411 karma · joined February 10, 2013
linkedin: https://www.linkedin.com/in/alexandercrooks
it has increased my speed to market massively. I mostly work on enterprise CRUD apps so getting things like SSO and API for free on Supabase is a great time saver
and then some apps require more specialized functionality and for that I’ll leverage the AWS stack
Due to the nature of LNG storage, a small amount of the gas “boils off” every day (fraction of a %). The longer a voyage, the more cargo is lost. So LNG carriers are incentivised to deliver cargo as quickly as possible so they can maximise the delivered value of the goods.
(Fortunately because these ships themselves are LNG powered, they are designed to recapture this boiled off fuel to operate the vessel)
But sometimes I come across a company that has set up an automation flow where you get a dozen emails over the span of a couple of weeks. This needs to stop. The only thing incessant spam will do is guarantee you get on my auto-delete list. A couple times a year is one thing. A couple times a week is unacceptable.
It took less than 2 hours of effort to update all of the code to work with the latest version.
One option is to call their bluff. But you should only do this when you have a fallback, so you should be able to push for a few week extension to make a decision. Since they'd expect you to move by August, there is surely some wiggle room for making a decision. And then start interviewing elsewhere, and tell them "I'm not moving" when you get an offer, and see how they proceed. They might say "ok", in which case you can review the benefit of staying vs. taking the offer. They might fire you, in which case you can take the offer and push for a payout from Tesla.
You could also propose a solution where you travel regularly (say 1-2 times a month) to a Tesla office between September and December, in advance of a January move. It's important that you are clear that this is approach must be funded by Tesla - flights, hotels, per diem. They might not accept it, but at least it potentially kicks off a negotiation conversation.
Regarding severance - it really depends on your employment agreement (which might specifically call out severance) + jurisdiction of employment. You may live in an at-will jurisdiction but have a contract that guarantees you a notice period / severance. Your contract also may say you are an at-will employee with no severance, but live in a jurisdiction where employees are protected (in which case the contract clauses may be void). All of this is highly dependent on your individual situation and for that you should seek legal advice.
My brother has been an addict for his entire adult life - for the last 20 years. His drug-of-choice started at cocaine, then crack, then heroin, and for the past few years has been exclusively fentanyl. So I have a lot of first-hand information perhaps biasing this point-of-view.
There was a big spike in deaths related to contaminated supply between 2015 and 2018 [1] which led to a lot of media coverage [2], which is completely understandable: many ordinary citizens ("normies" as my brother calls us) partaking in recreational drug use lost their lives during this period and so it created a lot of buzz.
You see though that there's another big spike starting 2020 [1], particularly starting April 2020 [3], which is not backed up by any growth in media coverage [2], suggesting that it did not have as much of an impact of "normies". And during this same period, there was a additional government assistance money available to addicts, beyond the normal welfare payments, which allowed addicts were able to purchase and use more drugs. Overdose deaths are 50% more likely in the days following receiving government assistance [4]. And this makes sense, when you're flush with cash, you can do more drugs. When you're an addict, you'll do as many drugs as you can. When those drugs are fentanyl, you're constantly teetering with death.
My brother claims to use a naloxone kit nearly every day on somebody - all people who are part of the community. And they're used on him very regularly too - even as an experienced addict, he overdoses regularly. In fact, he sometimes does this on purpose, because the high of an overdose is apparently second-to-none. But he says that he only does that when there's somebody else with him who can hit him with naloxone before he dies. This is not uncommon. I know a firefighter at the fire hall: regularly addicts will bash on their door and then immediately shoot up a huge dose, because they know the firefighters will rescue them.
That's not at all to say that a contaminated supply isn't the problem now: I'm just not convinced that it's the primary factor in deaths today.
[1] Page 3 of https://www2.gov.bc.ca/assets/gov/birth-adoption-death-marri...
[2] https://trends.google.com/trends/explore?date=all&geo=CA&q=f...
[3] Page 4 of https://www2.gov.bc.ca/assets/gov/birth-adoption-death-marri...
