Tidy- Offline semantic Text-to-Image and Image-to-Image search on Android powered by quantized state-of-the-art vision-language pretrained CLIP model and ONNX Runtime inference engine

Features

• Text-to-Image search: Find photos using natural language descriptions.
• Image-to-Image search: Discover visually similar images.
• Automatic indexing: New photos are automatically added to the index.
• Fast and efficient: Get search results quickly.
• Privacy-focused: Your photos never leave your device.
• No internet required: Works perfectly offline.
• Powered by OpenAI's CLIP model: Uses advanced AI for accurate results.

F-Droid

https://github.com/umutcamliyurt/GoCry

Works really well, personally only tested on Linux, but I love it!

At this point I'm very concerned about the open source industry relying so much on github. You have to remember that any project there can be swept away overnight because it doesn't fit into the agenca of a large company, for example.

I hope this project gets more contributors to help make it as good or better than Newpipe. Written in Dart using Flutter can allow it to be compiled for Android, iOS and desktop

https://gitlab.com/christosangel/magic-tape

Magic-tape is an image supporting fuzzy finder tui YouTube client.

UPDATE

Now introducing a new feature: the video description as well as the comments written by YT viewers will be shown in the terminal window, while the video is reproduced.

Thus, the user can be satisfied reading other viewers having a swing at the politicians/celebrities/stars they love to hate, or, watch closely to their heart's content, as cyber nuclear attacks are launched between self-righteous, valiant and livid keyboard fighters.

Comment loading is asynchronous to video loading, so it is possible that there will be some delay in the appearence of the comments. That depends on the number of comments, network speed etc.

Hello everyone! This is a small program I made yesterday to render the Mandelbrot Fractal with beautiful colors!

It isn't as fast as other programs (e.g. XaoS) but it is the first good program I have made using OpenGL. I may update it to render some other fractals too in the future (e.g. The Burning Ship).

I hope you like it!

Cubedex is a lightweight Progressive Web App (PWA) that connects to your GAN smartcube using Bluetooth. It's designed to help you drill, time, and master algorithms like PLL and OLL, making it easier to build them into your muscle memory faster and more effectively.

📱 How to Get Started:

✅ Visit https://CubeDex.app in your browser ✅ Add Cubedex to your home screen for an app-like experience ✅ You can use it offline - Cubedex works perfectly without an internet connection

Cubedex has been created with ♥ by Pau Oliva Fora using gan-web-bluetooth and cubing.js.

If you enjoy using Cubedex, please consider supporting the development on Ko-fi.

Hello, I started donating to my favourite open-source projects a couple years ago, but stopped about 6 months ago for different reasons and wanted to get back into it.

I wanted to ask if anyone here has a set system or process they follow when donating

• How much money do you donate? A set amount, whatever you feel like, a percentage of your earnings?

• When do you donate? Whenever you remember, on the first of the month, Thursdays?

• Do you have a minimum donation amount?

• How do you decide what projects to support? Do you forego donations if you've contributed in other ways? Do you keep a list?

• Do you donate to all equally or do you have some sort of ranking? Is it by amount of use, subjective preference, something else?

• What platforms do you prefer using? Liberapay, Opencollective, Patreon, ko-fi, Paypal, Monero, actual post?

So far the system I've devised for myself would go something like:

• put 2 % of all my earnings, whatever they are, in a separate account
• every quarter (on the first of January, April, July and October) donate the full amount of money in the account (with a minimum of 5 €, so as not to lose a big amount in fees)
• keep a ranked list of projects that I've used or deemed important or promising in the last three months (projects I donated to recently go to the bottom of the list), things at the top get more money than things at the bottom
• prioritise Liberapay since it's open-source itself

After trying to understand if Synapse was still the preferred community method for Matrix after the post on Element X/ESS, I thought I'd share this.

Element Starter is a game-changer, offering a free (as in beer) self-hosted version of Element Server Suite. It is a lightweight version of the officially supported Kubernetes-based components found in our paid packages (Business, Enterprise and Sovereign) - the very same stack used to power the biggest Matrix homeservers in the world; built by the team who created Matrix.

