ruh-roh

Yes - that mental picture of a divorced dad watching old family photos come to life, while sitting alone on his couch - is just gutting. Like a literal punch in the gut.

But having said that, is "loneliness" the correct (or only) takeaway on that? I mean, there are people toying with the idea of uploading emails and documents of the deceased to an LLM so relatives can still "talk" to them.

I don't have a word for what that kind of world looks like.

I joined a company last year that has a handful of legacy products, and we successfully moved one of them to a CD model a few months ago.

It's hilarious to sit back and watch every other team have the same monthly loop of big change -> lengthy test cycle -> "go-live" meeting -> fixes/big change -> more testing... etc. Loads of stress and finger-pointing.

Hardly ever hear a peep from the CD team. Sure, it's not all rainbows and lollipops. They do have urgent production issues time-to-time, but most are resolved by a rollback (or feature toggle flipped).

I used to travel ...a lot... Minimum twice a week, usually 4+. As an engineer (and process consultant), I spent countless hours sitting at airport gates looking for any optimization I could find, mostly focusing on wasteful and (seemingly expensive) boarding processes.

My favorite half-baked + stupid + fun one: Install the plane seating area at the gate itself, as a separate detached entity on a conveyor belt. Passengers can take their time, it's open, there aren't walls. You can just walk right into Row 20, seat A. No squeezing down aisles. Once everyone arrives and takes their seat, the belt moves and shuttles the seats+passengers into the fuselage. Kinda like a mini-roller coaster. Shut doors. Take off. Repeat in reverse after landing.

(There are obviously serious problems with this. You'd have to redesign planes and gates, at a minimum.)

The "once they start trying to change things" aspect is where I'm most keen to watch.

Surely we have all been involved with large-to-small systems where we were curious how well/long they would stand up on their own without intervention. It's kind of fascinating to watch a real-world example, even though we likely won't get a lot of detail. But still, :popcorn_emoji:.

But what's getting somewhat-ignored in this discourse is they aren't changing anything. All deploys are frozen. And the new CEO has made a ton of promises about new features and policies. Eventually they need to roll those out, and once they do, damn... Even if they have strong automated tests, those generally only tell you that new code is functionally ok. Anything that crosses interfaces, deals with end-to-end integration, or adds new tables/dbs/jobs/etc that would eventually hit their scaling solutions - all could have catastrophic effects.

(And that ignores the consent decree stuff on changes too.)

This is interesting - I have been passively monitoring LinkedIn and a few others for senior engineering management roles for the last year or so (VPE, DoE). I am not really looking, I have only pursued a couple of them, but my feed/inbox is noteworthy. Many of the "openings" have been out there for multiple months, some a year+. A good handful are tweaked slightly every few weeks to appear new.

I can understand fishing for individually contributing junior-to-senior developers, but for senior management roles? Would a company really troll waters to find a senior leader that they don't really need? "Hey, look what I found, let's find a spot for him/her!"

My theory has been either (1) those companies are super picky, which is understandable given the impact of those roles, or (2) they don't really know what they want, and they are using the interviewing process to figure it out. But regardless of 1v2, those companies likely have people fulfilling those management responsibilities today that they have otherwise deemed unqualified, or are very-senior C-level folks that are too busy to do it well.

Is there a term for the phenomenon where a supremely powerful person can comment on some event that may happen, perhaps innocently, and the fact they said it makes it more likely to happen?

In many ways, the health of an economy is a state of mind and emotional. Bezos has millions of people following his every word; many worship him and believe he has the keys to success (could also say the same about Musk & Gates). A pessimistic statement like this, regardless of merit, can/will cause others to be pessimistic.

In a vacuum I would be very interested in Bezos's thoughts on this topic - he's clearly super-intelligent. But I don't think I could get past his powerful position and the many reasons why he may want to sway opinion.

Comment on your edit: I don't disagree on the potential for a worse technique or tech on either side, but isn't there also potential for the opposite? As in, exposure to a technology/architecture/process at Company B that might help Company A.

As a consultant, I find many clients are genuinely interested in how other companies solve similar challenges.

While I generally agree about the annoyance with UI changes, and how that grinds on people, I hope we don't just handwave over the "CD for bug fixes is fine" bits.

It varies product-to-product obviously, but I'd guess there is also a significant chunk of folks that are pleased that their functional annoyances/blockers (bugs) can be fixed quickly, without waiting for a monthly or quarterly release.

These folks might not mind the fluctuating UIs, if it means they get the functional value they need asap.