[4] Page 7 of https://www2.gov.bc.ca/assets/gov/birth-adoption-death-marri...
I've had a mix of both. The times that I had I micromanager were the most soul-sucking experiences of my career.
They're definitely out there.
> Now we move on to a completely different topic, and maybe not the most pleasant one: about when engineers get fired.
So in context of the article’s purpose I think it is a fair representation - I very rarely see firings related to “wrong person for the moment” for engineers. But I very commonly do see firings for poor performance in engineering which, as the industry’s overzealous interview practices may suggest, is difficult to discern in an interview.
Even if you disagree with my perspective, I do still think you do yourself a disservice by jumping to the hyperbolic “it’s terrible” instead of writing your feedback in a more constructive manner, like you did in this subsequent comment.
So I think there are three downsides for me:
- The routine of going to the office daily made it really easy to fall into a healthy routine at the gym. I used to bike to work, and did at least an hour of HIIT every day - I was in the best shape of my life for a long time. It has been comparatively difficult to bring that rigour to WFH, despite objectively having more free time.
- There is definitely less connectedness throughout my organisation, and it likely has an impact on the quality of work we ship, how on-time we are, how much we understand the customer. I think during the office-primary days, there was so much learning through osmosis - overhearing another team solving a problem a few rows behind you, talking to other departments socially over lunch, etc. And so much of that is missing now, so I more often encounter situations where people aren't on the same page.
- It's so much easier and faster to solve problems in person. Encountered with a problem now, I spend a lot of time doing research, reaching out to people on Slack, avoiding creating more meeting fatigue, and when scheduling meetings needing to wrestle with availability. Whereas before it felt so easy to wrangle together the correct subject-matter experts for a 5 minute discussion.
It's really difficult for the industry to target the right spot of the supply/demand curve. A new-build project takes several years they need to commit in advance to the shipyards how many vessels will be built as part of a new class. Meanwhile, many other companies are simultaneously building vessels. And it's impossible to predict events like the port congestion in LA that drastically reduced supply for a period of time. And it's difficult to predict macroeconomic changes in demand. And a vessel might cost a hundred million dollars, so the operator isn't going to let it sit around and do nothing - they're going to put it to work and make as much money as they can.
I find it's an easy way to give the customer more budget certainty upfront and offer some level of flexibility. So as the customer asks for something out-of-scope that's small, I can just draw down the hours pool instead of debating between issuing change orders and eating the cost. And then you have some form of contractual protection saying "oh hey, your 10 tweaks took up the entire pre-agreed contingency, it's time for a change order" which is a much easier conversation to have.
- Because of my focus in particular industries, I tend to start projects already having a baseline of understanding of what my customers need.
- You can write a fixed price contract that leaves room for a customer to change their mind / not know what they want. If you articulate very clearly in the SOW what you will deliver for the fixed price, then you leave yourself open to issue change orders as the scope evolves. Of course, you need to be careful how militant you are here, because issuing a change order for every minor thing won't help you get repeat business. To hedge against this, I tend to include a bucket of hours for arbitrary changes. This gives the customer some room to change their mind still without allowing them to derive and negotiate against an hourly rate.
- In my experience, fixed price allows me to 2x or more the hourly rate I would otherwise be able to charge, so if I do a bad job at writing an SOW then I have a lot of breathing room between what I'm getting paid and what I likely would have gotten paid had I billed hourly.
RE: Pandemic: I don't develop the work product on customer sites, so I'm only really there for meetings (so a lot during discovery and a lot during delivery). I've found that it's been pretty easy to migrate this to Zoom/Meet albeit with a bit more friction. It's a lot more difficult doing discovery remotely, but I find running workshops in Miro has been invaluable to ensure I'm capturing everybody's voice.
My consulting mostly focuses around IT/digital strategy and the work product usually involves connecting together incumbent systems, and sometimes implementing new systems while ensuring they’re deeply interconnected with the rest of an organisation, aligning them closely with biz process, etc. My customer is always a business unit lead so the budget tends to be more aligned with top-line outcomes.
