So funny thing, a colleague of mine was arguing that there's no reason to do a compiler because he got Claude to generate ELF binary directly for a hello world...
Some of the AI bros are very extremely bullish on it too the point of saying compilers are a waste of time..
Hence my references to DNS records in addition to MX to determine if an endpoint is SMTP only or better. For compatible interop.
Looks like replacing IMAP rather than SMTP, which is significantly less thorny.
Frankly, I'm rather shocked it hasn't happened by now.
We've tried to bandaid over it with various weird things, but fundamentally SMTP is just weird.
I could easily imagine a domain having some SRV record to advertise "hey, I support NeoSMTP", and peers switching to that instead, possibly pinning on successful NeoSMTP negotiation.
Get rid of a lot of the relaying which has not been useful in modern times and only a liability, and perhaps offer an https option to be less likely to be blocked (a TXT record to indicate url, and a node could declare their NeoSMTP entry point as https://mail.server/NeoSMTP/ or whatever works for them).
There are pedestrian safety design points including pedestrian crumple and not having sharp edges.
Of course the goal is moot at 75 mph, but it does apply at speeds like 15 mph that might actually be survivable and isn't a crazy expectation in pedestrian heavy roads where the car might actually slow that much before impact.
So the cyber truck totally deserves criticism for discarding that in service of Elon's weird taste, but just a deer at 75 mph didn't illustrate it.
I was so happy when I ran out of tokens, I could just keep programming without management micromanaging how I got my job done.
The author seems to think this is some horribly intolerable scenario, impossible to keep going. That without tokens you can do nothing. Lamenting that it's impossible to understand the code the LLM put out. This is a bad sign. If you let the LLM get beyond your ability to reasonably understand and audit, then you run the risk of getting screwed at any moment. If I saw this perspective from anyone that did work for me, I told be avoiding letting them anywhere near anything potentially important.
Net this has been the worst thing to happen to the discipline of software development. Maybe after the bubble pops we can have the more reasonable takes, but right now it's just insufferable.
While the truck is certainly poorly designed for pedestrian safety, the truck hit the deer at 75 mph. There is no such thing as safe for pedestrians at that speed.
It's not the creators intent, is the closed captioner. The director does not play a role in what goes into the closed captions generally.
Yes, but those LLMs are not the official LLM of X directly tweeting on the platform publicly for all to see, attribute directly to the LLM instead of to whatever user induced the behavior.
Well a light refactor of deduplicating code would be a refinement that is generally easy to review and likely comfortably in the capability of the codegen.
Versus creating, which generally is harder to review and requires the submitter to explain the rationale behind the change.
Volume of code and nature of change.
Of that comes to pass, they can change it.
The codegen is a particularly big problem for famous open source projects. All of a sudden you have a flood of people trying to get their name on your project and flooding you with low quality stuff. Codegen enables folks without good ideas to generate submissions that demand a lot of review. Being about to shoot down obviously AI content without deeper review is pretty critical to keep the project viable. There may be good things tossed out with the bathwater, but it is a ton of bathwater. Besides, if there's something real behind it, a good human is likely to emerge to take it forward more credibly.
Closed source or personal projects can enjoy a bit more judgement. You can evaluate the people independent of the submissions and may be able to have an even more nuanced take.
Based on my AI enthusiast colleagues, I totally believe 100x productive.
100 x 0 == 0