0xfffafaCrash

The reply I received a few days later did me the favour of putting the violation on the record. Their position, in their own words, was that "in order to receive marketing / offers, it is a condition to be a member of the customer club." That one sentence is the whole case. They had taken a right I am entitled to exercise for free and turned it into the price of admission.

I don’t understand… it would be one thing if it said “receiving marketing/offers is a condition of being a member of the customer club” but that’s not what is being stated above… rather that being a member of the club is required to receive marketing — perhaps something has been misworded or lost in translation?

Just want to note that OP was in Australia, not Austria, but the broader point stands that it can be helpful to ask foreigners what they need rather than assuming your norms will translate over.

And yet, this was still a very generous gift and perhaps even greater value in sharing the experience and starting these discussions

Is the uuid generated in the frontend or backend? If frontend, I’d wager the likeliest explanation is that the client code or request was messed with to inject a previously known uuid rather than an entropy issue.

I think it’s fair to say that work on the experimental-strip-types option in Node was inspired/energized by a desire to try to catch up with the DX improvements found in Deno for Typescript-first development that is now the norm.

I’m not familiar with the author but something about this post just seems mean-spirited and petty.

Deno might not succeed as a project, especially with strong competition from Bun as an alternative to Node, but I would say that Deno has been more a force for bettering the ecosystem than not.

Many of those at Deno, including Ryan as well as some of those who have apparently left or been let go have been major contributors to the web development ecosystem. Thank you all for your work — we’re better off for your contributions.

moreover h* is just broken whenever dealing with more dynamic content — it simply can’t reasonably be made to work according to accessibility recommendations — and the accessibility guidelines around never skipping a level themselves are ridiculous given the practical reality that dynamic content exists and we have only h1, h2, etc. to work with — the readers and specs are what need to adapt here, not the entire internet

there should really be one header tag and its level should be based on some nesting depth

and don’t get me started on the maintainability mess that is z-index… better we have a system to centrally maintain an ordering list than a distributed one which only works reasonably consistently if you already know everything in the whole system

There’s already more human produced content in the world than anyone could ever hope to consume, we don’t need more from AI.

Even if you think the harms of AI/machine generated content outweigh the good, this is not a winning argument.

People don’t just consume arbitrary content for the sake of consuming any existing content. That’s rarely the point of it. People look for all kinds of things that don’t exist yet — quite a lot of it referring to things that are only now known or relevant in the given moment or to the given niche audience requesting it. Much of it could likely never exist if it weren’t possible to produce it on demand and which would not be valuable if you had to wait for a human to make it.

As the article says it’s used for age verification

Strange. I’ve only had really poor experiences with Amazon and Amazon deliveries in Sweden and really good ones in the US. In Sweden the delivery network seems full of other parties that frequently fail to show up at the last minute multiple times in a row. The translations are humorously bad and the selection is small.

thats a feautre! prevnets the pregmaing drunsk from getitng double drukn or having the drunknness cancle out or watever

If a nation wants to block certain content, let the nation deal with it by getting their own ISPs to block and make sure the citizen's anger gets correctly placed on their government and not the site operators.

I don’t really understand comments like these. Even if you’re exactly right about how it should work, how would you make this happen in the world we live in? Neither the tech community nor ISPs nor cloud companies decide these things. Just because a matter affects us doesn’t mean we have much of a voice in it especially if it’s legal.

Laws about tech are decided by (idiot) politicians/parties/governments and the consequences are enforced by massive fines, imprisonment, etc. by law enforcement and selective (and often politically motivated) prosecution. In some of the worst places the consequences could include death.

Afghanistan just lost access to the internet almost entirely. China and North Korea are famous for their firewalls. Much of Asia has internet blackouts whenever there are large scale protests. The western world’s government has more legal jurisdiction/economic influence on the companies that run these things and are increasingly leveraging that for their desired censorship.

If the answer to this is democratic influence, the populations of many countries don’t really have that, the majorities in countries that do have it certainly doesn’t know or care about these things and wouldn’t vote for the pro-censorship politicians in the first place if they’d then vote to cut off their nation’s access to uncensored internet while preserving the uncensored variants, and even if the majority ever did care to get the system to work in this way there’s a global trend away from having their opinions on such things matter anyway.