I am genuinely interested in this, please excuse my naivety.

My expectation would be that new Product work would, at some high-ish level, be sanctioned by company leadership. Or you wouldn't be spending weeks/months/years on it. Ideally Marketing would be involved in that leadership decision, and this is where they could raise that kind of concern. But even if they are not, Marketing's "boss" (CEO) could/should tell them they need to market it or go pound sand.

On a whim I picked up The Metaverse, Matthew Ball's book released last week (7/19). I am only about 1/3 of the way through it, so I can't recommend it yet, but I have two main takeaways so far...

1. If you can get through the marketing slog thrown out by FB, Epic, and others, many ideas behind a/the metaverse aren't terrible. Perhaps even good. As an old fogie myself, I cannot see spending hours/days in it, but I have to acknowledge there will be a market for it. (Granted, the author is a VC, so he might be a little biased here.)

2. We are a LOT further away from a functioning metaverse than people think/expect/hope. Many challenges exist across the tech (networking, storage, hardware, processing power, etc.). Not to mention identity, governance, and interoperability. Surely many more. Predictions in tech are always foolish, but I would be surprised if we approached anything like a 'Snow Crash' metaverse in the next decade.

> If we can't make the cars just as smart as an alert and capable driver, then maybe we need to meet halfway and make the roads a littler "dumber" (simpler) to accommodate the robots.

This aspect of FSD has always fascinated me and I'm a little surprised it doesn't get more discussion. Meeting halfway. At what point could/would/should FSD influence the environment around it?

For example - a poorly painted road sign*. Tesla/Waymo could say "We cannot support L5 FSD on this road until you fix this sign." If it meant a step forward in autonomy, Tesla/Waymo could even offer to share the cost of that improvement!

There are a million reasons why implementation of that would be problematic. Costs and incentives would be all over the place. But I am more interested in the framing: The machines are the ones that need to adapt. Which is essentially hoping for continued hardware improvements or a spaghetti mess of if/else statements. ie "do this weird thing if you see this other weird thing in front of you". Can we get rid of the weird thing and avoid the engineering challenge altogether?

* Yes, this is an overly simple example. Some environment changes could be so large that they would require a full redesign of a city/buildings/traffic patterns. But surely there are classes of improvements where some are easier than others.

Last Spring (21) when we thought the pandemic was subsiding, we started optional in-office days. I went in one day, stopped by the kitchen, and bumped into a peer I had not seen in a year. Not even on Zoom. Our teams had worked together on something a year or so prior, but the paths of our products diverged. We never had a reason to interact after.

And I'll tell you what -- it was _amazing_. A typical "5-second good morning" conversation turned into a 10-minute catch-up. Which turned into a 30-minute system design discussion. I loved it, we were both juiced. If you could guarantee that all in-person connections in an office would go like that, I would give up remote work in a heartbeat.

Unfortunately, that's not how it works. People get used to the grind, the commute, and their office environment. They don't (generally) bring that kind of impromptu excitement every day. Our interaction was a mix of curiosity, pandemic anxiety, and joy in our work. You can't snap your fingers and expect that to materialize every day.

The bean-counter in me can't justify the millions of dollars in the annual office spend to (hopefully) enable connections like that. Which is why I am fully on the quarterly-or-semi-annual-retreat train. Ditch the office, and point that money towards infrequent in-person events. Pay for people to come in from wherever they are (if they want to). I suspect that's a _much_ better ROI.

> Don't project into the future. What is now is not indicative of the future.

This is fantastic.

Nobody cares that the neighbor kid started walking at 8 months while yours hasn't yet. It doesn't mean anything. You'll barely remember yourself in a year or two.

Is it really not thriving? Surely it's relative vs SF/NYC, but when I got here in the 90s, it was all just banking & insurance. Also advertising, but very little tech there. It was super frustrating at the time, but the last decade has shown quite a bit of growth of "modern" startups, imo. It may not be those other hubs, but as Mr Bourdain would probably say, Chicago doesn't care.

On the weather... "A nice day in Chicago is a nicer day than anywhere else in the world." -- Anonymous.

75 degrees with no humidity is heaven here, and _everyone_ you see feels the same way. Outdoor dining, festivals, parks, etc. People grind through January just to see June, and it shows. 75 degrees with no humidity in San Diego is just another Tuesday.

I am not going to wade into the crime chatter all over this thread, people have their own stories and barometers on that stuff.