Element Starter is designed to allow anyone in the world (who doesn’t need to start with our powerful paid-for enterprise features) to adopt Element Server Suite for free. This option gives them all the real-time communication functionality they expect from a leading messaging and collaboration app with the added benefit of self-hosting to ensure data ownership and control, while building on a wrought-iron foundation which is futureproofed for commercial support and features on demand.

Really surprised I can't located migration methods anywhere, though. And that all the Element X app repos are still labeled pre-release but the blog says they are production ready.

Edit: This does require sign in and terms of use agreements, so I'll be sticking to my Synapse docker container for now.

Zuckerberg claims that “Open Source AI Is the Path Forward.” While it’s easy to agree with Zuckerberg’s sentiment, Llama 3.1 isn’t truly open source and wouldn’t meet the criteria for compliance under the OSAID. This raises important questions about how to engage with Meta: should the open-source community push them away, or guide them toward creating OSAID-compliant models?

In the second finding of the 2024 Tidelift state of the open source maintainer survey, we found that the more maintainers are paid, the more improvements they make to their projects.

...

In the previous finding, we reported that 60% of maintainers describe themselves as unpaid hobbyists, and 36% of maintainers describe themselves as paid (professional or semi-professional) maintainers, earning some or all of their income from their open source work.

...

When you break down the paid maintainers into professional (earning most or all of their income from their maintenance work) and semi-professional (earning some of their income from maintaining projects), it becomes clear that the amount of money a maintainer is making for their work has a large impact on the types of improvements they are able to make. Across nearly all major categories, professional maintainers are on average over 20 percentage points more likely to make key improvements to their projects than semi-professional maintainers.

...

In the previous study, 81% percent of professional maintainers earning most or all of their income from maintaining projects spend more than 20 hours a week maintaining their projects. This year, the percentage was nearly identical (82%).

Conversely, in last year’s survey, we found that the vast majority of unpaid hobbyists spend ten hours or less per week on their maintenance work (81%). This percentage also stayed consistent in this year’s survey, with 78% of unpaid hobbyist maintainers working ten hours or less per week.

...

We’ve heard from many maintainers that how they are paid for their work also matters. For many maintainers there is a huge difference between getting a one-time “airdrop” of money, perhaps right after a high profile incident where people are paying attention to their projects, compared to ongoing recurring income that they can count on. So this year for the first time we asked maintainers to tell us whether they would prefer to get predictable monthly income or a one-time lump payment.

An overwhelming majority of maintainers prefer to receive predictable monthly income, with 81% choosing that option.

https://github.com/umutcamliyurt/IrisChat

I’m just looking for a simple 2D floor planner that is easy to use and has the ability to show measurements.

The two most recommended open source tools I’ve seen have shortcomings.

Sweet Home 3D:

• Very 3D centric, which over complicates things.
• Project was recently acquired/handed off to a company and the iOS version is published by “AI Photo Editor Lab SRL”

draw.io / diagrams.net:

• Very finicky to set up and create floor plans
• To display measurements one has to add measurement objects to every wall/object

Does anyone have other recommendations?

Hi

I'm looking for a FOSS app to design icons

I heard of GIMP , Kirta and Inkscape ,are there other alternatives ? and which one of them is the best for this ?

Thanks

Can't imagine using my system without this.

Element is launching the world’s first communications platform based on the upcoming Matrix 2.0 release. The result is blazing performance which outperforms the mainstream alternatives - across a decentralised system that enables self-hosting and end-to-end encryption - as well as open standard interoperability to revolutionise real time communication between large organisations.

Built on Matrix 2.0, Element X now rivals the performance of centralised consumer messaging apps, empowering organisations to address the shadow IT issues caused by consumer-grade messaging apps in the workplace.

The new Element communications solution consists:

• Element X, our next-gen app with an array of new features
• Element Call fully integrated into Element X, for native Matrix-encrypted voice and video
• Element Server Suite, our backend hosting solution for powerful admin control and Matrix 2.0 performance

Hi, was wondering if anyone knew of an app where you can use your camera to scan documents (like Adobe Scan) which is FOSS.