Maybe it’s related to your location. I’m in Vancouver.
They tend to have fairly quick approval timelines and don't cost very much. This visa would allow you to try out Australia, build some connections there, and see whether it's a good long-term fit for you.
Within a route, it will suggest all the different options to get there. You can judge at a glance how much availability there is by looking at the available seats in the fare buckets. This is for an arbitrary flight from YVR to LHR on Air Canada:
J9 C9 D9 Z9 P9 R0 O8 E7 N3 Y9 B9 M9 U9 H9 Q9 V9 W9 S9 T9 L9 K9 F0
An airline splits its seats into different “buckets”, and the list above is essentially ordered from most to least expensive. J, C, D, Z, P are for business class seats. O, E, N for premium economy, and Y and beyond for economy.
J9 means that there are at least 9 tickets available in the most expensive business class bucket. N3 means that there are exactly 3 tickets available in the least expensive premium economy bucket.
Industry-wide, J means business class and Y means economy, but individual airlines tend to have their own fare coding within there.
So with all of those nuances in mind, you can infer how full the flight is: the second least expensive business class bucket has at least 9 seats available, and the lowest one has 0, so it’s pretty quiet but not empty. Premium economy is probably 50% full. And economy seems to have a lot of space (all but the cheapest bucket have more than 9 tickets).
Conversely, this is a sample of a fairly full flight (today on AC, YVR-YYZ):
J0 C0 D0 Z0 P0 R0 O2 E1 N0 Y4 B8 M6 U6 H5 Q3 V2 W2 G1 S0 T0 L0 A0 K0 F0
Expert Flyer also allows you to review the seat maps for each flight you see, but of course it’s more time consuming, because then you need to click into each option. So it’s best to learn the above syntax, and then use the seat maps to validate. It’s also worth noting that the seat maps aren’t foolproof: many passengers don’t select seats until they check in (because it costs $ otherwise).
Looking at the seat map for my first example above, I was able to validate that business class had 14/30 seats selected, premium economy had 12/21, and economy had ~70/~240 seats selected - lots of open space. A really good chance of having a whole block of seats to yourself.
Also another interesting fact: sometimes you’ll see something like Y3 but the flight is actually sold out. This is how airlines oversell their flights. Y is the most expensive economy fare, often not too far off from a cheap business class ticket - so it’s super profitable. So the airlines will generally keep a couple extra seats available for booking on a full flight. In the best case, a few people miss connections and they’ve pocketed all of that extra money. In the worst case, they need to bump one of the lowest paying passengers and pay them off. But the payoff doesn’t cost the airline much, if anything at all, because then having to do so meant that somebody bought a super expensive ticket in their place.
If you assume that your employer is holding 1 week of salary on average, then they’re holding $4,600 for your money at any given time.
Let’s assume an annual interest rate of 3%, which aligns with the IRS rate on overpayments.
So your employer gives you an additional $138 per year because of your money that they’re holding in escrow.
But actually it would be less: they would only need to pay you interest on the amount of your cheque that hits your bank account. The IRS isn’t asking for interest on the amount they withhold for tax.
Why not just ask for a $300/year raise instead? You’ll get more money in the end.
A less experienced engineer might push an impressive amount of code, but a more experienced and senior engineer will get to the correct solution much more quickly, and more often the first time. Over a longer time scale this will become clear as a senior engineer’s API spec stands the test of time (less versioning) or their architecture doesn’t require a rewrite or their data model can be extended without downtime-inducing migrations.
I’m saying this as a self-taught engineer since 8 years old who consistently closed more tickets than my peers. For a long time I wondered why I was less senior or paid less than them, and as I became more senior myself it became clear that seniority was more a factor of how often my instinctual decisions were the correct call, because I had solved that problem before, and less because of my raw delivery.
These rubrics are built to try to help managers make impartial, and objective decisions. This is not only more fair but more structure also helps reduce liability related to discrimination. The downside of this structure means that even though you may have contributed substantially, you don’t meet the specific guidelines for promotion.
