neilv

KeePassXC might do what someone needs, and I really appreciate the work of its developers, but when advocating KeePassXC, I should also acknowledge that the "UX" is rough in parts:

1. It has lots of features and complexity, but doesn't always convey affordances for common use cases to the user.

2. Some of the UI design feels very incrementally developed (naturally), and the implementation a bit quirky in parts.

(For one of many examples, when I had to do something involving adding TOTP secrets, once I found where to add them, I had to be careful in which sequence I clicked things, or it would just discard the secret I already put in the right place. If I hadn't been watching carefully, I might not have noticed immediately that it did this, and not been able to restore the secret before it was lost.)

Of course, in an ideal world, one would like to do a great holistic rethinking of the UI design (while preserving the data model), but that's a ton of work.

When advocating it to a "technical" person (who is not scared of, say, a legacy IDE), I would say it might do everything they need. When advocating to an ordinary user, I would look at their use cases, and see what they are going to see, and how confusing or quirky that might be for them.

I intended for the entire sentence to be in terms of the thinking of top leadership.

And to gloss over how that improvement would actually happen. (Not knowing what they've currently done and want to do, but for example, guessing: probably in partnership with vendors, consultants, etc., iterative and experimental process and tools improvements, and involving a variety of approaches and refinements.)

Pre-AI version: Oops, you laid off the higher-salaried people without having them train their replacements, so bring them back, long enough to do that.

Now, that training[*] will be for both AI models and lower-salaried hires.

Perhaps a second mistake by those who thought they didn't need their most experienced people: Now they think they just need to train the AI better, and then new-grad "AI native" hires will be the most cost-effective way to operate/oversee the AI and do whatever it can't.

[*] edit: originally typed "replacement" when I meant to type "training"

That sounds like a brilliant idea.

I wonder whether they already thought of that, and are exempting from monitoring the roles most likely to generate "smoking gun" evidence.

> [...] you get a degree of viciousness, competitiveness, and steel-eyed execution

I think there's a lot of truth to that. (Aside: Many manage without the viciousness part. It's not their fault their parents lined them up with an internship and a research paper co-author in high school, and they're not jerks about it.)

Though the current generation of students didn't invent hyper-competitive. Before software engineering jobs (and startups) were high-income and high-status, you'd see that mentality among many people on track for Wall Street, for example.

Another example: Before CS was a go-to for the hyper-competitive, a mentor of mine actually switched from pre-med to CS, at an Ivy, because a percentage of pre-med students were outright sabotaging other students, and it turned him off of the field.

> that a lot of older Americans just aren't used to.

Though, there have been -- and hopefully will remain to be -- people doing it for the love of the field, who are not impressed.

Other than the genuine people being crowded out of admissions slots and fratbro interviews by Wall Street types...

If a Palo Alto helicopter-parented overachiever McDojo black belt tries to pick a fight... with a humble rope-belted person in Asia, who's studied martial arts for the love of it... the latter will chuckle good-naturedly, and help the Californian up off the ground.

I want Anthropic to be the good guys.

I think Anthropic should explain why they're doing this particular partnership.

It works with teams, when you have the authority to do something, but do a non-blocking check as a courtesy:

"Heads-up: I plan to delete the old scratch volume at Tue 14:00 ET, unless anyone objects before then. (It only contains the old Debian APT cache, and 974 copies of the same YouTube video.)"

> The era of the standard behavioral interview (“Tell me about a time…”) is over; those answers are easily scripted by live-assist tools. Instead, organizations should introduce dynamic friction: sudden constraints, changes in project scope, or prompts that require candidates to defend a counterintuitive tradeoff.

Isn't this already easily faked with an ordinary general-purpose consumer $20/month AI tool?

> Cultivate a culture of intellectual honesty over polished perfection.

This is one good idea I saw in the advertorial. Or, better yet, start with honesty at all.

But you have to understand and believe in it, or it will immediately be twisted into yet another gamed performative bit of interview theatre, like most other aspects that emerge from big-corporate mentality of herding worker drones.

(Perhaps the authors, coming from Meta and Microsoft, appreciate that reality.)

> When the earliest filters in your hiring process stop working, your organization begins to systematically select for candidates who are best at performing the hiring process, rather than those best equipped to do the job.

Isn't "performing the hiring process" theatre what Big Tech hiring has been demanding for ~20 years?

And gifted to most smaller companies? (Because people already knew Google frat-hazing student style interviews, from their own interview prep, to try to get into a FAANG, so they mimicked that when they went elsewhere?)

Wouldn't hurt to also use European DNS TLDs.

I didn't use this company, but there are some legitimate purposes for scraping.

