Jimmy Utterström

Staff Engineer at Storytel 👨‍💻 Hobby musician 🎼

[Jimmy Utterström @jimutt]

I got inspired by @AllanWalkerIT to experiment with some #LIDAR data on a #Deckgl point cloud layer with a @Mapbox GL map. I've never worked with .las data before but after some trial and error it came together quite nicely. 👍

[Jimmy Utterström @jimutt]

@merill @kellabyte Wish it was the case for their Windows 11 devs 😅

[Jimmy Utterström @jimutt]

@MMelkersen @brody_n77 @_JohnHammond This solution seems to potentially circumvent the need of the particular Bitlocker key
https://t.co/kJo4No3uAs

[LetheTheForgotten @LetheForgot]

@SwiftOnSecurity What we did was use the advanced restart options to launch the command prompt, skip the bitlocker key ask which then brought us to drive X and ran "bcdedit /set {default} safeboot minimal"which let us boot into safemode and delete the sys file causing the bsod.

[Jimmy Utterström @jimutt]

@tagsalegirl Those clouds on the one with more of an autumn vibe! 🤩😮

[Jimmy Utterström @jimutt]

@T_Zahil @martindonadieu @Cloudflare Especially if the attacker rotates IP every 5 mins or so, which is a quite low effort thing to do. So also there some care might be necessary if you are using a metered service provider for smaller projects.

[Jimmy Utterström @jimutt]

@T_Zahil @martindonadieu @Cloudflare With default settings I find that CF are usually not super quick at mitigating DDoS attacks that aren't of considerable size, i.e. the type that might still be devastating for a side project that doesn't generate revenue, but that wouldn't be a problem for any larger actor.

[Jimmy Utterström @jimutt]

The timing between this announcement and the Windows Recall one is hilarious 😂

[Microsoft Edge Dev @MSEdgeDev]

Say goodbye to security breaches via screenshots👋

Data leak control capabilities in Microsoft Edge for Business now support screenshot prevention policies. Prevent screenshot attempts while using the Edge browser on a desktop! 🌟
https://t.co/Sg6DlsYMet

[Jimmy Utterström @jimutt]

@tagsalegirl Really cool how you managed to capture the depth of the scene, amazing! 👏

[Jimmy Utterström @jimutt]

@Oyin_dml1952 @gunsnrosesgirl3 That's not a raven. Probably some heron.

[Jimmy Utterström @jimutt]

@flour_and_arena Shortcrust pastry would be "mördeg" (tender/crispy dough, or something like that 😅)

[Jimmy Utterström @jimutt]

@flour_and_arena For cupcakes specifically we normally say cupcakes or muffins (but that's normally used if there's no glace)

[Jimmy Utterström @jimutt]

@flour_and_arena Looking forward to that! And it's indeed a better season for Stockholm than autumn & winter 🙂

[Jimmy Utterström @jimutt]

@bdefig @thekitze On the other hand it probably also increased the tolerance of different-looking phishing attempts 😉

[Jimmy Utterström @jimutt]

@jamonholmgren Examples like these might of course not be as related to the community management, transparency & collaboration aspects that I think you were referring to in the original post. Just wanted to comment on how few "real" obstacles there are to OSS contributions.

[Jimmy Utterström @jimutt]

@jamonholmgren The point being that there can be value found in publishing code that you yourself consider throwaway or something that you don't consider taking further. One can control the commitment level quite well

[Jimmy Utterström @jimutt]

It's admittedly a viable concern for the marketplace model and a future revshare model. However, if the value of your product truly lies in the raw data you possess, that's probably what you should be monetizing in the first place. Don't expose it through an OpenAI service.

[Jimmy Utterström @jimutt]

I can't really grasp the naivety that would make you believe that the data in the files uploaded for generating custom GPTs would Not be available to query for 🤷‍♂️ The whole reason anyone uploads a file is for the custom gpt to use the data from it, no?

[Kanat Bekt @kanateven]

Oh man -- you can just download the knowledge files (RAG) from GPTs. I don't know if this is a security leak or "just" a prompt engineering? @OpenAI @simonw

[Jimmy Utterström @jimutt]

Okay, maybe there's no need to make the original/raw file downloadable. But there's no reason the user should not be able to query for information from the files, anything else defeats a large part of the purpose. And then we're only talking security by obscurity anyway.