jonromero

It should work now!

I got the code, made it run in JS (still uses OpenRouter but I am working on including JS LLM) and you can even download the html and have a go at it.

OpenSource as always.

We try doing that with HeyFire.co - import from a screenshot that is processed on your browser! But with a high rate of hit or miss right now.

Thanks! I appreciate it!

I find it hilarious that 99.9999% of web3 events are not using NFTs as Tickets. Even more hilarious is that are using Ticketmaster or Eventbrite.

I decided to participate in a Polygon's hackathon and build something that is totally free, with no gas fees and more importantly doesn't alienate non-crypto folks.

The platform is called FutureTickets.xyz and you can claim a ticket here -> https://futuretickets.xyz/e/your-first-nft-ticket_91nox70YHN to experience the flow (you don't even need a wallet!).

If you want to post a new event, I would be thrilled as I am looking for more feedback, just send me a DM to enable your account.

A walkthrough of the platform is here -> https://www.youtube.com/watch?v=21IE5Tmh-nM

I used: Python (Flask), AlpineJS (should have used Vue or React), TailwindCSS, Solidity and hardhat.

The database is built using Erlang and C. Read more about it here: http://highscalability.com/blog/2012/11/26/bigdata-using-erl...

Processing data (stacktraces) from ~500M devices around the world in realtime is BigData.

Mnesia is by far the fastest (and easiest to use) thing we've ever tried. The only problem was that it was more expensive to keep data in memory and that pointer arithmetic is so damn fast. And (the most important thing) is the 2GB limit Dets has :/

Having a small (but very strong) R&D team helps you eliminate most of these "processes". The buzzword here is "iterations". Lot's of them.

Our clients were involved since day 0 (which I think is the key of our success btw). The first prototype (built in two days) was in Erlang using Mnesia.

The drawbacks are: + not having lots of engineers knowing C and Erlang + Outsiders having problem with LISP (that's why we implemented PIG)

Also, creating a HIVE/SQLish language on top of our LISP was super easy.

We are still using it for some other parts.

Yeap, the problem was the VM. And we are going to opensource most of it.

We tried to find something that was fast, cheap to scale (in terms of servers) and easy to extend but we couldn't find any other solutions. The problem is having everything in memory and doing correlations really fast.

Two people, prototype in 2 months (CTO of BugSense here)

I agree. Nobody said that having a small business that is making enough money to live a happy life is a bad thing. But lots of us are driven by the ambition to build something bigger. That's the driving force. Not some VC.

Yeap, I think that the two of them were the "dream team".

You just install the BugSense library. Check the docs (it's easy!) http://www.bugsense.com/docs/ios.

If you make an account - it's free -http://www.bugsense.com/signup/ you can see the UI. It's pretty cool!

thx, I'll update the site and I'll include a sample of the newsletter.

Good feedback! Fixing right now :D

Sounds crazy!

Yeap! there is already a :duration and a :volume element. You can map transformations on them

Thx for the post! Feel free to ask anything you like!

nice thought! I'll add it in the next version! thanks