So I recommend asking your manager for the career ladder information, if you’re in an organisation where one exists, and ask for feedback as to why they feel you don’t meet the criteria for the next step. That feedback will help you figure out whether they’re just “missing something” or if they see gaps.
In my experience selling to the enterprise things like SOC2 and ISO270001 simplify and reduce friction in the sales process, but they're often not showstoppers.
When you pitch your product to your business sponsor, they're buying based on the value you deliver, a trust in you as a seller, and a conviction that the product will solve their problem. And then once they start to involve IT and procurement and security, then all of a sudden the focus shifts away from value and towards risk and cost management.
By having SOC2, a rock solid pen test report, ISO270001, etc. you can fast-track a ton of these processes. But otherwise, as you note, they'll throw big Excel sheets at you with questions for which many of your answers will likely be "No".
Answering "no" on questionnaires is not a showstopper. Your response on financial questionnaires will most definitely raise business continuity red flags. Your response to the security questionnaire will definitely raise data security red flags. But ultimately these questionnaires are an effort to get a full picture of the risk profile of working with you, not a test where you need to score 100/100.
So in your case, your focus needs to be on the business sponsor. They will be presented the risks from your security and financial reviews, but if you have adequately articulated the value of your product, presented an aura of business stability and growth, operated from a customer-focused mindset, then they will have the conviction to accept the risks presented by the other teams.
And sure, sometimes one of your answers will be a showstopper. So then it's up to you to look into whether you commit to fixing that showstopper as part of the contract. One of my contracts required us to get ISO270001 certification within 2 years - but we were still able to close the contract.
We closed a contact for a business critical system with one of the top 50 companies in the world, who demanded a financial review when we had months of runway.
There's always a path.
Feel free to email me if you have more questions.
My everyday shoe now is from Casca [1]. They’re a Vancouver-based company with a specific focus on quality: I’ve worn them almost every day since fall 2018 (when they launched) and the shoes still look as good as new. And they’re machine washable.
They’re also incredibly comfortable. When you buy a pair of shoes, you have the option to take a few pictures of your feet and the shoes will be delivered with custom 3D-printed insoles. It’s amazing the difference in comfort when a shoe is moulded to your feet.
Within most shipping companies, there continues to be a significant hubris amongst charterers of "I know what I know" - a lot of people who have worked for decades in the industry and still trust their gut instinct over anything else.
As a result, since so few decisions are significantly data-driven, it's actually not that complex to build predictive models that take advantage of arbitrage opportunities in the market.
If you’re trying to load cargo in an area with an undersupply of vessels, you’ll pay a lot for the opportunity. If you’re sending a ship to an area with an oversupply of vessels, you pay more for that because the company renting it to you doesn’t have good confidence in its ability to get a contract once your cargo is offloaded.
Since around 2008 it has been very difficult for ocean carriers of all types (not just containers) to turn a profit and so they all embarked on different strategies to try to become more efficient and manage costs. Before the Alliances, large ocean carriers needed to have an active and full network worldwide to attract customers. It's kind of like the transit problem: you don't get customers until you have a large network, but you can't afford a large network until you have a lot of customers. So many carriers operated routes that were loss leaders hoping to profit significantly off their mainline trades (you see the same in air travel).
The Alliances were one strategy to try to help these businesses become more efficient while still offering customers door-to-door shipping services, so in THE Alliance, there are a number of carriers each with different primary focuses, which means that they can operate much more efficient routes and take cargo wherever it needs to go.
The price of shipping a container is highly commoditized, there's intense competition in the market, and the margins are razor-thin, so I'm fairly confident that price-fixing isn't a significant factor in this industry. As well, as a shipper, the lions share of your costs getting a container from A to B are not the ocean transportation but everything else. Container storage at the port. Transport to the port. Loading and discharging costs. A multitude of fees that easily double or triple the raw container shipment number. Carriers have invested a lot in this space over the past years to verticalise and able to offer more of these services to their customers, instead of just being a pure ocean carrier. That (freight forwarding) is where the real margins are, and customers appreciate it because they get one simple upfront price, vendor management is easy, etc.