hectormalot

A great blogpost on the architecture Stripe chose for calendar based versioning: https://stripe.com/blog/api-versioning

In essence, they transform requests from the latest API version back to the (older) versions by writing an adapter for each response for every breaking change. Not sure what they do for new endpoints, or when a future response no longer includes a field that used to be included before.

Some nice things though:

- timeline they suggest seems to be longer than GitHub (implied by the “a power company should not change its voltage every 2 years” statement)

- instead of a header, it’s linked to the account of the user and the time of the first API call. That might be a good default for GH rather than the oldest available version at that time as suggested elsewhere in the comments. (drives adoption, least surprise, longest expected validity of the interface)

Edit: formatting

No I think it is the sales that counts. That 200k salesperson will of course make PowerPoint decks showing all the support options. But, the main barrier to take is convincing 10+ people with different incentives to all sign off on your product, and that is what enterprise sales gets done.

This one, great reading too: https://www.joelonsoftware.com/2004/12/15/camels-and-rubber-...

It’s a long piece. The “credit card“ vs “3 months of hard core PowerPoint by sales” pricing is about 75% into the post.

I’ve used it for some designs and found it quite easy to work with. UI is very good (vs typical open source)

Challenging is the Figma ecosystem taking most mindshare. Eg trying to find someone on Fiverr to create PenPot designs is hard.

I'd pick PostgreSQL over CouchDB because it is so common, has great documentation, and many people say it is a great default choice.

For quickly launching an API enabled backend, maybe look at encore.dev? It's not one of your listed languages (it's go), but at least it looks somewhat like C.

I recently had to launch an API for a hackathon, and this was as fast as Ruby on Rails for me. Comes out of the box with Postgres and PubSub (2 lines of code if you need it). Encore also does the infra provisioning and hosting for you (either your cloud, or for free at their cloud). Tip: add `sqlc` for the queries, it works well together with Encore.

GraphQL/REST: it looks like your app will have a limited set of calls that are easy to predefine. In that case, REST wins hands down. IMO GraphQL really shines if you have (a) a data model with many relations, (b) possibly from different underlying systems/services, and/or (c) different people working on the front-end/consumption side vs the data exposure side. But that's beyond 'a simple app'.

Other tip: if you're thinking of building an admin interface. Don't, just use something like Retool first while you're getting users.

For some APIs you can get the access yourself. For example, here is the API for account and transaction information at ABN AMRO (3rd largest bank in the Netherlands): https://developer.abnamro.com/api-products/business-account-...

I'll concede that it's only for business accounts. But, that includes small single owner businesses and I've been considering to switch to ABN to automate the accounting of my small side business.

The PSD2 APIs (also for consumers) indeed need a PSD2 license and an EIDAS certificate.

Technically, for consumer data, I think you could use GDPR to request a computer-readable extract of your transaction data. However, I think most banks would then redirect you to some CSV/MT940 export option in the web interface that is hard to automate.

pros and cons, will depend on the type of thing a team/company is working on. I've been on the receiving end of this with some negative experiences as well.

Team1: "Hey, we've launched a new initiative a few weeks back. Important for us to be able to add one field to the response of your internal billing API. Can we discuss what that would take?"

Team2: "Hey, probably a few hours of work. But, we just had our quarterly planning last week. In 12 weeks you can submit for Q2, and _maybe_ we will plan it somewhere in May or June."

here you go:

* https://centreforaviation.com/news/pilot-union-accepts-emplo...

* https://www.flightglobal.com/strategy/klm-clears-path-to-bai...

VNV is the pilots union of KLM, the Dutch national airline. IIRC they were the very last of the unions to sign up to the 'commitment clause' the Dutch gov required as part of their bail-out during Covid. Not a union without teeth.

Wanted to share Penpot as well. It worked surprisingly well for my use case.

Ecosystem is limited unfortunately, eg hard to find someone on fiverr to deliver penpot files instead of figma

Fun fact: He left to join another company some time later, and I now work with him again.

I’ve had 2 exit interviews and both were good experiences.

The first scheduled the exit interview 5 months after leaving my job. Clearly said that they wanted to make sure I was comfortable in my new job before asking for feedback. I think the gap of a few months is a good mechanism to get honest feedback if someone is willing to provide it (and is what I now do for exit interviews myself)

Second time was with a manager I really trusted (I was leaving for other reasons). Gave him a list of the 15 things that I thought needed changing. When I visited almost a year later he showed me the list still on his office wall, with 12/15 items crossed off.

I think game theory might optimize for your own immediate gain, but overall many managers and companies want to do good, and your feedback might actually help if delivered the right way.

