Re: [seL4] Benchmarks of seL4

(sorry for the double-send, I used the wrong From email on the first attempt so the post didn't make it to the list) Jeff Waugh writes:

On Thu, Apr 16, 2020 at 1:20 AM Leo Gaspard wrote:

...

Thank you for your reply! I must say I find this stance strange in a way: sure, people can say you cheated if you published benchmarks for other systems, but… is it worse than them saying you make claims without providing any grounds for them?

I suspect Gernot would prefer to avoid being beaten with his own stick. :-D

Oooh I hadn't noticed the Gernot's here were the same! Well, then, thank you for the list & paper, they're really nice to keep at hand when benchmarking :) That said, if the problem is actually configuring the competitors correctly, then there's something that differs from the context of the paper I think: when writing a paper, it's written once and then fixed in stone; while data on a website could be evolutive. In other words, explicitly saying “here are all the configuration parameters we tuned for the competitors, if you know of a way to make them better please tell us” should hopefully be enough to alleviate any concerns that the competitors were unfairly treated… would it not?