But I cannot let Bean anecdotes slide without an alternate take. We take our kids (two under 10) down to the Bean/Millennium Park area 4-5 times a summer, without fear. It's a fantastic place for families. Super diverse, good exercise, breeze off the lake. The Bean + the fountains + Maggie Daley Park can be an awesome Saturday/Sunday afternoon.

Oh 100%, this drives my consultanty-productivity-efficiency brain mad. I get doing this so people can check that they are paying for what they ordered, but mistakes/inaccuracies are exceptions. The vast majority of restaurant payments have to be correct and should be seamless. Seems like we should be optimizing more for the latter. Not that the former isn't important, but the process seems skewed inappropriately.

I made a rule with my phone - only read content that was created > 1 year ago. This immediately rules out Facebook, Twitter, and (usually) WashPost/NYT. I am left with academic papers and books. (Podcasts and audiobooks are ok.)

Sometimes more of a guideline than a rule. I make exceptions for new works from authors I enjoy, or longform recommended web articles.

Using silence strategically is a superpower.

Best leader I ever worked with had a "7 second silence" rule. Just don't say anything for 7 seconds. Requesting input on a topic/idea in a meeting? Wait in silence for 7 seconds. Dropping an important point in a presentation? Wait for 7 seconds. Not sure of the answer to a question? Wait for 7 seconds. If you have doubt about what to say or do? Just wait and breathe for 7 seconds.

It is extremely difficult to do in practice. But it is crazy what a breather can do to help you organize your thoughts. Or what other people will do to fill the void.

(Note: It's not a novel idea, I've seen other folks use 5, 6, or 8 seconds. imo 7 is _just_ on the border of uncomfortable.)

I am probably 95% in line with this, SAFe is mostly terrible, but there are some nuggets in there that I think are good. Mostly around the cadence and checkpoints on progress in delivery. (Yes, we can argue whether some of these things originated with SAFe, but it's how I was originally exposed, so that's how my mental model is mapped.)

- Quarters (every 12 weeks): Big ticket things that should move the business forward. This requires Engineering participation because items like technical debt, system re-architectures, pipeline improvements, etc. should be on the table. - PI (every 3-4 weeks): Light checkin on progress to current items. A ballpark estimate on what can be delivered in the upcoming PI given the current constraints (capacity, quality, etc.). This works out to 3-4 checkpoints per quarter. - Sprint (every 1-2 weeks): Real commitments to tangible requirements are made here. If thorough discovery leads to changes to the ballpark estimates or development is delayed for other reasons, it is communicated at the PI checkpoint and forecasts can be adjusted.

(imo) SAFe is just a solution to problems that may or may not actually exist - eager-beaver middle management-types adopt it chiefly because they think it makes them look good, but also because Product/Engineering teams don't commit-escalate-communicate well.

I encourage folks to take a step back and ignore the specific SAFe guidance, but do consider the value of the inputs & outcomes.

More than two years into this, "*returning* to the office" is incorrect framing in a lot of cases. I have been hired via Zoom to companies out of state TWICE in this period, each headquartered more than 1k miles away. I had/have no office to "return" to. If either decided to roll back their remote policy and require physical presence, they are essentially firing me (along with ~100 other folks).

I guess I am on the other side - the more remote permeates, the harder it will be to claw it back. Not just because people like it and are now empowered, but also the practical/logistical headaches involved.

I don't feel that the realities of rent/utilities/snacks get nearly enough attention in conversations like this. If you have more than a handful of employees with office space in a major metro area, rent is easily equal to (or more) than the cost of multiple people. If you are looking to trim costs in a bad economy, or on the flip side - grow your company efficiently, rent seems like an easy spot to compromise on.

The FAANGs have spent millions of dollars building or leasing their current offices, it's not hard to see the Finance/Accounting folks escalating that those investments should to be utilized. (Not saying that it's a primary driver here - but surely it's a factor.)

"Number of engineers" has become another vanity metric in the VC ecosystem and it needs some pushback. I have viewed of a ton of tech leadership job postings in the last year+, and almost all of them have a requirement/bullet of "grew Engineering team from X to XXX", or something like it. As an engineer, building & supporting products with the right / a limited amount of engineers should be the goal. The Whatsapp example is fantastic and should be lauded. It might be extreme, but we need more stories like that, not less.

I get that most view it as a proxy for business growth, but I fear there's a growing group of non-technical investors/leaders/founders who see FAANGs with 4000 engineers, so you're only successful if you have that many too.

This is good advice for multiple reasons.

One I haven't seen mentioned yet - When Product is accountable & responsible for testing the outputs, they will understand the effort required and can therefore prioritize investment in testable systems and associated test automation.

