SocketCluster VS ideas2

Recently, I added an AI-generated soundtrack to my open source project's home page https://socketcluster.io/

It seems unconventional at first but I distinctly remember about a decade ago when Adobe Flash was still broadly supported, many Flash websites had soundtracks. I think the reason why regular HTML websites didn't have them was because it was difficult to implement and internet was much slower so they had to be streamed in a special way.

If redis doesn't satisfy your requirements or you're unable to make it work using adaptor, SocketCluster is a great package for this https://socketcluster.io/

Interesting how this feature set is pretty much exactly the same as offered by SocketCluster https://socketcluster.io/

This resonates with me 100%. Every bit of unhappiness I've felt in my career so far has been caused by a bad manager. The main issue for me has been the final point you mentioned about "Imposed artificial limitations" - I cannot tell you the number of times that I've been forced to use an inefficient tool or do something in a sub-optimal or downright incorrect way (knowing that it would have to be re-written later) by a bad manager... In some companies, it was a daily occurrence; that's why I never stayed at a single company for longer than 2 years. It's almost impossible to find a company that lets me implement things correctly.

Thankfully, nobody could constrain me in my open source work. I (with the help of community members) built:

- SocketCluster (https://socketcluster.io/): A distributed pub/sub framework.

- Capitalisk (https://capitalisk.com/): A lightweight quantum-resistant blockchain which is less than 5K lines of code.

- LDEX (https://ldex.trading/): A deterministic decentralized exchange (DEX) which can work with many different blockchain protocols. It's less than 4K lines of code in total and only has 3 small third-party dependencies (including sub-dependencies).

The two best lightweight solutions I have worked with are SocketCluster and NATS. SocketCluster is used to build out a custom backend that scales very well. NATS would be used as a central pub/sub system. If you need queueing, you'll need to add additional services like Redis.

In the homepage for SocketCluster it is mentioned that it is a Highly scalable pub/sub and RPC framework optimized for async/await. However in the documentation I have not found anything related to RPC. Am I interpreting anything wrong here?

At the time of writing, SocketCluster has almost 6k stars on GitHub and 7k downloads on npm weekly,

SocketCluster is a framework which allows you to use the WebSocket protocol the transmit between its backend API and client library. The client can be used both on a backend (E.g. a Raspberry pi) or frontend application.

My blogging/journalling setup is simple.

I just use GitHub. I just rely on the default repository view on GitHub.com

I create a README.md and add markdown headings to the bottom or to the top (bottom if its a journal, top if it's a blog) and then when I get to 100-800 I create a new repository and repeat.

https://github.com/samsquire/ideas (2013)

https://github.com/samsquire/ideas4

https://github.com/samsquire/ideas3

https://github.com/samsquire/ideas2

Thanks for posting this Ask HN question.

I journal ideas and thoughts about computers and software. I am interested in software architecture, parallelism, async, coroutines, database internals, programming language implementation, software design and the web.

https://github.com/samsquire/ideas (2013)

https://github.com/samsquire/ideas2

https://github.com/samsquire/ideas3

https://github.com/samsquire/ideas4 <-- this is recent but needs editing

https://github.com/samsquire/ideas5 <-- this is what I'm working on now

https://github.com/samsquire/startups

https://github.com/samsquire/blog <-- thoughts I want to write about, but incomplete

I use README.md on GitHub and create a heading at the bottom for each entry. I use Typora on Windows or the GitHub web interface to edit.

[3]: https://github.com/samsquire/ideas2#5-open-demand-mapping-an...

Thanks for this.

I love the idea of breaking up a flow into separately scheduled but still linear message flow.

I wrote about a similar idea in ideas2

https://github.com/samsquire/ideas2#84-communication-code-sl...

The idea is that I enrich my code with comments and a transpiler schedules different parts of the code to different machines and inserts communication between blocks.

I read about how Zookeeper algorithm for transactionality and robustness to messages being dropped, which is interesting reading.

https://zookeeper.apache.org/doc/r3.4.13/zookeeperInternals....

How does Mats compare?

LMAX disruptor has a pattern where you split up each side of an IO request into two events, to avoid blocking in an handler. So you would always insert a new event to handle an IO response.

I am more likely to journal and blog if the friction to creating a post is as simple as opening a document and writing. The important part of journalling or note software is that you actually create notes. I did use Hetzner to run a Wordpress blog but it had an overhead of server expenses and keeping Wordpress up-to-date.

I don't want my data trapped in a proprietary system where it is difficult to export, so I use plaintext. I looked into Publii [1] but I prefer my current plaintext setup. Today I journal software ideas, computer ideas, startup ideas and community ideas on GitHub in the open, as README.md files. My journal is all public on GitHub at the following links. There are over 550+ journal entries, I am sure you shall enjoy them.

https://github.com/samsquire/ideas

https://github.com/samsquire/ideas2

https://github.com/samsquire/ideas3

https://github.com/samsquire/ideas4

https://github.com/samsquire/startups

https://getpublii.com/