes it and asks for new features and without noticing it you might find yourself 15 years down the road with ancient tech and no one understanding anything current. Then you can still thrive if your clients like it... we have similar clients: a 1980s factory, another 1980s factory, a logistics app from the 1990s etc. Things deeply ingrained in some vertical, expensive but better priced than the SAPs etc of this world so it keeps going and going.
> Remote work eliminates a lot of problems with office work: commutes, inefficient use of real estate, and land value distortion. But software development is better when you breathe the same air as the folks you work with. Even with a camera-on policy, video calls are a low-bandwidth medium. You lose ambient awareness of coworkers’ problems, and asking for help is a bigger burden. Pair programming is less fruitful. Attempts to represent ideas spatially get mutilated by online whiteboard and sticky note software. Even conflict gets worse: it’s easy to form an enemy image of somebody at the end of video call, but difficult to keep that image when you share a room with them and sense their pain.
Every ounce of data proves this statement wrong. If you feel like you work better non-remote then do it. Don’t shill it as a panacea. I’ve been remote for 11 years now and if I wasn’t I wouldn’t have been able to take care of my family, go back to school part time, work on my health with better meals and reasonable gym hours, etc. even IF in office was better for the employer (even though all data says it’s not in terms of productivity) it is unequivocally better for the employees life to work remote as much as humanly possible.
This hot take is just simply insane. Humanity had no problem coordinating massive projects over IRC and mailing lists. It’s clear the author is a “nu-coder”.
> Remote work eliminates a lot of problems with office work: commutes
> Every ounce of data proves this statement wrong
There is no need for hyperbole. Because one thing we can all agree on is that remote work eliminates commutes, by definition.
> software development is better when you breathe the same air as the folks you work with
Ever heard of flu season? What if you have a family and don't want to bring diseases home?
> Attempts to represent ideas spatially get mutilated by online whiteboard and sticky note software.
Right... like, the Linux kernel team? Or any of the major open source key pieces of technology you use? Built by large teams that worked remotely for decades even when tools where orders of magnitude worse than the current state of the art? Some of them never meeting each other in person?
---
Remote work DOESN'T suck. YOU make it suck.
Remote work is great if you care about shipping.
Want to go for coffee or want to talk about our weekends? No thanks.
Did you see the distracting thing outside the building? No, I didn't because I don't have to go there anymore.
Is the heat too high or too low? It's your own home, just adjust it to YOUR convenience.
Worried about your pets being alone? Just be next to them. I care more about my pets than some stranger from work.
Want to be loud and flex about random stuff? Log into LinkedIn and talk to the other geniuses like you while I focus on doing my job.
Most people SUCK at drawing, suck at calligraphy and their whiteboard diagrams SUCK. Therefore, whiteboards SUCK. Unless you have great calligraphy and drawing skills, whiteboards are not helpful. You are just sad because you are not getting attention by being in front of other people looking at you.
My confession is that I actually love torturing people in coding interviews. Sick I know, but these are 6 figure jobs they are applying for.
I don't know that it is necessarily sick ... I suppose it depends what you do, how the people respond, and if you are playing a useful role in the interview process for your organization.
Tell us more?
On the positive side... Seeing how people respond to pressure (done within some bounds) might be useful information. Watching people dig deeper into their problem-solving toolkit (and even their determination and resolve) might be informative. But I have not recently reviewed interviewing research, so I don't know how strong the causal connections are between observing these things in an interview and on-the-job performance.
On the negative side: If one is primarily interested in finding flaws as some sort of intrinsically valuable thing or to boost one's own ego, these are probably hints to look closely at oneself and reflect and make sense of what is going on inside.