When those aspects are punted over to architects/developers/QA, you'll end up in a constant battle between technical testing investments and new features.

When I had a train commute in the before-times, the one thing I really leaned into was the train ride home. It was about 30 minutes. I used that time in a variety of ways - reading, podcasts, napping, meditating, drinking-a-beer. I let my mood dictate, I intentionally didn't over plan it. Just no work. This naturally let me slide into the non-work/dad/husband version of myself. More often than not, this recharged me, and I'd be in a good place when I walked into my home.

I admit to struggling with this with the pandemic and full WFH. I'd find myself working early and late, answering emails after hours that I really didn't need to, etc.

My solution? I bought a super cheap treadmill for my office, with a stand for my laptop. I don't really run on it. I get on it in the morning and walk for 30 to get my day started, 30 at the end to close it down. I think I still do a little too much work during this time, but it's generally prep for the day, pruning my todo list, checking out tomorrow's calendar, and so on. It's not the same as the train, but it's a lot better than having no boundaries.

In a weird way - this kind of thing has pushed me towards companies that aren't anywhere near my current location, even/especially international. It makes the conversation super clear, recruiters & hiring managers can't dance around it. International brings in the prospects of visas and such.

I am in the Midwest and just joined a company headquartered in SV in a remote role. If they had tried to pull some remote bait-and-switch I would have just chuckled and asked for first class tickets in/out one day a week, or asked for the moon in relocation fees.

My father has given me a version of the same advice in strange, disruptive, crazy economic times over the years: "At a certain level, the economy is just a collective state of mind. You can contribute to it, be a slave to it, or take advantage of it."

Look - I have small kids too, so I'm not advocating that you ignore it. Everyone with dependents should be proactive in considering how they are supported (in "good" economic times too). There are just so many interconnected pieces of a global economy that a wide brush of doom is far from inevitable.

The first question I would consider: What is the funding source of the company that I work for? Are we public and thus directly dependent on market swings? Or are we VC or private-backed, which is indirectly affected but could also weather the storm if forward-thinking? Or are we private + bootstrapped, which is generally a good thing until your customers start fearing the doom-and-gloom economy too.

When I took a full remote position late last year, I also invested in a cheap treadmill for my home office, plus a laptop attachment. I get on it first thing every day, and do my normal inbox processing, generally at a pretty slow pace. Responding to slacks, emails, PRs is easy in this mode, no different than sitting at a desk. Lately I've added a 30-minute end of day shutdown walk on top of it. Review my todo list, check tomorrow's calendar, etc. Altogether, I average 2-3 miles per day.

I haven't been able to consistently operate on it for deeper work/code or meetings, but I don't know if I really need to.

Early in my career I just did the standard '2 weeks' thing, but as I progressed I extended to 4 (and in one case I gave 6). In my mind, I thought it was respectful to the company, helped soften the blow on finding a replacement, gave loads of time to recruit/promote and transfer knowledge, etc.

But after reflection, that was a mistake. 4-6 weeks is an eternity in product & engineering time. Too much time to sit around and work-but-not-really-work, and was ultimately disrespectful to my staff, peers, and myself.

My current place has an unwritten rule to not allow notice periods of longer than 2 weeks. When someone resigns, just cut it off and move on. We suffered in knowledge transfer in a few cases, but that's been rare. If it takes longer than that to transition, we probably have other problems.

Not sure what you mean by "regardless of timezone".

I was lucky to be in a foreign country during an Olympics (Australia). They had no delusions of grandeur (except maybe in swimming). They interviewed winners regardless of country, they told stories regardless of country. They showed events live.

Dick Ebersol stubbornly kept the standard practice of tape-delaying major events to prime time in the US, long past viability in the Twitter/ESPN/Google ecosystem. In some cases (swimming), they have lobbied to change the schedule so live events can be live in the US prime time window. I believe Michael Phelps's 2008 Beijing races were at 7am local time. Surely other countries loved that.

Starting new at a startup-with-traction several years ago, I had reason to check out the feature flag configs for my first feature. It was hilarious.

- There were hundreds of them, some of them going back to the garage days, multiple years old. Some would turn on/off major functionality core to the product. For example, this was an ecommerce product - one toggle was "show/hide the buy button". (I think that one is staying in.) - This was all hand-rolled, pre-LaunchDarkly stuff, but at least they were all in one place. I diffed Production, Staging, and a one-off UAT environment - the toggles were MASSIVELY different across each.