Hacker Newsnew | past | comments | ask | show | jobs | submit | peanut_merchant's commentslogin

You do not need any form of photo or biometric id to work in the UK. I have never given anything of the sort, and have worked here for decades.

All that is required is a national insurance number (equivalent of Social Security Number in US).


Your employer is supposed to check that you have the right to work. This can be done in a variety of ways, depending on your citizenship, looking at your British/Irish passport is one of the ways.

https://www.gov.uk/government/publications/right-to-work-che...


You don't even need a photo ID to open a bank account in the UK

Personally been here about 15 years. There are ebbs and flows of different trends, but HN has never been pure tech or science.

My favourite thing about HN has always been it's sprinkling of the "interesting/thought provoking" amongst the more technical or science based e.g. 5 in depth articles about <insert language tech stack etc.>, blog post on seagulls in 1973 east Berlin, philosphical article ruminating on death and coding, Go compiler written in C# etc. etc.

I get the point about self promotion. But fwiw I think this list is quite insightful, and a fun read.


This is pedantry, temperature introduces a degree of randomness (same input different output) to LLM, even outside of that non-deterministic in a security context is generally understood. Words have different meanings depending on the context in which they are used.

Let's not reduce every discussion to semantics, and afford the poster a degree of understanding.


If you're saying that "non-determinism" is a term of art in the field of security, meaning something different than the ordinary meaning, I wasn't aware of that at least. Do you have a source? I searched for uses and found https://crypto.stackexchange.com/questions/95890/necessity-o... and https://medium.com/p/641f061184f9 and these seem to both use the ordinary meaning of the term. Note that an LLM with temperature fixed to zero has the same security risks as one that doesn't, so I don't understand what the poster is trying to say by "we have to treat LLMs as non-deterministic".


I understand the point that you are making, but the example is only valid with temperature=0.

Altering the temperature parameter introduces randomness by sampling from the probability distribution of possible next tokens rather than always choosing the most likely one. This means the same input can produce different outputs across multiple runs.

So no, not deterministic unless we are being pedantic.


> So no, not deterministic unless we are being pedantic.

and not even then as floating point arithmetic is non-associative


I get that maybe you meant culturally, but Ireland is a member of the EU whereas the UK is no longer. This forces a tighter alignment so makes your point about Ireland redundant.

The UK has continuously been pulled between it's dying imperialist vision of itself as a world power, it's close but conflicted ties with the US, and it's similarly close and conflicted ties with the EU.


I ran some quick benchmarks.

Ubuntu 24, Razer Blade 16, Intel Core i9-14900HX

  Performance Results:

  Initial Latency: ~315ms for short text

  Audio Generation Speed (seconds of audio per second of processing):
  - Short text (12 chars): 3.35x realtime
  - Medium text (100 chars): 5.34x realtime
  - Long text (225 chars): 5.46x realtime
  - Very Long text (306 chars): 5.50x realtime

  Findings:
  - Model loads in ~710ms
  - Generates audio at ~5x realtime speed (excluding initial latency)
  - Performance is consistent across different voices (4.63x - 5.28x realtime)


Thanks for running the benchmarks. Currently the models are not optimized yet. We will optimize loading etc when we release an SDK meant for production :)


on my Intel(R) Celeron(R) N4020 CPU @ 1.10GHz it takes 6 seconds to import/load and text generation is roughly 1x realtime on various lengths of text.


thanks for testing on the same hardware as mine, before me.


Not well versed in the field, what are the basic implications of this for health?


In the 1970s there was a lot of talk about ‘healthful negative ions’ and a fad for negative ion generators even though many of those also generated hazardous ozone.

Hydroxyl ions are a significant kind of negative ion in the atmosphere and they’re known to be good because they react with and clean out pollutants like methane

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

https://earthobservatory.nasa.gov/images/144358/detergent-li...


Here's some more research, since I have a tiny ozone generator in my fridge and I got worried:

Ozone concentrations as low as 70ppb are hazardous when you're exposed to it for several hours [1]. Estimates for Ozone's olfactory threshold aren't trustworthy, since you go nose-blind to it pretty quickly [2], but it seems like it's probably around 20-40ppb before olfactory fatigue sets in [3,4].

