I don't get what "Add" does

Try it and see! But, it adds the recommended book to the list of selected books.

>What nothing we've found out there do (except perhaps kedro) is model the "micro" -- E.G. the specific fine-grained dependencies so you can take a look at your code and figure out how exactly it works.

Just wondering then @elijahbenizzy - how does Hamilton differ from Kedro?

Ha! Kedro maintainer just joined on the hamilton slack, and we were talking it over :) here’s what I was thinking:

TL;DR hamilton is lighter weight and less opinionated about non-pipeline stuff, it also has a different way of specifying pipelines (that we prefer).

I think there are a few key differences in the approach: - currently, hamilton is lighter weight and less opinionated about directory structure/style guide. It’s just a library! - Kedro pipelining (from what I understand) has you define the nodes separately to specifying their inputs, whereas in hamilton it’s function-first, and the functions specify everything. It’s funny — Kedro is actually very similar to the framework I first designed to solve the problem and compared with @Stefan Krawczyk’s idea (that became hamilton), I called it “burr” - Kedro has a whole bunch of additional features that allow it to integrate with the outside world and hamilton is pretty lighter weight here (although we’ll likely be adding more)

Ah, cool, thanks for the clarification. Good luck with it all, and congrats on the launch!

Thank you!

>Pretty cool when you're added to a new project, have no idea what people have been working on for the past 2 months there and want to see where most of the actual effort was focused.

That's a neat idea!

That's a good point, but I guess for non-tech people activity => work and no activity => no work, so at that level it illustrates reasonably well, no?

This is just Goodhart's Law in action.

> activity => work and no activity => no work

Measuring activity, where by "activity" we mean "commits", means that a person who mentors other people, reviews thousands of lines of other people's code, and discusses that code in depth during long hours of meetings, does completely no work. I think you can see where such a measure can lead to.

Perhaps, just maybe, you might consider the fact that non tech people can understand things

Apple Music in a car when you are in an area with no network signal. Urgh! Try getting to downloaded music while you are driving (don’t do this)!

Go to Playlists, choose ‚downloaded Music‘ (don’t know the exact description in English, I’m German): You get a list with all downloaded music an can play it. If you’re in a area with just EDGE Apple Music is skipping all titles which are not on your device and is playing just downloaded music. At least as long you play a random list.

Disclaimer: both the author of the article and the OP for this post (me) work for Ably.

The post compares two cloud pubsub services: Azure Web PubSub and Ably in terms of ease of development, creating a collaborative multi-user pixelart drawing app with both services.

Conclusions: Both Azure Web PubSub and Ably offer a similar pubsub feature set, but the Ably API was easier to use (although comes with the disclaimer that the author has used Ably more in the past).

Being able to publish named events and subscribe to them (client-side) without the need of handling them via an upstream server and custom event handlers was an advantage, while a potential downside is the use of a third party library increases the download size of the app.

I've worked from home for years and agree with much of this. It isn't just the "office at home" it's a totally different mindset. It's probably only in my most recent job that I've really _got_ that, and that is because my company is fully remote so everyone is doing it. Nobody feels left out by being at home, and nobody expects synchronous communication or presenteeism.

Thanks for posting this. Full disclosure: I'm the author. I also work for Ably, which is what the post is about, and I'm a content writer so there is inescapably some marketing in it.

But I hope it's valid because, if you try to build a multi-region system for low-latency data transfer across regions, there are issues like discovery and synchronisation to consider.

I think it may be interesting to see how Ably is architected so its pub/sub messaging is realtime and reliable.

For example, how do the EC2 instances in one region find those in another? How does a published message in one region get to a client subscriber in another?

>>"After 50 one begins to recognise in oneself a growing resistance to change, if only because it so often nullifies years of hard-won training"

This really struck home to me. It's true: we teach ourselves life hacks and gradually they become redundant as different life skills are needed.

But also I'd like to think we get better at distinguishing between good changes and bad changes. We should resist the bad changes, or at least explain ad nauseam to the kids that their proposed change a) has been tried before and was crap, or b) is the beginning of the end, or c) is against the natural order of things. ;)

Yes, you are perfectly right, but sometimes it is overwhelming/tiring, the bad changes are becoming so common (and supported/accepted by so many people) that "fighting" against them appears to be (and very probably it actually is anyway) completely void of any relevance.

Over time I've gotten better at letting others fail. I failed a lot during my career, especially when I was young, and it seems to have worked out for me.

Short answer is infrastructure for event-driven architecture that is close to the edge so it can distribute events with low latency.

Longer answer is to take a look at some of the Ably blog posts linked in the post here, since they'll explain why Gartner have named this as a category for event-driven brokers.