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

Nico, while I have the chance, I'd like to thank you for all your work on Kobo mods and fonts


HTML and CSS is enough for 99% percent of epubs, and that's the only subset of epub standards that Crosspoint tries to support


Isn't HTML and CSS already a huge surface to support, unless you are happy with a subset?


Custom firmwares support hyphenation


For people with no other albums, Carplay was playing that U2 album automatically when they enter their car. So some people were forced to listen to it :-)


RAM-only viruses exist.


That's for when the uC is able to execute code from RAM. Some uCs can only execute code from a separate address space (Harvard architecture), that may not even be readable (let alone writeable) by the program itself.

On such a uC the "inject random payload" option would be out. Still possible might be denial-of-service attacks, altering or exfiltrating data etc.

But most stuff online are regular von Neumann machines where -in theory- anything is possible.


Termius is a great ssh client. For local work, Termux will give you basically a full Linux vm in your pocket (with a few caveats)


Awesome project. Do you use it to develop itself?


I fixed the eGPU disconnecting issue on my 2012 MacBook Pro by placing the entire motherboard inside an oven for 10 minutes (a rather unconventional solution for reballing). So, yes, MacBooks from that era were not as fragile as they are now. By the way, that laptop still works.


I cut my Apple-repairing teeth on an iBook G3 that would lose its BGA-soldered GPU if it overheated. The "fix" was to use a heat gun, propane torch (my choice), or even burning alcohol to gently reflow the solder:

https://forums.macrumors.com/threads/repairing-ibook-g3-grap...

I did it at least three times before that laptop died. Fun times.


On the other hand, many interesting links (IMO) I submit to HN also get zero comments


Worse, I've had submissions (both links and comments) get flagged in the past, and I have no idea why. I suppose they must have validated some HN policy, but if I had more information about the rationale, I could avoid making the same mistake again in the future (all of my submissions where that happened were for genuinely interesting contents or 100% non-offensive opinion comments).


They may have been SotA at the moment of writing


Sure, but they posted this 4 days ago. The minimum I'd expect for quality research is for them to skim the abstract before posting and change that line to:

"Models from leading AI labs" or similar. Leaving it like now signals either sloppiness or dishonesty


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

Search: