People have been talking about this for over twenty years, and here we are. Have you got anything better than the current certificate process? The truth is that certificates and their management suck, but that's the least bad approach that we have been able to come up with so far.

The truth is that we depend on many components that can fail at any time and that most of us will have no clue how to fix. Your are focusing on one, while apparently taking the reliability of the others for granted. Yes, your certificate renewal automation can fail, but so can your computer, modem, router, power, etc. That's life.

Well, Letsencrypt's framework looks daunting to the uninitiated, but the truth is that anything to do with this type of certificates does anyway, by their own nature. I set up mine a while ago, with safeguards to notify in case of failure, for three separate Linux systems driven from a single one, and it was not that complicated. Granted, I am more experienced than the average Joe, but we are talking Bash scripts twenty lines long or so, not convoluted stuff using lots of different files, libraries and whatnot, and requiring an expertise on lots of different issues, include the details of what the heck the blasted certificates do. I.e. something pretty much within the capabilities of a somewhat competent sysadmin.

Hardly surprising: if rigorously correct this would be a recipe for keeping a heart alive indefinitely. Which is, of course, absurd.

However, the HEP community will have to do significantly better than that to justify billions of dollars to be spent on new particle accelerators without a clear goal beyond, say, hoping to find super-symmetric particles.

May it rot in Redmond.

Coding falls into the realm of software engineering, not computer science. The teams of coders that develop a web site are, by no stretch of the imagination, computer scientists.

Bigly. Very bigly.

Not if you are well versed in cosmology.

Paraphrasing the immortal words of Mandy Rice-Davies, "Well, they would, wouldn't they?"

Well, they are gains nevertheless. This aside, you really have to be revolting if working for Zuckerberg.

We already have Xitter, which most certainly lives up to its name.

It's another incremental step forward; one of many more that will be required before a QC can be developed that is able to do anything of practical interest that cannot be done by digital computers at least as efficiently and far more easily and cheaply.

Human empathy is a trait that this bastard never developed.

I came to the conclusion long ago that the members of the Gnome team are a bunch of lunatics living in some La-La-Land of their own.