abstrct

I like playing with an agent as a team in a game. We discuss strategy, divide up tasks, review results. It’s helpful in a an always-on game to have an autopilot mode so I can go about my day.

I’m impressed the article managed to not mention Nathan Fielder even once.

Yeah, especially with a large knowledge base I find it important to keep a log of prompts/responses and perform team reviews of both. It’s honestly making more work than it’s saving at the moment with the hope that it’ll be more helpful down the road. On the plus side it’s made the team more interested in tasks around technical documentation and marketing material, so still a win!

I’ve absolutely explored this idea but, similar to lossy compression, sometimes important nuance is lost in the process. There is both an art and science to recalling the gently compacted information and being able to recognize when it needs to be repeated back.

The most limiting factor I’ve come across is hitting the context window. Eventually your new eager employee starts to forget what you’ve taught them but they’re too confident to admit it.

Is this insightful? I feel like this is trying to help me be a better founder but I don’t know what to do with this advice.

Yes, I can be successful personally without social media. Can I successfully launch and market a product without it though?

I will always miss those I’ve lost to benzos. Some of my favourite people that I’ll never again share a new memory with.

Heh, funny you say that. Schemaverse was built for my masters thesis which explored if the application layer and data layer could be successfully merged in a way that improved data consistency and integrity, and thus arguably improved security. I'll save you the 90 page read but short answer yes, but scaling becomes an absolute nightmare.

You're welcome! Feel free to drop by #Schemaverse on Libera.Chat if you have any questions.

One of the players dockerized it all too https://github.com/frozenfoxx/docker-schemaverse

I highly recommend running your own instance for battles between classmates or employees.

That's awesome! I'm glad to hear it helped. I was personally so sick of every database course I took using the same `department` and `employee` tables, I wanted something fun instead.

Those should be fixed now. Thanks!

Yeah, he's pretty unstoppable lately but there have been a couple new players starting to improve a lot too.

I'll happily share his secret. He recognized that it's not about any single metric, it's all about optimizing for the trophies https://github.com/Abstrct/Schemaverse/tree/master/trophies

Ah, it does look like GitHub changed their wiki url scheme. Thanks for the info!

That should link directly to the tutorial (link above) instead of that wiki anyways

Thanks! Totally forgot to update that. You can find us in #schemaverse on Libera.Chat

The very tiny server is getting a bit crushed at the moment. If you actually want to try it out, check the tutorial tomorrow when the traffic goes back to normal.

https://schemaverse.com/tutorial/tutorial.php

Heh, ok did not expect to see my old project sitting on the front page <3 The server is getting gently hugged to death atm but I'll try to keep it responding.

Some added info...

- The entire thing is open source: https://github.com/Abstrct/Schemaverse/

- This was discussed on hacker news 10(!) years ago https://news.ycombinator.com/item?id=3969108

- Some players have written more code to play the game than I wrote to build it

- We have a tournament every year at DEF CON

- Postgres is awesome

This is a pretty smart purchase, with it MasterCard is getting a fairly mature technology stack (by cryptocurrency standards at least), an incredible amount of data (mostly attribution data relating to addresses and transactions), and a team that’s experienced with the industry (which there is a massive void in skilled workers for).

I helped with this! :D

Happy to answer any questions people have. With the industry as crazy as it is, hiring is becoming difficult again. We hope this will help businesses find good candidates that understand the technology and, more importantly, know how to use it safely.

Bitcoin users are exceptionally gullible? I'm going to need a source on that one.

I'll preface the following by saying I fundamentally disagreed with rolling back the DAO as the entire purpose of these networks is immutability and their trustless nature.

The reasons for it, even on their own, all make a lot of sense:

- Ethereum has always wanted to move to Proof of Stake, a move that would have been impossible if 25% of the networks funds were suddenly in the hands of a nefarious actor.

- The DAO was intended to seed development into projects that would benefit ethereum and the blockchain space as a whole. It was an incredible amount of capital that was all intended to help evolve the network. Losing all this capital and momentum would have been detrimental to that goal.

- The DAO was arguably an unlicensed security and losing the funds of that many investors would have been a problem for many who had a hand in building it. By returning investments, they basically made this a non-issue.

So again, do I agree with the choice? Absolutely not, you should not proclaim "code is law!" until it's inconvenient for you. Did they have a choice? No, the roll-back was inevitable as soon as the DAO contract was deployed.

Over-performing via Bootcamp. The machine is a fairly new Mac Mini.

It runs horribly on macOS. Like a night and day difference. If I understand correctly the difference is primarily regarding the encoder.

I also switched my streaming machine over to windows recently.

Edit: ok, “horrible” is dramatic, but it was honestly shocking the first time I booted into windows and gave OBS a try there. Exact same computer that struggled with OBS in macOS produced a much better output without breaking a sweat.

Love the method used for I/O, super clever.

My long running pet project is a multi-player space battle game played entirely via SQL inside postgres https://schemaverse.com/

This scares me a bit. A part of email validation is ensuring they actually own that account. It depends on the sensitivity of the service to an extent but I don't think it's appropriate to let somebody interact as if they were a specific email account until they've proven it's theirs.

Edit: I do like the recommendation of showing them the email they're waiting to validate so that they can see typos.

You bring up a lot of important points and I do agree with you on many. I’d love to hear your perspective when sensitive data is involved, what takes priority? Is it the privacy of the employee, or the customer? Can there be a balance that respects both?

Without their knowledge? The use of these tools should be outlined during employee onboarding or explained during implementation/roll out. There is no good reason for them to be a secret.

That said, my statement was about helping people protect themselves. These systems will be used, and for legitimate reasons in many cases. Why not protect yourself from allowing them to overreach into your personal life?

I think I may just have a tainted perspective on this. I’ve worked in digital forensics/incident response for >10 years, so I have an appreciating for the level in which businesses need to protect themselves, and I’ve grown a fondness in never seeing my personal data end up under a litigation hold. It’s the same reason why I’m a strong believer in making sure to have different email accounts for work, personal, different side projects, etc.

Compartmentalizing is a valuable Op-Sec practice.

Then to me, this is the real issue. Businesses aren’t providing the right tools to help employees separate their personal lives with the business.

BYOD has indeed been a curse.

Don’t use the company device for personal use. It’s just not appropriate, for so many reasons.