Might be a cultural thing

I’ve both been rewarded for dissent from leadership throughout my career and had greater respect for and advocated more strongly for those willing to stick out their necks and disagree earnestly and productively when in leadership positions.

Dissent isn’t the same thing as sabotage. There’s healthy conflict and open disagreement which helps illuminate risks and gaps and uncover opportunities in productive ways and then there’s just stirring the pot or trying to tear things down without bringing alternative proposals to the conversation — being unwilling to contribute in positive ways if you don’t always get your way.

The latter kills the ability for the team to work well while the former is key to allowing colleagues bring insights and value to the team

https://danfabulich.medium.com/webassembly-wont-get-direct-d...

For one thing, language…

Have you considered that some people listen to text rather than read it? Do you think screen readers, non-native English speakers, or anyone else listening to these are able to reasonably differentiate between “Open Source” and “open source”?

Most readers won’t notice the difference or disregard it as meaningless and if you have to explain to everyone your position on a matter of communication it’s likely that you are the one who needs improvement with respect to your communication, not the rest of the world.

I really want JS/TS to adopt the pipeline operator which has been in a coma in stage 2 after eternal bikeshedding at TC-39

https://github.com/tc39/proposal-pipeline-operator

It would make it possible to have far more code written in the way you’d want to write it

While not exactly recent, I think most people have Dual_EC_DRBG in mind these days when it comes to NIST and backdoors

https://en.wikipedia.org/wiki/Dual_EC_DRBG

On a tangent, I don’t understand how Apple’s App Store allows the Google Photos app to insist on being given access to all photos on your device rather than selected photos to even work (even just to view photos shared elsewhere). Seems like an obvious violation of privacy and terms around not requiring unnecessary permissions. I think the app should be banned until they provide an option to gracefully respect users’ privacy settings rather than holding unrelated features hostage if they don’t get to spy on every photo on your device.

not just UTF-16 but potentially malformed UTF-16 (supporting invalid surrogate pairs or surrogate halves and with js string functions computing things like lengths independently of UTF-16 characters)

This is also widely known as WTF-16 (seriously, look it up!)

If you read the judgement it basically says that because the FBI eventually returned the money, she no longer has a live interest as an individual.

They also denied that she’s part of some well defined class for a class action. It’s not like they are saying there’s some other court she can go to.

US courts are generally infamous for denying justice whenever they see fit using technicalities like “standing” and other procedural grounds.

Environmental law violations, illegal surveillance programs, civil asset forfeiture like here, and constitutional violations are quite often practically impossible to get courts to address, especially if parts of the US government are the defendant.

I wouldn’t say signals is coming from React. If anything React is the latest to the party and being dragged there reluctantly by where the tech community is going. They’ve mostly been resisting the movement towards signals because it undermines a lot of the arguments they have made for their virtual dom, synthetic event system, and rerender all the things dogma. Solidjs is much much more in the signals camp. Angular, Vue, Svelte, etc also recognize the need for this sort of reactivity.

React could be argued to have abandoned the fight for being the best client side framework technically (though they have dominance pragmatically). They are really all focused on Vercel’s Nextjs/SSR/SSG/RSC stuff in recent years

I haven’t ever seen issues scaling with jotai even on very large apps working with extremely complex data structures.

I’ve seen far larger messes created in redux due to tight coupling of things across large apps to a global store and the inability to work with things like Maps and Sets or other values that are not trivially JSON serializable.

In the other direction I have seen messes with observable-based state management systems where things become far more complex and too far abstracted (how often do you really care about anything other than the latest value and maybe the previous one?) or with proxy based systems that have too much fragile magic (valtio, mobx) or require wrapping your components and going all in on oop (mobx)

To me signals hit the magic spot of reactive without being overkill and keep code testable in smaller units while retaining performance benefits of surgical updates

I like xstate in theory — it’s a good way to think about complex state transitions — but in at least half of cases in practice where you aren’t interested in a derived value, someone is storing a value/ getting the latest value or toggling a boolean and it’s just such overkill for that. The reducer pattern itself doesn’t meaningfully show up much for similar reasons. The other common cases are with fetching states (idle, loading, refetching, success, error) but you often end up wanting things like cache management or maybe even optimistic updates so eventually tanstack query covers that ground better than rolling your own.

