Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
croemer
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
27 ms
·
661.
▲
by
croemer
2y ago
You've got it completely backwards. MP3s just trick you into thinking it's the same thing. It's actually totally different if you analyze it properly, in a non-human sense. LLMs are able to often _preciesely_ recreate in cont
662.
▲
by
croemer
2y ago
Good find! I was briefly thinking what if techcrunch was using Automattic but quickly dismissed the thought.
663.
▲
by
croemer
2y ago
Which I created to read the article in the end :)
664.
▲
by
croemer
2y ago
https://archive.is/1YKki
665.
▲
by
croemer
2y ago
techcrunch.com seems down across the board right now? [20:27 UTC]
666.
▲
by
croemer
2y ago
Duplicate: https://news.ycombinator.com/item?id=42641987
667.
▲
by
croemer
2y ago
Nice example with the 1/(m-m_mars)!
668.
▲
by
croemer
2y ago
https://xcancel.com/SybreWaaijer/status/1875230654054752374
669.
▲
WooCommerce collects sensitive information without consent
(twitter.com)
3 points
by
croemer
2y ago
|
2 comments
670.
▲
by
croemer
2y ago
> Fun fact: WooCommerce collects your sensitive information without asking for consent. >Luckily, it merely consists of about 1000 data points. >Let's dive into it... >Disclosure: I'm a WooCommerce Marketplace partner.
671.
▲
A deep look at the WordPress Update API
(duanestorey.com)
2 points
by
croemer
2y ago
|
1 comments
672.
▲
by
croemer
2y ago
Mind blowing privacy violations
673.
▲
by
croemer
2y ago
No, still doesn't work. Progress bar now stuck at 95% for dozens of seconds after initial progress to 95% takes only 5 seconds. Broken.
674.
▲
by
croemer
2y ago
This is almost definitely against GDPR, there's no indication whatsoever of which legal entity is holding the data and how long it is stored on which servers where.
675.
▲
by
croemer
2y ago
The title is editorialized, it should be something like: "Anyvoice - AI Voice Cloning"
676.
▲
by
croemer
2y ago
Getting error: Failed to generate voice
677.
▲
by
croemer
2y ago
News report with full statement from university: https://boingboing.net/2025/01/10/student-says-university-of... > Thanks for your inquiry and for the opportunity to clarify this situation and our general
678.
▲
by
croemer
2y ago
First news report: https://www.king5.com/article/tech/university-of-washington-... It contains a statement from the university as well: > "On occasion, the Registrar will place a temporary registration hol
679.
▲
by
croemer
2y ago
First news report of the incident that got >1.3k points yesterday.
680.
▲
by
croemer
2y ago
Update on the Update: > I feel terrible writing this. But here we are. Today UW issued an official statement that was obviously targeting me and sent it to the media. In their statement they point to the "Tampering & Abuse"
681.
▲
by
croemer
2y ago
I don't get why OP didn't contact he TA directly before going public with it. It seems there's some major miscommunication happening. It looks to me like OP went berserk instead of resolving a misunderstanding peacefully. The
682.
▲
by
croemer
2y ago
Someone who says they were OP's teaching assistant for this project replied on LinkedIn (seems trustworthy) > JD, this has just been brought to my attention. I was the CSE 403 TA in charge of this project last quarter, and while I
683.
▲
by
croemer
2y ago
Trusted advisor doesn't need to be lawyer. Can be trusted professor.
684.
▲
by
croemer
2y ago
Hard disagree. It doesn't harm to try to solve it with the help of a trusted intermediary. One doesn't lose anything but a bit of time. Instead it shows good faith. You can always escalate to the lawyer afterwards.
685.
▲
by
croemer
2y ago
I got their code to run and it's not good. It's unusable as is and could be created from scratch in a better way in a day. It lacks all integrations with the University API, that part was never written. The existing matching of sw
686.
▲
by
croemer
2y ago
I also worry about this which is why I've been researching things and posting my findings here. I wish this got picked up by at least the student newspaper. UW should be able to present their side. Hard to predict whether we will ever
687.
▲
by
croemer
2y ago
Can be faked but that takes effort and raises the stakes. Now you'd not just misrepresent, you'd also be forging stuff. We could then also discuss whether the text was credible or had some signs of being forged. Data is always goo
688.
▲
by
croemer
2y ago
While I agree that by going public on LinkedIn and HN scrutiny is fair game, he might still not want the scrutiny and/or be surprised by it.
689.
▲
by
croemer
2y ago
Someone else pointed out that the framework OP used in the project is the creation of his dad as well: https://github.com/SharpLogic/LightNap
690.
▲
by
croemer
2y ago
Ugh, this is multiple correlation/causation fallacies in one: > Studies from Nedergaard’s group and others suggest vigorous glymphatic clearance is beneficial: Circulation falters in Alzheimer’s disease and other neurodegenerative
More ›