• 2 Posts
  • 566 Comments
Joined 1 year ago
cake
Cake day: June 16th, 2023

  • Scanning texts is OCR and has never needed modern LLMs integrated to achieve amazing results.

    Automated tagging gets closer, but there is a metric shit ton that can be done in that regard using incredibly simple tools that don’t use an egregious amount of energy or hallucinate.

    There is no way in hell that they aren’t already doing these things. The best use cases for LLMs for NARA are edge cases of things mostly covered by existing tech.

    And you and I both know this is going to give Google exclusive access to National Archive data. New training data that isn’t tainted by potentially being LLM output is an insanely valuable commodity now that the hype is dying down and algorithmic advances are slowing.







  • Um they are, and have been for almost 20 years, since the Wii. Or the N64 depending on how you look at it.

    What did you think Virtual Console was? How about the NES and SNES mini? What about the “Nintendo Game Pass” or whatever they’re calling it?

    Animal Crossing’s original Japan release had NES games in it, and so did the GC rerelease/psuedosequel we got internationally too.


    Even better: During the Wii era, the Wiis at the Nintendo Store in New York City ran official Nintendo made software to load games off a connected hard drive, so you could play multiple of their new releases without workers having to switch discs.


    It has always been about attempts to prevent piracy and keep control over how people access their games for Nintendo, and they are roughly 10 years behind the curve on modern tech trends.

    Either stop supporting them or get used to it.






  • This buries the lede quite a bit.

    Mullenweg effectively runs both the non-profit organization Wordpress.org and is the CEO of Automattic, a for profit conpany that sells support for Wordpress (and a direct competitor to WPEngine).

    A large part of Wordpress functionality is kept behind an Automattic plugin that forces any Wordpress site using it to collect telemetry/data for Automattic.

    The update servers for Wordpress plugins are hardcoded to use Automattic’s servers, and this is not configurable or changable unless you modify the Wordpress source code itself.

    With Mullenweg’s position over both the non-profit org and Automattic, he has direct control over these choices. If he’s doing this for the sake of open source, why is he gating things that should be core functionality behind a data collection scheme? If there are problems with load on the update servers, why has no effort been made to allow the community to host update servers themselves that check update hashes against Automattic? That would significantly reduce the load on the for-profit resources (that you called APIs). At the very least, the setting needs to be something exposed to the user and configurable without modifying the source code. Otherwise he’s complaining about a problem he has created.

    It’s also worth noting that at no point has Mullenweg tried to set up any sort of free vs paid tier of access to his update servers. This is a specifically targeted campaign. He has also not publically provided evidence of the increased load by WPEngine despite publically shooting off about a ton of other things that would be best saved for the courtroom.

    Mullenweg has also publicly stated some very questionable things about how the resources of the non-profit and his for-profit are intermingled, which may have some legal repurcussions. But that’s more of a footnote.


    Wordpress’s license makes explicit exception to copyright to allow anyone to use “WordPress” or WP.

    The initial reasoning (and I believe the lawsuit) for Mullenweg’s attempt to claim 8% of all WPEngine profit, is explicitly based on the claim that they are breaching copyright due to their use of WP.

    So while I agree that lack of upstream contribution and the amount of load on the upgrade servers are important and valid reasons to try and seek some contribution, that is not the angle he took to start this.


    At one point during all of this, he switched off the WordPress plugin update servers for all users with no warning.

    Now he’s done a direct hostile takeover of his competitor’s plugin. Of the two security issues, WPEngine disclosed both of them themselves and had already fixed one. There was no evidence that they were going to stop and not fix the other, and the issue is of questionable severity. The main change Automattic did to the plugin was to remove the code that checked for an upgraded/upsold license, effectively cracking the plugin to offer paid features for free.

    With the long history of WordPress, I find it incredibly hard to believe that there are not a considerable number of other plugins containing upsells, so the implication that those somehow are in violation of terms is weak.


    In my opinion, we have someone in the perfect position to make changes to ensure the upgrade server load (the only quantifiable reason for all this mess) never would have been able to be a problem in the first place. He has singled out the largest competitor to his own for-profit company and targeted them specifically instead of announcing blanket changes that would apply to anyone causing their level of load on his systems. He has taken incredibly poorly thought out and reactionary steps intended to spank his competitor that have had far larger negative effects for the rest of his users and customers. He has and continues to make very piblic statements that any sane lawyer would tell him to keep his fucking mouth shut about. Now he has once again singled out his largest competitor, taken one of their paid products, and modified it to be free rather than creating his own implementation with the problems fixed and no upsells.

    Matt Mullenweg has not done anything explicitly evil, wrong, or super obviously illegal. But he’s doing a hell of a lot of very concerning and questionable things when he had every opportunity to prevent any of this from ever being a problem in the first place.

    I have no love for WPEngine, but Matt isn’t a saint and is ridiculously mismanaging all of this.


  • Holy crap he is just continuing to build the case against him. His own for-profit automattic’s plugins have built in upsells and add additional data harvesting code that you can’t opt out of while you use them.

    He just keeps treating the non-profit .org stuff and his Automattic for-profit as interchangable.

    It’s time for the community to find a solution for distributed update servers that at most only rely on Mullenweg for hash checking to prevent tanpering. This is blatantly just a vendetta now.





  • It was a lot more than a single disagreement.

    Team members (both former and current) have made some very concerning comments over the years about Linus’s private and public behavior, and have repeatedly raised concerns over the breakneck release schedule negatively impacting quality.

    Whenever they get information wrong, or make clear mistakes on things that then effect how they review a product, they hide behind the tight release schedule of their videos, which is something entirely under their own control.

    Linus himself has shown shockingly poor judgement and communication skills multiple times as he has publicly ran his mouth off and been overall an unprofessional jackass when people have pointed out his mistakes in the past.

    They handle redactions and corrections incredibly inconsistently, if at all, and only when there has been significant backlash.

    The whole “getting a hand made protoype waterblock for free, failing to do the bare minimum to ensure they installed it properly, blaming their botched install on the maker, choosing not to contact the maker to discuss anything, releasing a scathingly negative review to most likely tank the fledgling startup who sent it to them with install instructions they ignored, being intensely belligerent with everyone telling them they installed the thing wrong, claiming they lost the prototype when the maker requested it back, then selling the prototype as a reward for a charity auction” that is only the biggest, most publicly called out fuck up recently, in an ever growing history of this sort of bullshit that goes back for years.

    Beyond all that, if you have ever watched a single video of his about anything you already know about, it becomes immediately obvious just how much he’s and his team are learning the bare minimum about things and then just skating by on production value and speaking as though they have authoritative knowledge.

    None of this is inherently damning, except for just how much money his organization makes, and that they project an image of being trustworthy authoritative knowledge sources when they really aren’t. Their entire business is based off the concept that they offer trustworthy and reliable information, but they explicitly don’t often enough that it is an inescapable problem. When called out on it Linus very publicly loses his shit and blames it on everything but their own failure to do proper research and take enough time to ensure quality control.

    This is a marked downward trend in their content and public behavior that has been going on for more than eight years.


    Lastly, I never said he was a garbage person. That was the comment I replied to.