They also consistently demonstrate massively worse blind spots around false communities. The blind faith towards "person from Discord / reddit" is truly extreme.

Part of it looks like the general social dysfunction from social media overexposure - conflating acquaintances with friendship and such. Hustle culture plays in, as integrity is just something for old people and "not the way the world works". Then they rage at companies and CEOs like they're personal friends beholden to conform to their worldview and in-group politics.

I mean it is all tribalism, but like zero effort, zero intellect applied tribalism.

If they have the technology to descramble the average reddit post, they're already sitting on trillion-dollar AGI tech.

It's about these things in balance.

Suburban homes sourcing most / all of their drinking water from bottles is insulting unless there is a Flint water situation due to toxic city leadership. Having a crate of bottled water as backup / leaving the home is more reasonable. Abolishing straws or going to paper ones was always at the opposite end an absurd, insulting virtue signaling. When it comes to changing what something is made from to make it easier to biodegrade, the first question should always be energy input. How much more energy and cost accompany things like reformulating straws' plastic.

Disposable diapers are a necessity of the dual income household. There are a variety of other disposables and consumables which fall in the same category where given the option and time, many could / should / would use reusables. The dark gray area are things like laundry and dish detergent pods. They probably shouldn't exist, but portion control isn't just an eating problem with many people.

Going from selling something a thousand times instead of once, the selling something a few times extra instead of once though is its own form of rent seeking by defect. We're seeing this in the automotive world with PZEVs, direct injection, and SCR systems all designed to incur costly repairs which shouldn't have been necessary.

A huge failure in F/OSS is the diversification of hardware and platforms which makes supportability a nightmare. And Broadcom / RPi as an organization has talked a substantial consumer base into toxicity against their peers by subscribing to the ephemeral hardware cycle. Ironically, Broadcom isn't the worst actor, or even a particularly bad one. It's ARM and the ARM community pushing the hardest for compatibility breaking at every incremental release.

In the hardware space as well as a lot of maker related software spaces, vendor support is live or die. There is a large enough consumer population to use the products and make them profitable. There is not a large enough consumer population to convert enough over to producer to keep F/OSS efforts alive. Easy software example from 3d printing where a lot of this semi-embedded / almost enough compute is used is Varislice being open sourced by Autodesk. Yes, that Autodesk... gave something away for free. And it promptly died. Literally a gift horse the recipients aren't cowboy enough to keep alive.

Education and tinkering was the original sales *claim* up through the Pi 2, but it was never how they operated. PiOS and other upstream development has always been about creating a market of embedded general use cases, but that must shoehorn and conform to absurd platform limitations. The whole Compute Module effort was about admitting it's just a SoC board and that more serious uses needed big boy adult hardware when getting into GPIO and shared busses among accessory hardware.

Heck, some of the major "F/OSS" software verticals effectively only run on RPi hardware without a LOT more tinkering to get running on other platforms and are openly hostile to even mentioning x86 general PC cross platform use.

<quote><p> tailored for financial gain...</p></quote>

Somehow people seem to forget a monetized YouTube channel or podcast fall under this category and think because it's only making them a few bucks, the rules don't apply. And parody as well as review / criticism fair use cases are similarly misused to define whatever the thoughtless want ad nauseum. They wouldn't know what transformative meant if a high voltage power line fell on them.

Do they know they're over 20 years too late?

Probst did the killing, but Riccitiello started packing disks with vacuum-desiccated remains long ago.

Pi 3-5 should have all supported Wake on LAN and 4-5 Wake on Wireless, but they're all mediocre products sold on brand recognition at the end of the day.

Once the higher end RPis broke $80 (higher RAM 4s, but I recall 3Bs having odd price spikes too) and the NUC-like x86-64 PCs exploded with fully functional (SSD/NVMe + RAM) units <$250, the value balance for Pi was over -- and this was before 2020 issues. The Pi still needed $30-120 in parts to make "complete" whereas the PCs just needed $10-20 in an Arduino, STM32, or Pi Zero for GPIO, if you needed that sort of thing.

<quote>rehearse the demo ad nauseam until you've found a solution that works and then stick to it absolutely verbatim. </quote>

In that case, the demo succeeded. I am absolutely sure one of the selling points at the C-Suite and a few of the more malicious advertiser and MBA crowd just beneath is that AI brings just enough pseudo-randomness while seeming useful to increase engagement and eyeballs.

This started getting noticeable with Google Search only a couple years after acquiring DoubleClick. Search results started getting progressively more kneecapped to keep people iteratively refining the search, while hopping over to another device with the same search string InCognito would yield better results with the original string.

Being directed by Google and similar search engines over to AI blog spam is not due to the AI blog spam fooling the algorithm. It's the algorithm has intentionally been reworked to produce poorer quality results.