> Simply because the credit system is _in my opinion_ imperfect and punishes people harshly. It encourages borrowing. i.e. if a person is responsible and chooses not to open a credit line, then their credit is considered bad, and banks would consider them risky. That's just one example.

That's a US (and maybe some other countries as well) perspective, but not the same for all countries. e.g. in The Netherlands, your credit report is mostly concerned about how much debt you have (vs capacity to pay) and if you have debts in arrears. I understood historic debt payments don't really factor into it here.

Note that different debts do have different weights. e.g. a car loan for 20k can reduce the limit of your max mortgage by 50k because its a bigger impact on your capacity to pay.

I’ve once had an amazing view on the northern lights flying icelandair. Crew went around waking people up so they could see it.

If it's a true concern, the answer is probably iodine tablets and a respirator. Increases your chances of survival (if you would survive the first wave).

But, what more can we do? It's an interesting question to think about. Nuclear war would probably be the result of a significant escalation, and in all likelihood will involve many nukes back-and-forth, rather than a single nuke hitting a single city. With that in mind, I'll give it a go. I'm assuming your goal is to survive the war, and be able to continue living afterwards.

Step 1: A location likely to survive the war. You need to pick a location that allows you to survive the initial wave of nukes, then provides sufficient shelter for the resulting radioactive fall-out, and then finally allows to you rebuild some resemblance of living, once you're able to go outside again. This probably eliminates all countries that have nukes, or are part of alliances that have nukes. You'll either get a direct hit, or will likely be close to the fall-out.

Maybe French Polynesia is a good place. It is in the southern hemisphere (vs US, Russia, EU in the north), very remote, has limited military value.

Step 2: Shelter. Although you've picked a location unlikely to get nuked, it's still good to be in a sturdy shelter, underground (maybe even under water), that also holds your supplies. Assuming no direct hits, I'd go for a concrete basement with an extra thick roof.

Step 3: Stocking up to survive the aftermath of the war. You'll need various supplies: (a) water (for days, assuming you have a way to get new fresh water, e.g. a well or small-scale desalination), (b) food (months, maybe years, assuming global supply chains no longer exist), (c) basic defensive weapons (when others find out you have supplies), (d) materials to survive the increased radiation outside (iodine tablets, respirators), (e) medicine (antibiotics, painkillers, etc.)

Step 4: Building a new life: seeds, fertiliser, building materials that are hard to get in the post-war world, fishing equipment, a boat, etc

Bonus items: Communication (FM/AM antenna on a weather balloon?), Transport (Sailing yachts with solar panels and desalination equipment can go a long way, might even be a good shelter if you find the right place to sail), pooling up with a group of people so that you can maintain a small 'society'.

Sink directly into it for the summer. Heat exchanger between the source (approx 10-15C) and the water from the floor heating/cooling system. Because the heat pump is not involved it's very efficient (only the power of the well pump)

I'm actually considering if it makes sense to push a lot more heat into it in the summer (Deliberately leaving curtains open, then moving this heat into the well; or even PVT panels), but I'm not sure if the well would retain heat long enough to make a difference towards the winter.

Correct, it's big but I wouldn't say huge (maybe 4.5x2.5m). This is the machine they used: https://conrad-stanen.nl/nl/producten/boxer-200

There is a great (German) database with heat pump performance. You can switch to English on the right top: https://www.waermepumpen-verbrauchsdatenbank.de/index.php?bu...

There's quite some monthly data on Air/Water systems (extracting heat from outside air, into a central heating system). I suppose good Air/Air systems could be similar. Indeed above >0, or at least >5C it would be beneficial (even more so if you also have PV installed)

your comment about average yearly temperatures is of course correct, but in this case it biases in favour of GP:

* It's colder than average today, to reducing the setpoint has more effect vs an annual average. If sun takes part of the heating, then the ratio of dT to gas saved is even bigger.

* The formula actually does work that way, maybe you're thinking of a different formula? i.e. Q (flow of heat) is a function of some heat transfer coefficient (U), area (A) and the temperature difference (dT) between inside and outside. So the (simple) correlation between dT and gas consumption to maintain a temperature is linear. This will be influenced by e.g. wind (increases U, therefore gas consumption) and other factors of course.

Anyway, it's winter. If there is any time where you can reasonably reduce your gas consumption by lowering the thermostat, it's today. In fact, I'll try this now and see if it works. Maybe time to start a movement.

(Also note that residential heating and office account for roughly a third of gas consumption in e.g. the Netherlands. Reducing that by 20% is still 7% reduction in total, and potentially a 17.5% reduction in Russian gas given the current gas mix)

I have one of those. The well is 170m deep. Took 2 guys about half a day to drill it, so that actually doesn’t seem excessive in terms of time required (mostly sand, no rock below my house).

