apparently this git cheat sheet PDF doesn't render properly in Chrome? https://wizardzines.com/git-cheat-sheet.pdf (here is what the 4th column looks like in Chrome vs what it is actually supposed to look like)
so I guess that's my side quest for today instead of whatever else I wanted to be doing. PDFs are weird.
(edit: it was a relatively short side quest, I think it's fixed now)
FedSoc justices disgraced themselves and their fake pet theories in the oral argument on presidential immunity.
Remember when they were “minimalists”? Oh, but now they embarked on a long policy peregrination so as to make what Gorsuch called a “ruling for the ages.” (He actually said that.) Behind the stunning pomposity, it’s miles from “minimalist.”
Remember when they were “constitutionalists”? The constitution says they’ve got to stick to the “case or controversy” before them, and yet they went on their wild hypothetical wanderings. Some “constitutionalists.”
Remember when they were “originalists”? Follow the text, never mind the outcomes?
All gone, in hand-waving about what various rulings might portend, and what effects they could have.
Remember their recent switch (Dobbs, Bruen) to “history and tradition”? The “history” is that no president but Nixon and Trump committed crimes; none sought immunity.
The “tradition” is presidents for centuries got along just
@skykiss This is so ahistorical it borders on conspiracy theory.
There is a grand total of one currently-sitting Judge who even heard Bush v. Gore. There's no "they" here. It's an entirely different Court.
The United States has never delayed seating a President. They had only a few weeks until the inauguration deadline. Failure to decide Bush v. Gore could have jeopardized that timeline, and that shit matters. You want to see a democracy go to hell? Delay sitting its new Executive by one day and see how long until people start crying "It's a dictatorship!" In contrast, we're still months out from the Presidential elections. Speed is not of the essence right now. Thoroughness is.
Microsoft Recall in Windows 11: in what way can this be POSSIBLY compliant with the requirements of GDPR?
(Same goes for Office365 requiring autosave to stash files in OneDrive, and Outlook slurping all your emails into Microsoft's cloud and using them for AI training.)
@cstross Does the Recall data ever leave your local machine? I was under the impression that all the search etc. is implemented local-side, so the GDPR ought not apply.
Google's sorry excuse for a "Web filter" demonstrates how much they plan to stuff AI down our throats and arrogantly starve for views the very sites that their AI is getting its info from. DISGUSTING. I'm starting to feel about #Google Search the way I now feel about #Twitter / X. Gone into the dumpster.
@lauren Isn't this a good thing? People who want to bypass the generative content can just go to the "Web" tab now. Granted, I'd like a way to shortcut into it like images.google.com taking me straight into Image search, but I imagine writing a Chrome extension to wallpaper that shouldn't be too hard if clicking the "web" button becomes a burden.
@lauren But what does that mean? The algorithm itself (guessing from usage patterns what humans probably mean when they give a query) is "AI crap." Onebox has been around forever and it's "AI crap." From Google's standpoint, this is just an evolution on a tech stack they've been building for decades now.
I agree with you though that it'd be nice-to-have to be able to sticky that "Web" option---really, any of those options. Or get to them by a keyword ("in:flights") for much more convenience on the input line.
@joeo10@lauren Sure, but accusing Google of pulling some kind of shenanigans because they added a UI feature to focus your query (over a default that they believe is most useful to the most people) seems... Odd?
Google's mission statement is "To organize the world's information and make it universally accessible and useful." The mission statement has never been "To send people to websites." Websites were always the means because that's where the information was... Is it anymore? If Google can give you your result faster without navigating you away, it's in their mission statement to do so.
@lauren Sorry, I think you're operating at the tip of a chain of reasoning that I don't see right now.
That's okay. I'll think on it and maybe I'll come around to your view. Your passion on this suggests to me you're perceiving a danger I'm not seeing here.
@mrcompletely I think "negatives are immense and varied" would need to be unboxed to really confirm there are concerns here. The "often poorly" is a much larger concern---if the tool doesn't work, that's a problem.
@mrcompletely I think the biggest concern is inaccuracy---if the tool doesn't work, it doesn't work. And it is, I agree, hard to predict what the "information economy"(?) will do in response to this change... But people have predicted the death of the web before, so I'm default-skeptical any time someone says "This will end people's incentives to put information online" as if there aren't people doing that because they believe in the core mission of making everything convenient (how much do Wikipedia editors get paid?).