I've run across guys in bars using them to peacock. Whatever stereotypes and conclusions you may draw about that particular cohort is likely accurate.

Oddly they left human resources off the list this time.

The issue with these lists is these are job titles and wildly inaccurate job descriptions are obvious "provided by HR experts". You can tell it is some "HR expert" fabricated job title and description every slide show article discussing jobs and any kind of top / bottom 10 list and this one matches the pattern.

Traditionally, technology identifying jobs replaceable with better technology placed HR at the top of the list. This goes back to punch card computing helping to eliminate payroll jobs. Since payroll and benefits are largely handled by Workday, Oracle, and ADP these days, next up on the list will be recruiters, who have traditionally been maliciously incompetent at best at their jobs until you interact with that top 5%. You know, the same cohort providing the job titles and descriptions above.

PageRank started breaking after Google acquired DoubleClick in 2007/8. If you survey search savvy people about when Google search results started getting noticeably worse, they'll usually point at 2010-2015 - basically as DoubleClick's influence internally took hold. Google search feeding absolute garbage AI generated content heavily was natural pivot as Google has their own vested interest in the success of current generative systems. This was a rising tides raise all ships decision from the top. And steering human users to artificial content allowed an extension of what "PageRank" had become since Google went full into being an advertising company first instead of a technology company -- having users curate what is good versus bad AI content. Steering consumers to ALL the AI trash give Google an incredible competitive advantage data set as it lets them identify which AI content generation strategies are successful.

As for Reddit, they are not going to be able to execute on a good search system. It's simply not in their culture. Actual Reddit employees exemplify inefficiency, sloth, and lawful evil, basically every negative cliche about DMV and similar government bureaucracy resting on their necessity. Reddit is a collection of fiefdoms wherein the stock price and value is based on the incredibly large arm of free labor they get from moderators. The moderators in turn tend to be petty tyrants, and ironically the exception to this rule is most consistently vendor run subs for products. Reddit is already aging out the way Facebook did several years back and Reddit has a more entertaining fight in the next 3-5 years -- when the moderated rise up and want democracy and freedom from the vassals and start calling for moderator elections.

I don't program professionally and never stuck with a language long enough to build genuine competency. I work in tech and do IT in the support direction and even write bugs for programmers to fix from time to time.

My issue with programming is as a skill it requires an excessive amount of practice time and effort to get to a point of doing anything reasonable beyond an address book. I compare every session to sit down and write code is too immature a process, it is like teaching a small child how to build a Lego set, but each time, you have to start over with teaching what Legos are and how they work.

When I last took a programming class around 2009, I was bewildered with how little the IDEs did as far as basic logic analysis. Couldn't catch simple things like loops that never end or obvious off-by-1 errors. The tools had effectively no real evolution at that time from my high school days over a decade earlier.

Started playing with AI code generation a few weeks ago. This is a usable tool for programming, and it's really much closer to where early advanced IDEs should have been much sooner. Not just the language models, but as far as simply describing the math and operations in psuedo-code I want to accomplish against an input and being able to frame up the code framework and explicit code itself.

If we compare this to other technological tool advancement, like say woodworking, we started with the Woodright's Shop (PBS show) from the 50's through the 80's with programming. The 90's through 20's failed to move to electrification and power tools for 30 years (despite knowing how electric motors work) instead dawdling on perfecting Japanese joinery instead of evolving. AI driven code is electrification and power tools.

From maybe a couple years after monetization, YouTube Search became an experiment. Unlike Google search of that time, it intentionally did not provide the best results for the requested content and information when it can get away with provide slightly incorrect content. The goal is to keep you on the site and engaged checking each video result and dropping comments. The reason why title and description generally vaguely work is it would be too easy for other search engines to legitimately index this data and for users to realize something was not just broken, but intentionally misleading in YT Search results.

This is similarly why it typically delivered some catchy unrelated left turns about 4-6 videos down the scroll as well as what I can best describe as "adversarial" content. By adversarial, I mean they provide or lead you to content the algorithm knows you will fundamentally negatively react due to any number of technical, theological, ethical, moral, or presentation and style grounds.

Google Search adopted some of these methods. The explosion of blogspam ~6 years ago and subsequent growth in Google Search results isn't an accident and it isn't that blogspam suddenly was indistinguishable from legitimate content. They started promoting the blogspam and ignoring the previous relevance / click through / success metrics.

I did this through college with a couple different Windows tablets. The original Surface Pro was a godsend and dropped at just the right time as the first year with an Atom / netbook architecture tablet was a bit limiting (Q550).

Made "do you have notes from..." -- "email? Here's a PDF." an almost amusing joke with friends. I even did homework long hand in OneNote and print out the pages to turn in.