TheBigRoomXXL

How the data is stored and persisted ? Is is just an index dumped to the filesystem?

You can add X-Clacks-Overhead as a custom header. Nice.

tinyfeed creator here, looks like it's already been a coople of days since this has been shared, but I am available if you have any questions.

It's always lovely to see tinyfeed appear randomly on the internet :-)

A fun hack to improve my workflow.

Very interesting concept. Im going to keep an eye open for a new project where I can test that.

I was about to comment there is some interesting intersection with residuality theory but then you mentioned it ^^

> Traditionally, lightning protection has relied heavily on lightning rods. However, their protective range is limited, and in some cases—such as wind turbines or outdoor event venues—it may not be feasible to install them. At NTT, we are exploring the use of rapidly advancing drone technology to create a new approach: "drone-triggered lightning"2.

I can't believe that's a practical solution. Surely just installing more lighting road is simpler et more effective. They just want to do something cool and try to justify it sideways.

I’m really curious to see how productive it actually is to use something like that. This combination of a framework, database, and infrastructure all in one package seems to simplify things, but I wonder if everything falls apart the moment you try something a little 'outside the box' or unusual.

One year ago i started to work on TinyFeed, I didn't like existing solutions for feeds reader. Many of them seemed overly complex, most of them require a database (and didn't event support sqlite). For a simple feed reader, it felt overkill. What I wanted was an easy to setup app and with a clear and minimal interface like HN. Now, I’m happy to announce that it has slowly improved since then. It’s now reaching its first stable release, version 1.0!

The latest release highlights:

- Daemon mode: Supports background operation for continuous feed updates, allowing TinyFeed to run as a service on a server or a laptop.

- Interface redesign: A simpler, more readable web page.

More generally it protects against anybody who has access to the database, including bad actors if it's leaked.

I don't think it protects against timing attack because the common way of doing it is just to use sha256 and use the resulting hash to do a lookup in the database. This is not a fixed time operation

I have been thinking about the sync-only-server model for sometime and providing a way to use application specific logic, like CRDT, while keeping the server generic has been a major headache for me.

This solution/pattern is the best I have seen so far. It does not require anything special server side, all the logic live on the client, which also make plugins way easier.

Thank you!

Look great. I am going to try it this afternoon!

I remember looking at it sometimes ago and indeed it was not easy to use. Maybe it's time to try again!

Awesome! I looked for GRPC in the issues but didn't think to look in the discussions.

Next month my team will be dedicated to improving our testing capabilities, maybe some of that time could be dedicated to contributing GRPC capabilities ( if I am capable enough).

Indeed.

Also buying power is just not the same based on your location. A hundred dollars in the US and in the Netherlands will literally not buy the same things.

Not an expert on ML model, what does the smallest box represent exactly? Function calls?

A weird artwork. It feels like watching the landscape during roadtrip.

Also a great example of brutalist web design.

404 page.

If you care about alt in markdowns, isn't it simpler to use the <img> tag directly in the markdowns file? Markdowns does support inserting html.

Nice read. Very detailed.

That kind of optimisation make a big difference when I am trying to load a webpage on the train. Thank you for caring!

The directive has not finished the legislative process yet so it's difficult to be exact but according to the briefing on the New Product Liability Directive redacted by the European Parliamentary Research Service

> In the aim of not hampering innovation: (i) free and open-source software developed or supplied outside the course of commercial activity, as well as (ii) the source code of software, should be excluded from the definition of products covered under the proposal.

https://www.europarl.europa.eu/RegData/etudes/BRIE/2023/7393...

Be liable doesn't only mean that if the software you sold *harm* people then they can sue you. It doesn't mean they can sue you for every little bug.

So yes if you sell software, whether it is open source or not, you better have the balls to be liable.

You couldn't sue me for 2 reasons:

1 - This regulation only concerns commercial activity. So you could only sue the company I work for, and only if you've bought their products. Also by definition that excludes my personal projects.

2 - You can only sue for defects (in this legal context it means unsafe to use) or damage (physical or material). You can't sue for simple bugs.

These kinds of liabilities already exist for all the objects in your life and yet you don't spend your time suing people every time something does not work as expected I imagine

No you are not liable. Liability is linked to a commercial activity because it is meant to protect consumers.

The article is very ambiguous in the way it describes the regulation. I recommended this one for more clarity : https://www.euractiv.com/section/digital/news/eu-updates-pro...

This is great. Software is important, software has an impact, and so we need liability.

This regulation ensures that whoever sells the software to the consumer is responsible, and that's the way it should be. The creator of a library doesn't know how his library will be used in the wild, he can't anticipate all possible problems, the product maker can. It is the product maker's responsibility to integrate external components properly, having validated that they are up to standard.

If you're a manufacturer, you can't just pick components at random and then say it's not your fault if your product doesn't work. That's why manufacturers have whole teams of people working to ensure that what they receive from a supplier is up to spec.

What really get me is the FantilatorOS. They even open sourced it: https://github.com/justinasr/FantilatorOS/tree/master/kernel

I open it on mobile and got an entirely white page

Great article! As a self-taught web developer I find that kind of resources very valuable. It help me to move away from all the abstractions and towards a more concrete understanding of how thing work. Thank you.

> And most importantly they require streets that are designed for cars, not people or transit

I think this is one of the most interesting point the protestors make.

This kind of solution doesn't seem to integrate very well with new practice of urbanisms where pedestrian, bike and traffic intermingled in the same space and travel distance are shorter.

But I am European so I am probably bias because our city are already partially car free.

I also think that validating an API against it's OpenAPI schema is a great methodology. You should checkout schemathesis, it's fantastic for doing that.

https://github.com/schemathesis/schemathesis

Simple and clean, I like it.

But the "let's get to work" button doesn't have enough contrast in light mode.