React doesn’t really concern itself with state management. Of course it has context, state, and props but the mental model for it predates focus on fine grained reactivity in frontends — useSyncExternalStore helps enable others to fill that void in v17+. Fine grained reactivity was notably also missed in the development of web components — only now is there a tc39 proposal for signals (in its early stages).

Jotai, mentioned briefly in the article, may not be built in but is as intuitive as signals get and isn’t even tied to React as of later versions.

I’ve very rarely met a state management problem in clientside state management where neither tanstack query (for io related state) nor jotai (for everything else) are the best answer technically speaking. The rare exceptions are usually best served by xstate if you want to model things with FSMs or with zustand if you actually need a reducer pattern. There’s a tiny niche where redux makes sense (you want to log all state transitions or use rewind or are heavily leaning on its devtools) but it was the first to get popular and retains relevance due to the fact that everyone has used it.

You can go a long way with useContext and useReducer/useState but few would opt for alternatives if jotai came batteries included with react.

A splat is basically a point in a point cloud. But instead of a gaussian splat being just a point it isn’t infinitesimally small but like a 3d gaussian where the point represent the mean. It also has color and opacity. You can also stretch it like an ellipsoid instead of having it have perfect radial symmetry.

Strongly disagree here. While I haven’t published e.g. particle physics work, I have authored/coauthored a number of peer-reviewed papers in other topics generally considered hard science (and published in ”high impact factor” scientific journals). This article and series is just as accurate about how ”Science 2” works outside of ML in my experiences. Branding and marketing is a very major factor in everything from grant funding to research publishing in academia.

Here’s some relevant discussion about some of the footguns:

https://github.com/typescript-eslint/typescript-eslint/issue...

https://github.com/tc39/proposal-explicit-resource-managemen...

I imagine there will eventually be lint rules for this somewhere and many of those using such a modern feature are likely to be using static analysis via eslint to help mitigate the risks here, but until it’s more established and understood and lint rules are fleshed out and widely adopted, there is risk here for sure.

https://github.com/typescript-eslint/typescript-eslint/issue...

To me it seems a bit like popular lint libraries just going ahead and adding the rule would make a big difference here

I’ve also had my writing misidentified as being LLM-produced on multiple occasions in the last month. Personally, I don’t really care if some writing is generated by AI if the contents contain solid arguments and reasoning, but when you haven’t used generative AI in the production of something it’s a weird claim to respond to.

Before GPT3 existed, I often received positive feedback about my writing and now it’s quite the opposite.

I’m not sure whether these accusations of AI generation are from genuine belief (and overconfidence) or some bizarre ploy for standing/internet points. Usually these claims of detecting AI generation get bolstered by others who also claim to be more observant than the average person. You can know they’re wrong in cases where you wrote something yourself but it’s not really provable.

I had the same thought. Also with square roots hiding 2s behind notation, etc. The whole project isn’t really very coherent without specifying what specific operators you can use (and how many times).

Doctors are at the very top of my list of people who don’t care. Not necessarily that they got into the field not wanting to care, but that in practice they quickly get to the point of caring largely about getting through their day — maybe a few select patients stir them out of their bizarrely intense waking slumber where they go from patient to patient and immediately prescribe nearly the first thing that comes to mind for nearly the first diagnosis that comes to mind. Given the volumes of patients they are expected to churn through though it’s not surprising that they become desensitized and divorced from the ramifications of shoddy work with minimal research — for many (especially nonspecialists) it’s effectively impossible to do thoughtful work for every patient. I think overwork desensitizes many/most and few actually have the time or energy to do more research or think deeply about an individual patient, but ultimately decisions which consume minimal resources from them drastically affect the lives of patients.

Healthcare professionals know this to be true. This is why when their own loved ones are the patients they have such a strong tendency to become very actively involved —- it’s not necessarily that the person attending to their loved one is incompetent, but chances are that their loved ones will similarly be just another face that occupies another physician’s mind for a few minutes.

Artificially high barriers of entry in the field may lead to massive compensations but also to a huge ratio of patients to physicians — this takes a toll.

I agree with the issue of ”is the test dataset leaking into the training dataset” being an issue with interpreting LLM capabilities in novel contexts, but not sure I follow what you mean on the quantum computing front.

My understanding is that many problems have solutions that are easier to verify than to solve using classical computing. e.g. prime factorization