Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ask HN: Who wants to collaborate? (April 2022)

news.ycombinator.com

Post reply on HN
Print view View original

101–110 of 125 posts

Re: Ask HN: Who wants to collaborate? (April 2022)

#101

Post by skywal_l »

illegalbyte wrote: post #96
skywal_l wrote: post #83

I'd like to make a website that provides movies and series recommendations based on your preferences. Like letterboxd but extended to series (and other things?) and with open data so that anybody can improve on the recommendation system. Got this idea browsing rotten tomatoes and realizing you can't order lists based on users scores. And the API is private. So, putting your score there is basically donating it to a p…

Heads up, themoviedb has a very robust API which is free and used by big services such as JustWatch and some streaming services. https://www.themoviedb.org/

The data still belongs to TiVo corporation. It's not a Creative Common license like Wikipedia. The API is free (as in free beer) but that might change at anytime in the future.

Re: Ask HN: Who wants to collaborate? (April 2022)

#102

Post by marginalia_nu »

If anyone is looking to dabble in the search space, I'm considering opening up Marginalia a bit. It's written in Java but it's also based on REST services so other languages could be integrated, no big deal as it's not within some particularly hot code loop.

The entire project is basically an excuse to dick around with NLP, low level programming, weird algorithms, large datasets; you know, the sort of programming that's actually fun. I also have a metric shit-ton of real world data and metadata to play with.

I'm half looking to open source it partially or entirely (probably AGPL), but I'm working on ironing out some details about the format. Not really sure what model works since it requires niche hardware to run, manual maintenance to operate, it's a nightmare to test, and so on.

Re: Ask HN: Who wants to collaborate? (April 2022)

#103

Post by dcz_self »

HeyLaughingBoy wrote: post #87
dcz_self wrote: post #52

Earlier quoted context omitted.

A bike computer is the name for a device you put on your bike to show you useful info like speed etc. Typically they are about as flexible as calculators, but this CPU has been shown to run Doom, so it's a general-purpose computer too :) https://github.com/next-hack/nRF52840Doom

Doom on the 52840 is impressive! I don't know if this is something I'd want to build, but I spent 5 years at an outfit that designed/built a variety of fitness devices like this, including some of the consoles on gym equipment you may have used. You mentioned 'figuring out Bluetooth (BLE, please)" in another post. What are you having trouble with? The sample code in the Nordic SDK is pretty good in most cases.

The Nordic SDK is out of the question. It's not Free Software, and there's already plenty of non-free-software devices out there.

I will have to port the NimBLE stack to Tock OS, and that's guaranteed going to be complicated, so someone who knows at least how Bluetooth works in general would be invaluable.

Re: Ask HN: Who wants to collaborate? (April 2022)

#104

Post by thirdtrigger »

tartakovsky wrote: post #72
imranq wrote: post #65

Earlier quoted context omitted.

Vector search is getting a lot of attention these days. Why is that?

Read about pinecone.io. Cut off a neural network that does image classification at the second to last layer and you have a system for creating meaningful vectorial representations. Then if you apply some distance metric to a bunch of these it turns out you can retrieve images that look similar to some query image. Problem is, scaling and doing this in real-time is hard. Use cases abound. This was only one. There are…

Agreed! You can check out Weaviate too. It includes out of the box vectorization modules

Re: Ask HN: Who wants to collaborate? (April 2022)

#105

Post by chaosprint »

I am working on Glicol, the next generation computer music language that runs in browsers and many other platforms. The goal is to help people with zero coding and music production background to get started with music programming. Yet it is also quite innovative and revolutionary as few languages can have these features together:

- run in browsers with a near native speed

- rethink music interaction in live coding from the audio library design perspective

- come with a new collaboration mechanism with decentralized demo

- run on different DAWs as a VST plugin

- run on development boards such as Bela

- support both high-level music sequencing and low-level sample-level audio synthesis

You can try it here:

https://glicol.org/

It would be great that if you can give me some feedback here or on GitHub:

https://github.com/chaosprint/glicol

Re: Ask HN: Who wants to collaborate? (April 2022)

#106

Post by bellwether »

I'm a hands-on full stack engineer with a passion for product research, management, and development. I can do everything from Figma mockups, to database schema design, to backend and frontend development, to cloud infrastructure and scaling.

Does anyone need collaboration on a DeFi, web3 smart contract, or other FinTech project? If so, please reach out to me (email is in profile).

