Defining, detecting and ultimately stopping fake news.
That sounds very interesting. Tea?
That sounds very interesting.
Well it's ambitious, so it has that going for it...
Tea?
I'm not sure I get the reference. Yes?
Ask HN: What are you working on?
211–220 of 751 posts
Re: Ask HN: What are you working on?
#212Project: Hosted & On-Prem fast full-text search with faceting, filtering, multiple ranking algorithms and plenty of other features. Not yet ready for launch but built a simple demo trying to get into startupschool ( failed unfortunately :( ), which lets you search every hackernews post while letting you filter based on domain / user / story type. http://searchhn.com
Cool project. Could you briefly talk about * the backend you use and how it will scale to sites with large amounts of data across servers * can third party sites integrate your search service? * How is it different from eg- Algolia Good luck with the project!
Thank you !
Backend is custom built written in C and assembly. Supports sharding and replication which is rack aware and data-center aware.
> can third party sites integrate your search service?
Yes of course.. that is the end goal.
Algolia is awesome.. but you end up paying a lot based on how many ways you sort / rank data. This operates with an on-the-fly ranking model and rank on any field in any direction. Also different ranking algorithms, extensibility with Lua and a lot more when I officially do a showhn
Re: Ask HN: What are you working on?
#213Re: Ask HN: What are you working on?
#214Currently working on quering OpenStreetMap using graphql.
Working demo over here http://kushanjoshi.com/owesome/ .
Github https://github.com/kepta/owesome . Would love to talk more about it.:)
Re: Ask HN: What are you working on?
#215Re: Ask HN: What are you working on?
#216I have a project to get to know something about ML in computer vision. The project I'm doing that with is OCR of a noisy, low-resolution image (~32x16) Main problems: * Since the resolution is very low, I found it hard to use traditional CV approaches to segment individual characters and classify those (for example: is this rectangular blob a few pixels diagonal a dash symbol or just noise) * End to end CNN approach…
Have you tried this. Convolutional recurrent network in pytorch. https://github.com/meijieru/crnn.pytorch
I think this is the current state-of-the-art (or close to it) for OCR on noisy low-res images.
Here is a broader list of resources/papers for scene text recognition https://github.com/chongyangtao/Awesome-Scene-Text-Recogniti...
Re: Ask HN: What are you working on?
#217I built a mobile app for fitness enthusiasts to connect with like minded people around the world.
Mark your gym on a map and send messages to people that are working out near you. Post funny gym stories and pictures to motivate others.
Website: https://gympulsr.com
The biggest problem during development was the framework itself. As there aren't many people out there that are using Qt for mobile app development, it was sometimes a little bit difficult to find help. In case you are interested, here is a lengthier post about my experiences with Qt for mobile app development: https://gympulsr.com/blog/qt/2017/02/23/working-with-qt-mobi...
The biggest problem now: Get people to use the app.
I honestly didn't thought this would be that hard, but unfortunately it's really tough to get some traction and create a solid userbase.
Re: Ask HN: What are you working on?
#218Problem: My experience taking audio tours at various museums showed the use of antiquated and expensive hardware. Most alternate solutions used mobile apps which are inconvenient to download and take more time to release. Project: I went about developing a web app that allows anyone to quickly create an audio tour for free: https://www.youraudiotour.com/create I also integrated Amazon Polly to automatically generate…
Nice! I'll be passing this along to my museum friends. I could see this being especially useful for smaller museums that may not have the budget to buy a full application or edit the audio. Do you think you'll still keep the custom-audio option, or only Amazon Polly?
Re: Ask HN: What are you working on?
#219Two projects recently started that I expect to have available in the coming weeks:
* a MUD-style game to teach the principals of Erlang concurrency and the actor model.
* Take the half-dozen blog drafts I have, clean them up and publish them. Fix my broken personal site in the process.
Re: Ask HN: What are you working on?
#220Earlier quoted context omitted.
That sounds very interesting. Tea?
That sounds very generic too. How do you even define fake? How do you know it's not fake?
I think a better question to ask is: how do we structure discourse around content to make our society better immune to the proliferation of low-value content? (Whether that be fake news, clickbait, disguised advertising, etc.) Also, how do we do this without creating political silos, each with their own propaganda?