tlarkworthy

I can see it's missing some important commands like WATCH, and maybe the WAL for high durability contexts.

I started using VR as a brain refresher, I feel that should be considered too. It's sorta exercise but with more modalities

I have done both and IMHO a Wednesday off where you can totally disconnect is worth more for mental well-being than shaving some time off everyday but otherwise doing 5 days a week.

Yeah the load balancer bit sucks, I forgot about that. I use terraform for multi region deploys but yeah, the load balancer is a major cost.

No cloud Run instance supports 1000 concurrent users and stays active for 15 min after no traffic, so only a small subset of users should ever hit a cold start.

Cloud Run is an alternative to fly.io that scales to zero so it can be much cheaper but with added cold starts.

Many of these competitors are missing: offline persistence, optimistic updates, local cache, push updates, command pipelining and casual consistency. Which is cool but to me this is missing the main point of Firebase which was about making modern, reactive apps simpler to develop.

Appwrite and supabase make CRUD apps easier to write. It's different. I guess Firebase expanded to being used for everything but it was all about realtime apps originally (ex-Firebase developer here)

I like these apps but I find the positioning weird given their feature set

Hi HN, I struggle documenting complex distributed algorithms. Sequence diagrams are great for precision, but you end up with the prose on one page, explaining each step, and the reference diagram elsewhere, which is really hard to read. The readers eyes have to go back and forth long distances.

This notebook is my revelation of using interaction to limit the prose and to highlight the active step on a sequence diagram. This seems to aid in readability for me a lot.

Observable is an IDE. It is code. You can definitely use it to generate GIFs [1] and whatnot too. So it's possible to build what you want, but you need to roll your sleeves up :/

[1] https://observablehq.com/@mbostock/canvas-to-gif

Early days but here is an open source Firebase database server which you will be able to self modify including JavaScript security rules https://observablehq.com/@tomlarkworthy/firebase-server-prot...

yeah it's the technical name for the intuitive sort order you need for a graph of dependencies.

thing is, if you google for it, you will find it as a technique https://docs.oracle.com/database/121/VLDBG/GUID-5A76B6CE-C96... which I can imagine is used to justify the path they were on.

and we must note that the business side did eventually bring in external consultants because the project was clearly distressed. So it did eventually get corrected and I naturally tend to see the worse examples of this kind of thing, but yeah, it is possible to assemble a team of 8 developers and get unlucky such that not a single one knows anything about indexes or cloud databases in general.

You would think! But during my years as a cloud consultant I was parachuted into a problem project. I observed a large well paid team doing weird shit like archiving all their data to keep their database fast.

The database was 10gb RDS with no indexes.

Kinda but observable is more because the notebooks cells reactively recompute. And it's written in javascript and ran in a browser so you have the languages proper development tooling integrated (e.g. chrome Dev tools). And you can easily share it as its the web.

I am a just fan and rapid user. I am so enthusiastic about it I am an (unpaid) ambassador.

https://observablehq.com/ambassadors

I am trying to build a serverless compute layer on top of observable.

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

If I was shilling webcode.run then I would think a disclosure would be appropriate, but as I was just talking about how great Observable is as a literate programming env, I do not think I have anything to disclose.

the topological dependency order of cells is computed, and only (logically) downstream dependencies are recomputed on a code change. See https://observablehq.com/@observablehq/how-observable-runs

I am super excited by https://observablehq.com which has made an out-of-the-box literate programming environment for Javascript zero configuration (like Knuth's it has a non-linear execution order). Since switching to literate programming I have found myself adopting a documentation-driven-development methodology, where I ponder about the purpose of the notebook in the introduction, and because I reread it every time I open the notebook, the clarity of the software goes up, which also feedbacks into the ongoing development process too.

It takes a little while before habits change, but I think I have learnt more in the last year at age 40 than I did in my first year of undergraduate aged 19.

An aspect of Observable which is particularly compelling since Knuth is the fact that inter-notebook dependencies are hyperlinks, so these are not standalone literate programming artifacts, but they are a graph of explanation too. You can learn a lot by surfing notebook dependencies. Bundling code with documentation is such a win.

The other thing about Observable is the cells are reactive, so your documentation is not necessarily static either. You can provide animated interactive explanations too.

More info on the different types of content you can embed

https://observablehq.com/@observablehq/cell-modes

More info on the spreadsheet like execution ordering:

https://observablehq.com/@observablehq/observables-not-javas... More info on the literate programming support

Quite a lot of medium links are on hackernews or on Google searches. I organically encounter it enough I don't need to follow anyone but I do because I also write on it.

You could follow me! https://link.medium.com/w4fPh9dzmpb

But really i don't think I actually consume content by following people . I consume content via social media or aggregators like HN and enough of that contains medium that makes it worth it.

I, on the other hand, am happy to see a non advertising based business model and thus pay for medium and find it very good value for money compared to a newspaper.

Reactive robots inspired by insects (Braitenberg vehicle)

https://en.m.wikipedia.org/wiki/Braitenberg_vehicle

HTTPS://ObservableHQ.com is rows of cells where cells can be code or DOM, input widgets. Cells update themselves automatically like spreadsheets. It's is fairly insane what can done and the information content possible with these interactive notebooks that run normal JavaScript (in a non-linear fashion)

https://observablehq.com/@tomlarkworthy/notebooks2021

Its not unverified. It's trust based on the email domain and hopefully DMARC etc. That's an ok trust model, it proves the request is coming from a government agency. In my opinion the issue is that police forces are not securing their email properly, which is a big issue, but not necessarily the tech companies fault. Domain ownership is the trust model of the internet after all.

Use lambda as the control plane over s3

Google cloud storage (and S3) are the dumb KV stores with great read latencies

It's a problem using US services, coz customer IP which is protected gets leaked to an organization subject to Cloud Act.

Encrypt your network logs and storage with a customer managed key and keep that in your control. It matters less where the data is then.

When you chunk it you have problems about what happens if that process is interrupted. So it's not trivial (though solvable) but it's the kind of atomics you want the new engine to do.

A good resource, some others I consult with

https://cloud.google.com/apis/design

https://opensource.zalando.com/restful-api-guidelines/

I've used https://all-inkl.com and https://hetzner.de both worked, all-inkl I thought had pretty good support.

Skip breakfast

In games the fastest moving object is bullets, so model them as lines, not points.