Re: Ask HN: Who wants to collaborate? (April 2022)

#107

Post by EZ-Cheeze »

skywal_l wrote: post #85
EZ-Cheeze wrote: post #21

From my recent post: "The thing that could bring the most change to the world is a system that finds everybody a thousand or more best friends based on psychological proximity and compatibility. Don’t tell me it’s impossible to predict whether two people are likely to get along or not based on a bunch of factors that can be analyzed with various techniques including AI. What would the world look like if every single…

What would be the kind of data you would need for that? Also, how would you kickstart it? This kind of site needs a huge amount of data to start becoming useful so there is always the problem of recruiting users while at the same time not providing any useful service.

-start with some rough seed data

-make matches based on our initial hypotheses of data vs action vs outcome

-learn from success/failure of matches

-adjust matching strategy & request more or different data as needed

-search for best predictors by babble and prune [1] with extremely prolific babblers and extremely discerning pruners.

GANs & transformers & hashtables, we'll grow em quickly out of the right SUBSTRATE; I think representation efficacy makes for the biggest performance jumps in machine learning training and inference speed. Constant and well-directed experimentation is key in innovation, let's get all the best people spazzing out constructively on a constant basis

Let a thousand distributed "Bell Labs" type bloom. (If you don't know, Bell Labs is known for many innovations and was ran with a spirit of collaboration, tinkering, and interdisciplinary work. And also it had all the right people in the same spot!)

[1] https://www.lesswrong.com/s/pC6DYFLPMTCbEwH8W "Babble creates a constant stream of coarse material, while Prune cuts these ideas harshly without the slightest hint of remorse."

Re: Ask HN: Who wants to collaborate? (April 2022)

#108

Post by EZ-Cheeze »

erikerikson wrote: post #74
EZ-Cheeze wrote: post #68

Earlier quoted context omitted.

It wouldn't be just friends, but that's how I phrased it there. I called it "supporters" in these 280 characters: "Some people just get along. They are compatible. They mesh. I think it's possible to predict whether any two people will hit it off, and to create groups composed entirely of highly compatible, highly synergistic people. What will happen when EVERYONE has a THOUSAND SUPPORTERS?" And it wouldn't just be f…

Okay. The problem is something like "social knowing doesn't naturally scale" but the entirety of the problem is how to make it do so effectively without ill side effect. I'm curious if you've gone the further step on how the system may simply iterate the current state. What ends do you seek? Most important... What have you done to start solving this problem?

emailed, plus see reply to skywal_l

Re: Ask HN: Who wants to collaborate? (April 2022)

#109

Post by Uninen »

I've been working for the past few years to build better (and free) tools for DJs and artists. I'm interested in the mixing (pun in tended) of arts and technology, and in creating communities in healthier spaces than Facebook, for example. It would be fun to find other artists to share ideas and collaborate on projects.

In addition of the most visible project which is a streaming site for DJs called Slipmat.io, I've also built a ton of related tooling around the process of planning, preparing and hosting (live) radio shows, for example:

- Playlist parser for converting playlists from DJ software to text and/or HTML for aftermarketing: https://playlists.slipmat.io

- POC show scripting app for collaborative planning of radio shows: https://scripter.labs.slipmat.io

- POC app for planning and organizing (virtual or physical) festivals across several timezones: https://eventplanner.vercel.app

- A tool for DJs to take requests during live events using their DJ app music collection as a source for listeners to search from (early demo from 2017: https://www.youtube.com/watch?v=T9HkLoKrTqs )

- A simple desktop soundboard app w/ MIDI-control: https://github.com/slipmatio/soundboard

I have a vision of building basically the GitLab for artists: a comprehensive tool stack for collaborative planning and playing live shows, marketing, and connecting with listeners, all in one easy to use place.

The current stack of all my projects is Python (Django, Starlette/FastAPI) and TypeScript (Vue). I've recently started to re-write the older stuff and in the process open source the parts. All the recent projects are now on GitHub: https://github.com/slipmatio

If you're interested in collaboration or testing out some of these tools, you can contact me either by email ville @ the project domain, or ping @uninen on twitter :)

Re: Ask HN: Who wants to collaborate? (April 2022)

#110

Post by ReallyAdjective »

vinner_roy wrote: post #62

I'm building pagespace.app, a platform to create, publish and sell longer form HTML-based content. The vision is a cross between Substack and Github for longer form content. I mention Github because much like code, good content and knowledge should improve and change over time and have lots of "commits". Tech stack is Elixir, Vue 3, Prosemirror/TipTap on Kubernetes. Looking for cofounders/partners.

This is an interesting idea.

I can imagine textbooks on this platform but we a collaboratively constructed knowledge graph.

I'd love to jump on a call and chat.

Post reply on HN