"IP theft" is... I think there's a concern about copyright infringement, to be sure, and if Google turns out to be infringing copyright may they pay through the nose. But unfortunately, one rarely finds out whether something is infringement without trying it. Copyright is not common-sense obvious like other areas of law (like theft or murder) are. It hasn't been considered "IP theft" for Google to build a search index, so it's not a priori obvious that it's IP theft if they take that exact same index and train an AI on it.
Significant environmental concerns
I think we can't make that assumption without something to compare to. I used to work at Google; I have a rough number in my head for the number of computers involved in every search query, and I think one would have to consider the number of glasses of water they drink before asserting AI drinks more or fewer.
abuses by bad actors
Always, but that's already Google's problem. People put false information (and intentionally-misleadingly-constructed websites) online every day.
New development policy: code generated by a large language model or similar technology (e.g. ChatGPT, GitHub Copilot) is presumed to be tainted (i.e. of unclear copyright, not fitting NetBSD's licensing goals) and cannot be committed to NetBSD.
@netbsd Figuring out code is tainted by use of copyrighted code from another source is as straightforward as string-matching, maybe some fuzzy matching.
How would one identify code generated with the assistance of an LLM if the contributor doesn't admit to doing that?
At #Google I/O today, the firm is publicizing an array of new projects. Some of them seem flashy and relatively useless, others seem like they could be very worthwhile. How many of either category will still exist five years from now is of course a crucial question given Google's history.
But Google I/O is merely the gloss, in many respects what has become the so-called "lipstick on the pig". Because Google executives have permitted their race for the golden and in many respects false prize of "Artificial Intelligence" to cloud their vision, and to permit an increasing number of basic services that billions of Google users depend on every day to, in effect, rot away.
The collapse of Google Search, once a global technological wonder, has been profound. Often incorrect or even inane generative AI responses now often supersede links to the very sites from which Google is obtaining the raw material for their AI systems (usually without any form of compensation, while driving down user click-throughs).
A similar decline is obvious in various other core Google services.
Of great continuing concern to me is the very foundation of how virtually all Google users access most Google services -- Google accounts themselves. I continue to be flooded by persons who have problems with their Google accounts through no fault of their own, including lockouts and permanently lost crucial personal data, with Google's automated systems providing them with no resolutions, only horrific frustration. Google's frankly poorly conceived and rushed implementation of passkeys -- and the pushing of users to them who typically do not understand them and have more problems as a result -- is making matters even worse. What good are fancy new services when your Google account needed to use them may lock you out at any time with effectively no genuine ability to appeal?
Some groups of Google users -- such as seniors and other users with special needs who may not be technologically sophisticated -- are especially affected by these sorts of problems and suffer mightily as a result. I don't think Google actually "hates" these users -- I think Google simply does not want to make the minimal efforts required to help them, basically treating them with much the same disdain as you might flick a bug off your shirt.
There is so much that would be relatively simple for Google to do that would vastly improve the user experience for these users and others -- but Google seems to only care about the majority, and if you're in the minority, well, if you swing slowly in the wind locked out of your account, too bad for you. Google's got other fish to fry to keep the profit centers humming.
I could go on, but you get the gist. I don't hate Google. I still have major respect for the firm and especially for Googlers (Google employees) in general. But I am enormously disappointed with the direction executives are now taking the firm, and this seems to be getting worse at an accelerating rate.
@lauren I worked with a colleague at Google who's byphrase was "Don't optimize for the unusual case." Every time I hear about someone losing their account permanently and being lost in the labyrinthine halls of the recovery process that assumes a particular type of spherical point-like user, I think about the fact that he still works there and his philosophy is closer to the centroid of Google's philosophy on software than mine.
... anyway, a friend's parent got his Yahoo account compromised a little while ago, and he was able to reclaim it with one phone call. Someone at Yahoo checked the recent access logs, confirmed that it was obvious that someone had radically changed the access pattern of the account immediately after a password change, and just reset the password back. No muss no fuss.
So now I tell people to get a Yahoo account if they can.
@cfiesler I feel like this guy is engaging in a one-man (well, one-man-and-the-whole-fucking-company-he-owns) crusade to prove right the libertarian-leaning idea "Internet services are infrastructure and should be politically-agnostic."
Which is a shame, because things could be so much better if we stopped deluding ourselves that we have to keep handing people weapons to hurt each other because it's only fair if everyone has them.