tlarkworthy

but they do...

https://www.cnbc.com/2021/09/21/silicon-valleys-quest-to-liv...

'Amazon’s Jeff Bezos, Alphabet’s Larry Page, Oracle’s Larry Ellison and Palantir’s Peter Thiel are just a few of the super-rich who have taken a keen interest in the fast-emerging field of longevity`

you should try to talk to existing experts too. Sometimes they can utter a few words that gives you a helpful mind model that saves days of original thought.

flag the other one (https://news.ycombinator.com/item?id=29350983)

the best model of a cat is a cat, preferably the same one

I am building a distributed FaaS using the WEB as the normalising environment. This has several benifites like the developer's browser transiently becoming "prod" unlocking debugging-in-prod tooling, but it also addresses portability. 20ms deploys. it's 100% optimized for DX, no toolchain either

https://observablehq.com/@endpointservices/webcode

I used to work on Google's FaaS and I intend to take those lessons for generation 2.

At Firebase it was sometimes called the client-database-server architecture. The pattern was documented in 2013 [1]

If you use Firebase as an ephemeral message bus it's a great pattern. It has problems if you use it like a traditional database because migrations are very tricky. DBs that support views (or GraphQL) can make migrations much easier

[1] https://firebase.googleblog.com/2013/03/where-does-firebase-...

yeah that's not correct sorry

you have a charitable donation of 10M so you can deduct 10M in tax, at a cost of 4M. Tax rate is not relevant, you are subtracting 10M of your tax liabilities

interledger protocol is for exactly this, and it's not a Blockchain, https://interledger.org/

This is great, I was just working on a forked fluid simulation and this has definatly helped me understand why the pressure calc is found a 25 iteration loop!

https://observablehq.com/@tomlarkworthy/ink

It's missing IAM conditional policies so it's a no for me.

there is peer-to-peer decentralisation and federated decentralisation.

the delta between a graph DB and a relational DB is graph can formulate queries such as X is connected to Y, and stuff that is some kind of walk over the graph. None of that is relevant for CRUD apps.

Releational can model a graph but can only answer queries like X is an immediate neighbour of Y. It lacks the walks (SQL recursive is kinda there but not really).

The primary purpose of a DB is to serve queries. But graph DB's queries are not so useful for most "lookup" type workloads that dominate business. So a graph DB is overkill, and when you get into the nitty gritty, cannot do many performance optimizations as a result. So a graph DB is a worse relational DB. But immediate relations are what is interesting to business.

Think about the the practical value of queries. Queries are the main point of DBs. Now think about most applications. Are graph queries useful for CRUD? .... no. So that is why.

relational is good for lookups and adjacency queries, and thats the main query semantic in CRUD

That ACL filtered list seems like it could be super useful. Its extremely horrid if a UI is full of controls you don't have permission to use.

There is subversive power in choosing what to connect

not sure about event sourcing, but replicating your state immutably on write can save your bacon if there is a bug. I try to skip the D and make the U fork in CRUD for systems where the data is very valuable.

classic event sourcing require playing your events in order to calculate state. I think that's a complexity. Just do normal state computation but write it into a new timestamped row

Funny, retros are the one thing I like. If you can get a team to honestly say what went well and what went bad, and the step that can be taken to improve, every two weeks, with some light follow up. In 6 - 12 months, you can turn a flaming wreck of a project into something that resembles best practice.

I have do it as a consultant. I have taken a project on life support that they were gonna cancel, regardless of our performance, and resurrected it so hard they actually decided to keep it. We changed their minds

The most effective tool was retros. I dislike agile in all other aspects, and do not use story points. Retros are the best bit! It's the core part of the learning loop.

I learnt retros from Google. Big tech does retros. Its honestly magic when done well.

my kids should "dad, look at me" before they do anything. the desire for external validation runs deep. If no-one hears a tree falling, does it make a sound?

Mixed integer programming?

Here is a browser version https://observablehq.com/@tomlarkworthy/mip

I built this https://webcode.run/

I made a service like this on Observable

https://observablehq.com/@tomlarkworthy/minecraft-servers-be

it never really took off so I mothballed it, however, I do use it at home for our personal server and it has saved me a ton of money! It makes perfect sense as you can have quite a good spec machine when you are paying by the hour. you just disconnect the disk from the VM and pay for disk storage which is very cheap.

It was based on the following terraform recipe (which I wrote)

https://github.com/futurice/terraform-examples/blob/master/g...

Things like CD/CI or data processing pipelines are also annoying to develop for similar reasons. Lets ditch the YAML and have an environment we can attach debuggers and properly develop. So webcode is a building block to easier to develop future cloud services. Thanks to Observablehq its also very easy to document, fork, and code share. So this is just one aspect of a improved development workflow.

The main contribution is that this includes a symbolic rearranger so you don't need get your problem into a certain form (unlike all other MILP solvers in the JS ecosystem)

OMG the pointless telemetry. I got asked to install some on a project I took over. I checked, and said "there is already telemetry integrated for account XYZ". Nobody knew who had access to XYZ. Like, they install it, and lose it. Its not even fulfilling a useful purpose.

because it's a brain booster. Too much boosting leads to problems. I really expect there to be a line underneath which sugar consumption is net positive

I think sugar demonization is gonna be yet another fail by nutritionists. In 10 years time they will say, oh actually it turns out that sugar in moderation stimulates the brain directly and increases focus. Its especially useful to animals that rely on cognitive function, like humans.

It will be one of those facts we will all immediately recognise as obvious in hindsight and wonder why we went along with the fad of cutting out sugar, which had a detrimental effect on our ability to think.

Meanwhile, nutritionists will have moved on to saying calcium isotopes are shortening our lifespan.

My theory is its data driven decision making. It's messed up the corporation to be purely metrics driven. Everyone hunts the metrics and misses the qualitative reasons for anything. We distrust internal models now, and try to be purely empirical, which is a mistake. We are overly technology now, but the technology lacks nuance and long term reasoning.

no, it's no good for big data sets because lambda warmup time and RAM consumption is proportional to binary size.

Just reference the data as a URL to static storage, problem solved.

no the blink detector is not free and massively adds to the overheads compared to the LED output.