My takeaway is that Ozone generators for rooms/basements/etc are definitely a bad idea. The best-cited olfactory thresholds are all in the same order of magnitude as that 8-hour hazard threshold, and with nose-blindness being a significant factor, you just don't want to mess around with that.

Inside a fridge, though? As long as you don't actually smell any ozone when you open the fridge, and you don't just shove your head in the fridge for hours on end, I'd think you're probably fine.

[1]: https://ozonewatch.gsfc.nasa.gov/facts/SH.html [2]: https://www.ecfr.gov/current/title-21/chapter-I/subchapter-H... [3]: https://www.cabidigitallibrary.org/doi/full/10.5555/19602703... [4]: https://spartanwatertreatment.com/ozone-safety/


How can something be a negative ion generator without simultaneously being a positive ion generator?


You're right but a lot of times the positive ion is far less reactive and/or more massive than the negative ion. Not so much for OH-. Charge is not the only thing that matters.


Well, in a similar way to how you can't generate a negative ion without simultaneously generating a positive ion... how do you use the negative ion in a reaction without simultaneously using the associated positive ion in the same reaction?


Here's one way

https://en.wikipedia.org/wiki/Ion-exchange_membrane

Another, that you might be interested in, but it's more confusing

https://en.wikipedia.org/wiki/Proton-exchange_membrane


See the figure

https://www.fuelcellstore.com/introduction-ion-exchange-memb...

Each ion of salt participates in a different reaction


.


That isn't how chemistry works.


Isosaccharinic acid has the same chemical formula (C6H12O6) as glucose, which isn't acidic. However, they both have the same net charge.


When something is an acid, it dissociates into both a positive ion H+ and negative ion (rest of the molecule)

HA ⇌ H+ + A-


FWIU hydrogen plasma in water for hydrolysis would produce OH Hydroxl radicals. (and H2O2, O3 (Ozone), and NO_x).

TIL that Hydroxyl ions bind to methane and thereby clean the air?

Air ioniser: https://en.wikipedia.org/wiki/Air_ioniser :

> A 2018 review found that negative air ions are highly effective in removing particulate matter from air. [6]

But the Ozone. Ozone sanitizes and freshens, but is bad for the lungs at high concentrations.


The article does not come to any health conclusions, just studies the impact on indoor air chemistry.


if only there was a 'Discussion' section in the article, that goes over the basic implication of the study results... if only.


Yeesh, who taught you to debase others.


Even in Northern Scotland (further south than northern Sweden) this is the case. The latest aurora showing was vividly colourful to the naked eye.


One of the examples here strikes a parallel with the US government's "Enhanced Interrogation" used as a euphemism for torture starting during the bush years.

Straight from the Nazi playbook: "Verschärfte Vernehmung ("intensified interrogation"): torture"


We also use the phrase “concentration camp” (Konzentrationslager) to refer to death camps.

Which is really weird as it’s not a common English phase, and seems to be a straight copy of Nazi propaganda.


Not entirely correct, I'm afraid. Concentration camps were inaugurated by the Spanish and popularized by the British. Co-location of killing facilities and camps was not a common occurrence even in Germany, so there were many more concentration camps than death camps.


Now reframe the above, normalising as a portion of wealth.

Tax analysed in absolute dollar terms is misleading. What should be analysed is the tax's effect on a person's quality of life. A person with a million dollar income paying 50% tax still has 500k to burn. A person paying 30% at 50k a year has 35k left. That 30% means a lot more at the lower end, even though the dollar amount pales in comparison.


> normalising as a portion of wealth.

I don't follow this framing. You are saying we should look at taxes as "how much money does the government allow someone to have"?

This seems to assume a government has a right to take an arbitrary amount of money from its citizens. Do you believe this?


Where in the world would that not be the case?


If you mean practically speaking, everywhere? At some point production incentives are eliminated, leading to quality of life reduction and inevitably revolt or external invasion. Additionally, as government controls a more significant percentage of buying decisions, we shift closer along the spectrum to a command economy which at some point is computationally infeasible, though perhaps ChatGPT will make command economies work finally (mostly kidding).


Do we tax the rich?


Sure we do. Then they hire tax lawyers and use tax havens and end up paying next to nothing.

They also don't get audited.

Do the rich pay taxes? No.

See: rich people asking to be taxed more before 'something bad' happens.


The top 1% of rich people pay 39% of income taxes.

Do rich pay taxes? Yes.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: