BaronSamedi

Great article. It would seem that prompt injection attacks in the name of "anti-LLM" measures are now something to protect against too. OWASP has a pretty good page on the topic of prompt injection. I suppose we are going to now need a kind of uBlock style plugin to strip these attacks. Probably also some code agents can use to strip these attacks and filter them out when they fetch pages.

LLMs can write Common Lisp just fine. I have no idea where the myth that they can't because of the all the parentheses comes from.

If any of the alternative search engines would offer this and a feature to block pages with affiliate links--like those "review" sites--I would be happy to pay for them. They need to offer ways to customize the search experience.

It is possible to agree that content creators should be paid and that ads are bad. Substack, for example, allows you to pay and doesn't have ads. Ads are one business model, not the only one.

Thanks to the magic of modus tollens, absence of evidence can indeed be evidence of absence. "A 100kt nuclear bomb just went off in your front yard!" Me, looking out the window, nope. It depends, of course, on the argument being made. It could be true in some cases.

As cliches go this one is better than "you can't prove a negative"--not true. At least the "evidence of absence" one makes you think about causality and the nature of evidence.

"The point of experts is they have thought about a field more and are better at making sound arguments in that field."

This is an excellent point. It is so good that I think it may serve as the definition of an expert: An expert is someone who is better at making sound arguments in that field. Sound arguments being what matters and not the person making the argument, just as the calculations matter more than the mathematician.

Ironically, if we apply your first statement generally you are likely excluding yourself from the debate.

"If you are not a logician your opinion on fallacies doesn't matter at all."

I thought his statement that all requirements should have a name assigned to them was very good as well.

I agree with you. The idea that using e-prime makes one's writing more clear isn't a "fallacy". People might disagree but it's hardly fallacious. There are many items in this list that aren't fallacies (e.g., "Dog-Whistle Politics"). This list seems to be a mix of actual fallacies and "Things I don't like".

If it is a fallacy we should be able to construct a simple argument from it. With the "Be-verb Fallacy" what would that be? Perhaps, "You used e-prime in your argument but all people who use e-prime are part of a cult, therefore you are part of a cult (and are wrong)." That's a fallacy all right just not the "Be-verb" one.

I also found the definition of a fallacy at the start lacking. It conflates good argumentation with rhetorical technique. I prefer T. Edward Damer's definition from my old logic textbook:

"A good argument must have premises that are true or acceptable, premises that are relevant to the truth of the conclusion, and premises that together constitute good or adequate grounds for the truth of the conclusion." Fallacious arguments violate one or more of these three principles.

For your home network DNS: a server with dnsmasq -> stubby -> DNS over TLS -> NextDNS works quite well. Fast with dnsmasq caching and reasonably secure, plus NextDNS can block unwanted domains (ads, malware, etc.)

After many years working with these tools I agree. Declarative and template DSLs are OK for basic things but you quickly bump into their limitations when you want to do advanced configurations. At this point administrators should know how to code so being able to use a normal programming language with IaC modules should be the standard.

Amazon's CDK for AWS is a step in the right direction. Microsoft's new Bicep for Azure misses the mark in my view because it is yet another DSL and not a real programming language.

Scalability is just one reason for using serverless. There are many others. Most of which have to do with being able to focus on the part of your code that actually does something instead of on infrastructure and integration. For example, not having to configure and maintain VMs, containers, or Kubernetes, off-loading a part of security onto the cloud provider, and built-in integration with supporting services like logging and metrics.

Serverless, like every design decision, comes with its own set of trade-offs. Whether those trade-offs make sense for your environment is something you have to determine. At my company we have seen a productivity increase and shorter development times using serverless. Of course, that was after the learning curve which is certainly not trivial.

One rarely discussed effect of drug legalization is that pharmaceutical companies would start developing recreational drugs. I expect they would create milder versions of current drugs specifically for the recreational market. That market would almost certainly be huge. Because of this, Big Pharma ought to be one of the biggest proponents of drug legalization. Whether the net effect on society would be positive or negative I do not know.

Your are conflating what is practical with what is true in logic and argumentation. It is practical to accept our doctor's diagnosis because we trust that the medical system produces competent doctors and because of constraints on our time. If you claimed that a diagnosis must be true because it came from a doctor then you would appealing to authority.