HN user
Quite interesting. I just woke up (it's 7AM here). A quick check and everything works as before, so public pages are showing up without requiring login and with the old design. I don't understand what kind of human experimentation we were facing.
Thank you for your valuable contribution.
Please read the question again. It's not about my page, it's about all pages.
This way X and music also works: https://github.com/kpocza/LoWe
Of course, it's a bit circumstantial...
There are no specific requirements for the C# version, the library was written in v7. Go# compiles to three targets: .NET Standard 1.2, .NET Standard 2.0, and the full .NET Framework 4.5.2+. The same three are available in the NuGet package published to nuget.org.
There are not so many allocations in the library where Span would help still it's already quite performant at the moment. If you have any production feature requests please don't hesitate to file an issue on GitHub or send a pull request.
I haven't compared it to TPL Dataflow however made some pure performance measurements. The results are at the end of this: https://github.com/kpocza/GoSharp/blob/master/src/GoSharp.Pe...
Basically Go# can pass millions of messages in a second.
And yes, a Go#/Go channel is similar to something like BufferBlock however both are designed to fit into different "ecosystem" and way of thinking.
Go# channels are thread safe and have been designed to operate in parallel environment.
Just a quick note. This is a small library for .NET/C# that implements the most important concurrency elements of go like channels, selects, timers, tickers. Don't hesitate to tell your opinion if you have suggestions, comments, ideas, etc. Thanks.
Try this one: https://github.com/kpocza/LoWe
Originally there was no actual need, I was just curious about the technology, the limits and I wanted to push these limitations.
But: Later I realized that there is a need for this. With apt, it's much easier to install and update (GUI) apps than with the Windows installer, so the need has grown from nothing to a real-world requirement. Moreover it happened several times that I was looking for some software and there were many alternatives for Linux but nothing valuable for Windows. LoWe will help in these situations, as well.
Most probably to employ a third-party X server is more robust at the moment.
However in this case we can achieve the same effect with the Ubuntu bits of the X server. It's true that here we require other tweaks but this is a bigger step towards letting Ubuntu on Windows on its own to run X apps.
LoWe provides a generic approach and idea that can be applied in a specific way to other functionalities. Eg. it's possible to play music (PCM) by "emulating" some ALSA devices.
I know that there is room for improvement.
What do you think, where should I improve LoWe first? Should I add more apps, more devices, or make it easier to use?
Thank you.
Thriot is an open source IoT platform. There are no real usages yet, but since it's open anybody can make use of it. The main advantage of this platform is that it has configurable and extendable storage system so you can store the configuration, telemetry data and M2M messages where you wish besides the fact that it has been tested on Windows and Linux, as well.
Code: https://github.com/kpocza/thriot
Documentation: http://portal.thriot.io/
Demo site: http://central.thriot.io/