We moved from IBM Connections and WebEx to Sharepoint and Teams. I still think Teams is terrible software[1] but I’ve learned there’s definitely worse software out there.

[1]: the 100% CPU, the fan noise in the background of many peoples calls, the 500-2500ms lag between clicks and a screen update, the UX disaster of teams vs chats vs private channels, etc.

Indeed, most pyramids are less than 5k years old. A 100k years means making the step from the first Egyptian dynasty to today 20 times. That’s an incredible timespan. I don’t think we can design rules and agreements around this type of storage and expect them to last that long. Statistically it’s unlikely the country you are living in still exists in just 10k years, let alone in a 100k years.

Maybe not that far apart. The challenge with subsidies is the administration and making sure you exhaustively subsidize different options. It’s also infeasible for smaller investments such as radiator foil.

My underlying assumption is that you would reduce the 2k compensation in my example. So the incentive for any saving remains.

On duration: I installed solar for 0.80€ (ex vat) per Wp (which generates about 1kWh p.a.). In NL electricity rates recently spiked to 0.45 euro per kWh. Based on that, payback time is now less than 2 years (that includes everything, panels, inverter and installation, etc.)

but I think we agree the challenge is to find a way to create a good market incentive for consumers and industry to reduce their externalities/CO2.

No it doesn’t. Example, imagine you currently have a energy bill of $1000p.a. With the proposal above, we’ll move you to a $3000p.a. bill with CO2 taxes. And, we’ll give you $2000p.a. to compensate for the increased energy bill.

Would you really just keep paying more for energy because you have that? The payback period on solar panels just went *0.3, batteries just became a affordable option for day/night balancing, many alternative investments in renewable energy just became profitable.

Many people will use that opportunity to get ahead financially. I think that’s what GP meant with “the free market will sort out the rest”

Yes, but is it a problem worth solving beyond what we have today? What’s the last time we saw an database integrity failure at the DMV/… leading to cars being owned by people that shouldn’t?

True. I’m hiring lots of data scientists and have become somewhat disappointed with the phd profiles. It happens too often that they lack the broader skills to be successful, vs people who did do “non-core” activities and are now more well rounded.

> I don’t think you’ll be well-served carrying this mental bias around.

I didn't mean to pick sides, or suggest anyone picks graviton by default. I meant to comment on how the communication around these new architectures by AWS has shifted, and what - I think - it signals about Intel's current and future market position with AWS.

For reference, my stuff runs on Heroku, I'm clearly not that concerned about price/performance ;)

The specific designation of Intel with the ‘i’ suffix is interesting. I’m surprised how fast this went from “everything is Intel” to “only take intel if you need something like SAP”.

I remember the first AMD instances a few years back and the 10% price/performance was nice, but at the same time didn’t feel like it would convince enterprise users to switch. Now with Graviton2&3, they’re apparently at 30-40% advantage and any big user would be crazy not to explore moving workloads over.

I expect the Intel-AWS negotiations today are _very_ different vs 5 years ago. This new naming is just one of many signals of a further downhill trend to Intel, and it’s likely to gain momentum in the next few years as the experience and ecosystem around ARM grows.

If you read the description you’ll see that it’s optimizing around certain syscalls like memcopy (which is otherwise limiting and reduces speed by 5x). I think it’s less language choice, and more low level optimization that is making a difference here.

(To check, I wrote a Go program that doen nothing else than fmt.Print($largebody) in a for loop and it tops out at around 10GB/s)

If we want to put it this way, yes.

140TWh out of 170000TWh is about 0.8%. relative impact is bigger than 0.08%. Coal is 36% of the power mix. Under the assumption that most bitcoin mining happens connected to the grid (seems reasonable) and that these grids contain coal powerplants (also reasonable with the exception of Iceland), that 0.08% translates to 0.22% of global coal based emissions. On a global scale, being responsible for 1/500th is a lot (for me at least)

the numbers are more relatable when looking at absolute impact. Assuming a average coal power plant of 600MW (you can find bigger, but this is a representative plant near me[1]). That’s 5TWh/yr per plant. Bitcoin then represents 28 of such coal powerplants running at 100% capacity 24/7. That - to me - seems like a massive amount of CO2 for a niche application.

And if I read your source correctly, that’s just bitcoin

1: https://en.wikipedia.org/wiki/Amercentrale

Coal is about 36%

Sorry, but 0.1% of global energy consumption still adds up to a lot of coal power plants.