cyclonereef

Introducing A/B testing for videos seems to be along the same line of thinking as subscription payments for your car seat warmer[0] or surge pricing for fast food[1], where a concept has been hamfisted into something because a closely related concept or technology was able to utilise it well. Sensibility be damned

--- [0] https://www.theverge.com/2022/7/12/23204950/bmw-subscription... [1] https://www.foxbusiness.com/media/wendys-roll-uber-style-sur...

this is one of the few areas i feel like a blockchain-style solution could work. Mint a token, transfer it on sale or resale to the new wallet. Validate a wallet with the key is associated with the account. Same for old-school MSFT licenses or other systems where you buy a license key.

There's probably a shortcoming I'm missing here, but it's the only genuine case I could see a blockchain having use above other technical solutions

I can only speak for myself here, but if I feel like I've read a whole paragraph that should have been half a sentence, that's my signal there's possibly AI generated content in there

Got a referral to work as a Solution Architect at a company from the Senior Sales Person who I will literally be paired up with to talk to customers. Outgoing architect also said I would be perfect for the role. Got on a call with their internal recruiter who cut me off partway through most of my answers about my background and then received and automailer rejection.

That someone who would be bearing the impact of a bad hire recommended me and some recruiter who openly admitted she didn't know much about that side of the business made the call to drop me left a pretty bad taste in my mouth about the company

I've worked with plenty of companies that provide some sort of hosting for enterprise customers, and the number of times I've seen even senior admins use only CPU Utilisation and Memory In-Use investigating an issue is disheartening. And given that CPU Utilisation is an aggregate of all time != CPU idle, the same utilisation number can mean very different underlying system states.

There's something like a dozen different CPU metrics that can be referred to by the OS alone.

The author mentioned this in passing: > The tasks were menial but doable: ferrying credentials between services, clicking Deploy, watching something fail, pasting the error back to Claude, repeating.

It's no different to a company outsourcing code development for an application to a consultant or another company, just it's using an LLM instead. If that code breaks then the company will need to find another developer to fix it. Sure there are drawbacks to it, but it also lowers the barriers for people to try things out and prototype at low cost and low impact

I always gets the sense around a third of the way through his articles that whoever reads his drafts just gives up. It goes from wordy and repetitive to wordy, repetitive, filled with rage-bait exasperation and more filler than content.

Give the man a 2000 word budget and he could probably write a better article and cover the same information

It's the same sort of pedantry as correcting someone's use of their/there/they're. Yes, technically you're correct in what you're talking about, but arguing minutiae that is probably not relevant to the overall discussion.

I'm going to call.

This format of writing.

Partial sentences.

LinkedIn haiku.

_allegedly_ Until there's evidence of the CVE's the model found and the severity of them etc. it's just a statement and "trust me bro" vibes wrapped in a PR puff piece

His articles conflate quality with quantity. An aggressive edit with a more coherent structure would improve the message and sound less like a stream of consciousness rambling. Advertising his newsletter as "over 7000 words" is like bragging about LOC, it's an impressive number but doesn't itself indicate whether it's a necessary amount or if it's useful.

Unfortunately though I can't really find anyone else looking at this same information, so for now I have to wade through these newsletters to pick the gold from the shit

Most of the time checking for "typical" thresholds for infrastructure will yield more noise than signal. By typical thresholds I mean things like CPU Usage %, Memory Consumed and so on. My typical recommendation for clients who want to implement infrastructure is "Don't bother". You are better off in most cases measuring impact to user-facing services such as web page response times, task completion times for batch jobs and so on. If I have a client who is insistent on monitoring their infrastructure I tell them to monitor different metrics.

For CPU, check for CPU IOWait For memory, check for Memory swap-in rate For disk, check for latency or queue depth For network, check for dropped packets

All you want to check at an infrastructure layer is whether there is a bottleneck and what that bottleneck is. Whether an application is using 10% or 99% of available memory is moot if the application isn't impacted by it. The above metrics are indicators (but not always proof) that a resource is being bottlenecked and needs investigation.

Monitor further up the application stack, check for error code rates over time, implement tracing to the extent that you can for core user journeys, ignore infrastructure-level monitoring until you have no choice

There's a point at which you need to actually specifically refer to what low-hanging fruit there is, "trust me bro" and "do your own research" shows you either don't know or don't have conviction in what you're saying

Of those 800+ commits, how many were for the prompt progress capture as compared to the actual code? It also feels contradictory to state something was both "production ready" and in the next sentence say "respectable alpha version". To my mind at least the two are very different levels of completion.

The idea that there's a global cabal of ultra-wealthy real-estate owners who have globally conspired to control people to go back to the office holds about as much weight in my mind as the idea that there's a global cabal of ultra-wealthy who have globally conspired to create a virus to control the population.

Covid was an outlier event, and the outcomes that occurred as a result of it will revert to the mean over time.

Is it that Sun was particularly respectable, or that it disappeared before the current run of subscription-driven lock-in revenue strategies took off? I can't imagine any reason to think they wouldn't be doing the same thing as everyone else if they were still around as an independent company today

I hope it was, but in taking a quick view at the authors other works, it seems like that's just his writing style