raphman

FWIW, I wrote a small paper on this general topic a few years ago, collecting earlier work and own ideas.

"Files as Directories: Some Thoughts on Accessing Structured Data within Files"

https://dl.acm.org/doi/pdf/10.1145/3191697.3214323

a) because it is a general problem, observed time and again

b) I don't see how calling them out by name in this forum might be helpful

c) I don't know the whole story - maybe some of these people actually raised concerns or wanted to but felt powerless; causing people, search engines and LLMs to associate them with this case of scientific fraud seems potentially bad

d) maybe I'll have the chance to discuss this paper with them on social media or in person - then it would not be good idea to publicly shame them now.

e) anyone can quickly find out the names by reading the paper (https://arxiv.org/pdf/2412.17866)

Apparently, none of the 21 people mentioned in the acknowledgments questioned the source of the dataset. One of them also wrote a quite popular Twitter thread about the research. When notified of the recent events, he curtly replied that "It indeed seems like the data used in the paper is unreliable." [no need to mention them by name, I think]

Thanks. On Twitter, Ethan Mollick seems to imply that Robert Palgrave might be the scientist that triggered the investigation.

Oh, he also claimed that he got IRB approval from "MIT’s Committee on the Use of Humans as Experimental Subjects under ID E-5842. JEL Codes: O31, O32, O33, J24, L65." before conducting this research, i.e., at a time when he wasn't even a PhD student.

FWIW, in the q&a after a talk, he claims that it was a GNN (graph neural network), not a GAN.

(In this q&a, the audience does not really question the validity of the research.)

https://doi.org/10.52843/cassyni.n74lq7

> it had already been mostly abolished

c.t. is still standard at many German universities (and at all Bavarian universities I know). However, I know at least one university of applied sciences where lectures start at full hours.

Something like this might work. When pushing the knob in, a spiral inside it rotates around the tongue that is attached to the potentiometer (or vice versa). Once it is fully pushed in, the groove in the spiral rests against the tongue, allowing the knob to turn the potentiometer. Would probably need a lot of fine-tuning to get all forces exactly right.

https://winterwind.org/_media/pasted/20250426-220417.png

Yes, this sounds like a simpler solution - the major drawback/advantage would be that you'd need to insert the knob in just the right orientation so that the shaft end fits into the potentiometer's slot. Might be a bit fiddly. On the other hand, the poti would keep its state even when the knob is removed, and you wouldn't accidentally change the value when inserting the knob.

Just to nitpick the nitpicker: the knob's shaft needs to be longer than 15 mm, right? "3.5 mm" is the diameter, not the length of a small T(R)S jack.

To offer a counter-example to the many anecdotes about being gatekept(?) by veteran Wikipedia editors: I have the opposite experience.

I occasionally contribute to various topics, and in many cases experienced editors silently fixed formatting errors I made, allowing me to focus on contributing to Wikipedia without having to keep up with the best practices.

I also participated in a deletion discussion once, and - despite being inexperienced and in the minority position (keep) - the experienced editors considered my arguments and responded to them.

Two thoughts:

- The pope was not only a very important religious and political leader but also wrote and spoke about the relationship between humans and technology [1, 2]

- I joined Hacker News due to its links but stayed for the community of smart and thoughtful people (and the great moderation). Oftentimes, a HN submission acts as a seed crystal for "off-topic" discussions that people want to talk about. As the people that make up this community get older, and as the times change, the topics we discuss change, too. At some level, technology always has political, moral, ideological implications. For me, HN is one of the best places to discuss these.

[1] https://press.vatican.va/content/salastampa/it/bollettino/pu...

[2] Tim Cook on how Pope Francis influenced his thinking: https://www.facebook.com/watch/?v=1541230109287507

They are apparently not profitable (yet):

> Discord has raised a total of about $1 billion in funding. It has more than $700 million in cash on its balance sheet and the goal to become profitable this year, according to a person familiar with the matter.

https://www.theverge.com/2024/1/11/24034705/discord-layoffs-...

I'd argue that humans also more easily learn from huge mistakes. Typically, we need only one training sample to avoid a whole class of errors in the future (also because we are being held accountable).

I'm pretty confident that it is not AI-generated but simply a traditionally animated 3D model.

https://cdn.flipperzero.one/BusyBar_First_Block_Video_Deskto...

Welcome, Tom! Y'all are making HN a place that I still love to visit every day. I find it awesome how dang et al. not only manage to keep spammers and trolls in check but also actively improve discussions by merging threads and asking people to behave.

Hmm, my first guess would have been that you have been a target of "brushing" [1]. In a Reddit thread from 2020 [2], multiple people mention that they received surgical masks they did not order.

[1] https://www.bbb.org/article/news-releases/20509-amazon-brush... [2] https://www.reddit.com/r/tulsa/comments/hpe8s1/just_got_a_su...

The guys at Jolla (developer of Sailfish OS smartphone OS) are currently trying something like this¹:

https://www.jollamind2.com/

¹) however, their focus is less on home automation, more on personal information management, AFAIK.

Context: https://www.tarlogic.com/news/hidden-feature-esp32-chip-infe...

Both HN discussion [1] and blogs [2] mentioned that it was a nothingburger.

tl;dr: there are undocumented debugging interfaces on the original ESP32 chip that can only be accessed by code running on the microcontroller or via another computer connected via the physical UART interface. No remote exploit possible. Espressif will now provide a patch to disable these debugging interfaces and document all yet-undocumented interfaces.

[1] https://news.ycombinator.com/item?id=43301369

[2] https://darkmentor.com/blog/esp32_non-backdoor/

Sometimes, even a whole conference gets compromised, see e.g. ICIMTECH '21 [1]:

"NOTICE OF RETRACTION: While investigating potential publication-related misconduct in connection with the ICIMTech 2021 Conference Proceedings, serious concerns were raised that cast doubt on the integrity of the peer-review process and all papers published in the Proceedings of this Conference. The integrity of the entire Conference has been called into question. As a result, of its investigation, ACM has decided to retract the Entire Conference Proceedings and all related papers from the ACM Digital Library.

None of the papers from this Proceeding should be cited in the literature because of the questionable integrity of the peer review process for this Conference."

[1] https://dl.acm.org/doi/proceedings/10.1145/3465631

Ooops, yes - I misspellt my search query.

The transgender censor seems to be simple string matching (if query == "transgender"). Just putting the forbidden word in quotation marks makes it searchable, too [1].

I wonder if the filter was intentionally implemented in a shitty way by NIH staff so that they formally comply with an internal order but offer easy ways to circumvent it.

[1] https://search.nih.gov/search?utf8=%E2%9C%93&affiliate=nih&q...

Thanks, you are right!

According to a comment on Reddit, search problems have been existing for some time - they did not just appear after the DNS issue [1].

Also, I just searched for "transgender" on nih.gov, and got lots of hits [2], the first one being a publication on PubMed [3].

[1] https://www.reddit.com/r/NIH/comments/1j28ytk/comment/mfs14d...

[2] https://search.nih.gov/search?utf8=%E2%9C%93&affiliate=nih&q...

[3] https://pubmed.ncbi.nlm.nih.gov/34993517/

Cool, thanks!

DNS resolution seems to work again now. https://dnschecker.org/#A/pubmed.ncbi.nlm.nih.gov

I submitted this post and can guarantee that it was not my intention to associate this technical issue with DOGE. Quite the opposite: both in this thread and on Bluesky (where a lot of people blame DOGE for 'shutting down PubMed'), I have advocated for Hanlon's razor. My aim was to document that this was just a DNS problem, and to suggest a workaround to affected people.

I actually find it a little bit sad that the DOGE angle is so prominent in this thread.

Yes. As noted elsewhere in this thread, they still respond to TCP requests. I did not know this at the time of writing.

FWIW, nameservers lhcns1.nlm.nih.gov (130.14.55.72) and lhcns2.nlm.nih.gov (130.14.55.128) still resolve nlm.nih.gov subdomains.

I disagree. The nameservers still respond to requests on 53/tcp. It seems like someone messed up a firewall configuration or rolled out a bad server configuration. Shit happens - maybe exacerbated by the DOGE insanity. But there is no indication that this was done on purpose.

Can confirm.

Thanks for pointing this out. This makes it much more likely that someone messed up a server/firewall configuration than that Musk is ripping out network cables at NIH.

It seems someone shut down all NIH DNS servers. Right now the NIH website, PubMed and BLAST are up, but not resolvable.

Cloudflare's 1.0.0.1 DNS resolver seems to still have cached the records. Google and most others I tried did not. This probably explains why some people on social media could access the sites while others couldn't.

Workaround via /etc/hosts :

  156.40.212.210 nih.gov
  96.17.96.9     www.nih.gov
  34.107.134.59  pubmed.ncbi.nlm.nih.gov
  130.14.250.10  ftp.wip.ncbi.nlm.nih.gov
  130.14.250.10  ftp.ncbi.nlm.nih.gov
  130.14.29.110  blast.ncbi.nlm.nih.gov

EDIT:

While ns.nih.gov, ns2.nih.gov, and ns3.nih.gov do not respond, the nameserver at lhcns1.nlm.nih.gov (130.14.55.72) does.

Also see https://tldr.nettime.org/@ww/114089972404202687