For example, at a startup a few years ago, one of the many technological things we needed to do was to monitor marketplaces for suspected counterfeit and contract-violating gray market goods for ~100 brands. And we couldn't just ask for data feeds, because, well, the marketplaces make money off of all those sales. And the off-the-shelf third-party data solutions were useless crap quality, worse than your average vibe-coding. So I made a bespoke crawler that gently and accurately tracked the data we needed, including global geofencing. So gently, I never got a whiff of disapproval or countermeasures (like throttling, 403, nor data poisoning). We were putting insignificant load on the marketplaces, for the purpose of helping to make the market better for both consumers and legitimate businesses. It was like a single "secret shopper" unobtrusively walking around some parts of a store. (And I also made an iOS app that did something different for actual secret shoppers in physical stores, for legitimate supply chain traceability for customers' brands.) Personally, I love the marketplaces, and hate the counterfeits, and this was my version of PG's advice that startups should be a little bit naughty.

Two of the problems with the current AI scrapers, which are destroying servers, and inviting backlash:

1. The gold rush situation brings out many of the crappiest people in the world. And also many who aren't crappy might behave in a crappy manner. (The latter, maybe because they're just emulating what they see, or extrapolating from the ethical temperature of prior industry norms, like surveillance capitalism in everything.)

2. Many of these scrapers are shockingly bad at what they do, and grossly inefficient. Almost like they're just pounding the same unchanging resources to DoS the servers for competitors. Or to drive sites to a protection racket company that's set up so they can also monitor cleartext. Or (Occam's Razor) just plain bad at what they do, and the people who pay for the salaries and computer resources either don't know or don't care.

1. What happens when schools start hiring faculty with dissertations full of LLM output that's essentially the same thing as the plagiarism that this person is alleged to have done?

2. Well, at least it's an option in their back pocket, if they ever need to get out of a regretted contract or tenure commitment.

3. The article mentioned "weekly radio show". While presumably not the case here, the idea of academic as public public influencer suggests that rogue politicians could use allegations of academic plagiarism (due to LLM) to neutralize an academic who is saying something they don't like, when funding levers can't be used.

(Examples of topics for academic positions that some politicians might not like: climate change, race/ethic, sex/gender, geopolitic, safety scandal of campaign donor, planetary flatness.)

(I'm guessing this attack might work, since academics tend to profess more (dare I say it) conservative ideas about integrity than the overall population, and the universities value their own reputations highly. And academics have some rules and hierarchies, and they enforce rules when a transgression is called out in a way that can't be ignored. So, even if the general population goes full AI acceptance of people putting their name as author on things that aren't their own work, you'll have a lot of academics holding their own to a higher standard. And they already have a belief that plagiarism is bad. Fabrication is another bad. Popular LLMs are currently spewing both heavily.)

> In my opinion, the most useful next step is to contact Volkswagen support in a coordinated and technically precise way [...] Smartphone: Google Pixel Operating system: GrapheneOS

I strongly recommend saying that the operating system is one of "Android" (there are many variants), "Android (GrapheneOS)", or "GrapheneOS Android".

But if you say only "GrapheneOS", you are practically telling VW to respond that they do not support that operating system.

And not only crime. I think it's going to be a lot of people trying to hustle via mass-automated fake-human interaction -- whether it's sales, influencing/advertising, recruiting, dating, or whatever.

And they're being encouraged towards that: even tech companies with reputations to lose are already aggressively offering to take over as someone goes to write an email or text, and proactively "summarizing" the one-to-one human communications of others. But there will actually be demand for one-to-many fake one-on-one interactions, in the hustle culture, and doing it to strangers will seem no worse than what they've already done: corrupting pre-existing interpersonal relationships.

Just this week, I got my first AI-automated one, following a pattern of "liked your X, especially Y".

I hope the people doing things like that realize that's awful for humanity, and they stop doing that.

This is fun, and I wouldn't be surprised if someone at Google did something like this a couple decades ago, as a 20% Project.

Outside of Google, around that time, I used Google Earth for a 3D visualization tool for real flight data recorders, integrated into a larger browser-based system.

(Stack: Google Earth Plugin did the heaviest lifting, especially before there were better ways to render 3D in a browser window. The frontend used JS, HTML for instruments, and some kludges to work around some limitations of off-label use of Plugin. The backend was in Scheme, and retrieving and serving up cached data for this was one of the simplest of the things that the Scheme did in that large system. Aircraft 3D models were off-the-shelf, which I tweaked lightly in (IIRC) Google SketchUp.)

Hypertext to the rescue! Here's the lede sentence:

> In [Structured Query Language (SQL)](https://example.com/sql/), you can solve Unusual Complicated Problem with Super Advanced Thing.

That said, one time I had in mind a reader archetype, for whom I added an appendix of one basic concept, which ideally they'd already know, but likely didn't.

<https://docs.racket-lang.org/roomba/index.html#%28part._.Ass...>

I could've linked the some mentions of "association list" to a chapter of some textbook they'd never seen before-- and maybe they would read it, and maybe they would come back.

But instead, I decided to give a quick overview, in terms of an example relevant to what I was documenting, and leave them with a code pattern they could use, to get on with programming a robot like they came to my document to do.

(Though I wish I'd put an accessible showing-off demo example near the beginning of the document. After the intro, it reads a little too much like the glorified inline API docs that it is.)

> Both marks come from Scheme, where [...]

Reminds me of an email I wish I still had.

Circa 2000, I wrote that I was leaning towards moving to Scheme, for more rapid R&D work than I could do in Java.

Some nice-sounding person I didn't know emailed me from Japan, to mention a language I hadn't heard of, called Ruby.

I don't know whether the person was Yukihiro Matsumoto himself, but it's a small world.

> There are other ways to get rich than by starting startups. Some of those do require you to exploit people. But startups are the most common way to become really rich, and if you want to start a successful startup, the key is not exploitation but empathy. What do users really want?

To have illegal hotels that then help keep a generation out of home ownership?

To have an exchange for cryptographic tokens that are used almost entirely for financial scams and organized crime payment infrastructure?

To have an online forum that made so many long-time contributors who built the content and appeal feel so betrayed, that often the top solution to a posted problem (you find in search) has been deleted in protest?

To have a non-profit spun off, ostensibly for the benefit of humanity, and attract talent and funding that way, then coup and rug pulled?

Other big successes?

For players who are new to the game, there should be a 1/4 chance you go to bed proud of an honest day's work with your hands, and wake up the next morning having strained a muscle you didn't even know you had, and you can't chop wood for the next couple weeks.

If the article's description evoked applications like old missile guidance system methods based on geographic features... this Wikipedia screenshot's imagery looks like it would also be good for precision drone attacks against urban small civilian structures and select individuals within/around them:

https://en.wikipedia.org/wiki/Pok%C3%A9mon_Go#/media/File:Po...

Wartime propaganda poster: "Loose Surveillance Capitalism Children's Game Apps Sink Your Own Darn City, to an AI autonomous drone swarm assault that surgically neutralizes whatever the worst people want to neutralize".

AI, please rework that into a catchier slogan, and render it as a printable US WW2 OPSEC poster style PDF, but without storing my prompt and-- Hey, what's that buzzing soun--

Relevant quote:

> I am sure it is not perfect (I only spent an hour working with the results), but a software engineer would iron out the remaining potential bugs that I could not find quickly [...]

People have said things like this many times in the past, and, in the past (perhaps not now), it's always been a misunderstanding of what is good and bad, what's difficult and easy.

For example, someone would draw a UI in a GUI painter that generates code (or a resource file), and a manager would see it and think the majority of the work towards the product is done. (Incidentally, then there seemed to be a reaction, towards making your UI mockups look abstract or otherwise different from runnable code, helping the nontechical to understand that this isn't 90% of the finished product.)

Or a student intern hacks out a homework-grade demo, and a manager who understands neither software engineering nor product domain says "we just need some engineers to polish it up for production", and thinks the student is a star and why can't their engineers be as brilliant and productive. (I might have once been that energetic intern, who was happy for the encouragement, but then learned more, and saw it was a thing.)

This common misunderstanding was sometimes self-correcting -- when trying to ship became a disaster of misery and regretted-attrition, or the product was poorly received by the market because it wasn't thought through nor implemented well, or building subsequent functionality atop it was a nightmare. (But adverse effects of bad approaches is one of the reasons for management and ICs to job-hop, before the unwanted effects affect them personally.)

What might be different now is that some of these AI tools are outputting better-engineered work than some software engineers, and much faster.

At the back of my mind, I'm wondering how the really great software engineers will continue to stand out, as the discipline is being devalued in the minds of most leadership, and anyone can prompt an AI to generate something that superficially appears to them like what they assume a great software engineer would produce. (Even if the great engineer would do much better quality of implementation, have innovative ideas that ML from open source code would not, and maybe arrive at better product concepts as they worked through the problems.)

This was the secret sauce of my best startup idea: something that once existed, but had been forgotten.

(Because, I believe, either the flood of people into the market space never knew it, or it wasn't the dominant model for exploitation of the user base.)

Import-only is the lazy way. :) If you want, GnuCash also has features to support manually entering each transaction as it occurs.

Then GnuCash has features to match up financial institutions data exports against your transaction splits, see the discrepancies, and then reconcile against their PDF statements.

It's a way of life, but you always know what money you have, where. Although that sounds like something for wealthy people, I'd say it's actually more important the less money you have.

OK, just please be careful how you frame what you're asking for.

For software development use of Claude, I'd be happy if the `claude` CLI executable does everything I need, within the Linux KVM VM sandboxes I create for the work, without a desktop client. The cleaner and more trustworthy, the better.

Also, for random interactive use of Claude for asking questions, I use it from my host desktop, sandboxed within the Web browser, and I want that to be well-supported. Someone marketing/product person at an AI company will naturally want to dark-pattern push people towards a proprietary desktop client, but that's one corner of abuse potential that we can still keep in check.

For agentic automation of my host desktop things and the things they have access to... no, thank you, the state of the art is not ready for that.

What I'm wondering is -- although an item would be found to be legally considered public domain in all applicable jurisdictions when examined by law clerk -- what are the practical rules for passing copyright clearance checks/challenges by Amazon, Kobo, etc. for public domain art.

You don't want a book's carefully choreographed launch sabotaged by various companies' Kafkaesque bureaucratic processes, while you're trying to finesse the algorithm and the publicity you've scheduled.

(I should've said upfront that I know a bit about public domain and copyright law as a layperson. And I also know that I can't necessarily point a CSR to law surveys if something gets flagged. I need to know the "do X and avoid Y" that is proven to actually work smoothly with all these companies' ideas of "copyright clearance", so that I hopefully never have to talk to a CSR.)

Exactly. The site mentions that and some other problems, and just disclaimers their liability. One problem I didn't see mentioned directly is that some venues have additional bureaucratic rules.

For example, I saw a writer-artist complain that they couldn't show the big ebook store proof that they'd licensed the art for their cover, since they painted it themself, and there was no documentation licensing it to them.

So one could use practical advice from someone who's figured out the rules, and has the battle scars, like: "With public domain art, just make sure you do X and avoid Y; otherwise, there's a 50% chance that a random copyright check the first week will result in your book temporarily being shadowbanned, and you won't get the crucial early sales while the algorithm is trialing exposure of your book in searches and categories, and then the algorithm will pretty much never show your book to anyone ever again."

Does anyone know how easily you can do "copyright clearance" for these supposedly public domain images?

For example, the page for the first image I clicked on said:

    Date
    1833

    Underlying Rights
    Public Domain Worldwide

    Digital Rights
    No Additional Rights

    * Source states “no known restrictions”
    * We offer this info as guidance only

with a link to: https://pdimagearchive.org/reusing-images/

If, for example, you design the cover of a self-published book around such an image, is Amazon KDP going to reject it, because they don't accept that screenshot as sufficient proof of rights?

There are plenty reasons to be critical of Microsoft's AI strategy and tactics (and especially of many other things MS has done), but the linked article seems to be targeted at gamer, rather than at people who care about non-gaming tech industry or public policy.

What seemed a bit more relevant was one of the linked 404 articles, concerning CEO's denial and attempts to dismiss the document, before the document was revealed to be co-authored by the head of the strategic project. But even that article sounds more like social media or political mud-slinging in style, rather than journalism:

> In attempting to distance himself from his own company’s executives and strategy documents, Nadella has revealed that he either does not know how to read or does not know what is happening with some of the company’s highest-profile products.

But what I didn't see what a smoking gun that they were truly looking for addictive (like, say, Facebook/Meta has been caught engineering) rather than something they could've described as essential if they weren't using amped-up business bro language. So rage-baiting over the word "addictive" seems to be missing better questions.

Ideas for what individuals can do about this and related awful tech:

1. Ask your local and state governments to completely ban "stalkerware" and "Big Tech surveillanceware" (like will use this and other face recognition), as well as ban using hidden cameras (including in these glasses) to photograph/video people.

2. Tell everyone, before their buy the glasses, what a "glasshole" is.

3. Social negative feedback to people who wear these. Tell your friends if they're being inconsiderate. Tell coworkers it's inappropriate in the workplace. Frown at strangers who do it. Tell apparent creepers to stop, and/or consider calling the police.

4. Social negative feedback to people who work at the companies pushing this tech. There's plenty of tech talent on the job market. Why consider someone who continued to work for one of the companies, in some cases after years of sociopathic abuses of society?

5. Be skeptical of influencers and astroturfing shills promoting the products.

Other ideas?