On arXiv today: An approach to detect fair queueing in the network and adapt your congestion control approach depending on whether you find FCFS or FQ at the bottleneck arxiv.org/abs/2010.08362

Christian Huitema @[email protected] This is a good start towards just using ecn as a suggestion. This paper (pre-print) arxiv.org/abs/2206.10561 proposes a mechanism to monitor the presence of FQ continuously during a flow’s lifetime. This can be used to change the congestion control depending on the presence of FQ

I was wondering if my cable modem supports fair queuing, so I wrote a simple measurement tool. It can test for fq_codel, cake and fq (and possibly more): github.com/muxamilian/fq-… In my simple testbed it achieves an accuracy of >99%

github.com/muxamilian/poo…: Using the front camera and tensorflow.js, the smartphone becomes a “window” into the real world. Video and image content appear as if they were seen through this window. This is especially useful for content captured using ultra-wide lenses.

Robo99 is a readability-optimized font whose glyphs were generated using Machine Learning. It uses a custom GAN which has an additional loss term to encourage maximum readability. The resulting font looks a bit like Comic Sans muxamilian.github.io/Robo99/

Variational Autoencoders (VAE) use complex Bayesian hocus-pocus to achieve their goal of generating new data. The Straightforward VAE can achieve the same goal using a far simpler formulation: github.com/muxamilian/str…

"duralava" is a neural net which can simulate a lava lamp in an infinite loop. The recurrent GAN can generate an arbitrarily long video of a (virtual) lava lamp, without diverging even after thousands of frames. github.com/muxamilian/dur…

WYSIWYH is a neural network that transforms an input video into an audio sequence in real time. It does so by compressing each image using an autoencoder and interpreting the resulting code as a frequency range. github.com/muxamilian/wys…

One way to smooth Internet congestion control are agreements between content providers (like NVIDIA GeforceNOW) and ISPs. An ISP guarantees that the network bottleneck has fair queuing. Content providers can then use mellow congestion control. Customers benefit.

Can Bing Chat (GPT-4) come up with original thoughts about research? Not so much but it mentioned my paper, so I'm still flattered 😀

Fast fair queuing detection in C: github.com/muxamilian/fq-…

Apple's fq_codel implementation (github.com/apple-open-sou…) seems to do "true" fair queuing, meaning that all flows are completely separated (unlike fq_codel or cake in Linux, which assign multiple flows to one bucket if they have the same hash).

Only smart LLMs can understand good essays BusinessIntelligence Dr. Khulood Almani | د.خلود المانع ipfconline Dr. Theophano Mitsa ☦️🇬🇷🇺🇸 Tony Moroney 💙 #TechForGood 💙 Françoise Morvan #CES2025 Fati Sule Eveline Ruehlin Prof. Sally Eaves Dr. Monika Sonu I Founder Healthinnovationtoolbox muxamilian.github.io/essay-entropy-…