Elektrine
Log in Register
Paige Chat Timeline Gallery Friends Email Drive DNS Private DNS Domains VPN Kairo Nerve
Remote

Georg Weissenbacher

@GeorgWeissenbacher@fediscience.org
mastodon 4.6.9
  • Open on fediscience.org

Computer Scientist at TU Wien.

0 Followers
0 Following
29 Posts
Joined November 12, 2022
Website:
https://www.georg.weissenbacher.science
DBLP:
https://dblp.uni-trier.de/pid/15/3636.html
Google Scholar:
https://scholar.google.com/citations?user=1KfaH-QAAAAJ&hl=en&oi=ao
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 8mo ago

Over the last year, I've wondered how to reduce my dependency on US tech giants. Especially for mobile phones, this is quite difficult - you either go for a de-googled system such as #GrapheneOS (but might end up having to buy that device from Google) or you accept some compromises (regarding the quality of the device or functions of the OS).

I've finally decided that perfect is the enemy of good and bought a #fairphone5 on which I installed #ubuntutouch. While a number of things are still missing (#Signal - @signalapp@mastodon.world and banking apps, for instance), a lot of the stuff that I do on a daily basis already works (phone calls, text messages, emails, calendar, podcasts, photos, web browsing, Mastodon, and Matrix, for instance). Yes, I had to change my habits a bit, and I do still depend on a secondary device for some Apps, but overall, I'm quite happy. I am confident that in the not so distant future, #ubuntutouch and #postmarketos will become viable alternatives to #iOS and #android

And I haven't enjoyed playing with an electronic device that much in a long time.

fediscience.org
55
10
17
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 7mo ago

@krismicinski@types.pl @shriramk@mastodon.social @jfdm@discuss.systems @csgordon@discuss.systems @lindsey@recurse.social @jeremysiek@types.pl we have been claiming for decades that we are not just educating coding monkeys, so it shouldn't really matter that LLMs can now do all the coding. As far as I see it, it's still necessary to identify and clearly formulate verifiable requirements and specifications, come up with a modular design, and verify the whole thing, because I still believe the ultimate responsibilty lies with the developer. So students still need to understand the fundamentals. But yes, it has become much harder to check *at scale* whether they actually grasped them.

8
0
2
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 5mo ago
Replying to
@regehr @dpiponi that’s a problem with any kind of verification. In fact, that’s the validation problem (is the specification describing what we want) vs. the verification problem (does the implementation satisfy the specification). Validation would deserve a lot more focus from our community, especially in times of LLMs.
5
0
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 5mo ago
Replying to
@lindsey @regehr the work by the philosoper Imre Lakatos ("Proofs and Refutations") illustrates nicely how mathematical proofs aren't bullet-proof because of implicit or missing assumptions.
3
0
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 4mo ago
Replying to
@kemotep@mastodo.neoliber.al @dangoodin@infosec.exchange well, in the xz case, it was actually one of the main contributors who slipped in the attack. There is little that can be done against that.
2
0
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 5mo ago
Replying to
@GrapheneOS@grapheneos.social The European Union is developing a digital identity system that incorporates FIDO2 standards for secure authentication. Some governments (like the Austrian) already support FIDO2 for authentication and identification. Banks should just do that, too.
2
0
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 7mo ago
Replying to
@ErikJonker@mastodon.social "Legitimate interest" is their favorite catch-all phrase. The Austrian credit rating company Kreditschutzverband has a full history of all addresses I have ever lived at (at least those in Austria). When I asked them to delete it, their lawyers sent me a letter stating that they have legitimate interest. #ksv1870 In fact, I thought this had been ruled illegal already (https://noyb.eu/de/gewonnen-grossteil-der-datenbank-von-kreditauskunftei-crif-illegal) thanks to @noybeu@mastodon.social , but apparently the Kreditschutzverband disagrees.
noyb.eu
4
0
2
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 6mo ago

Tony Hoare, in addition to his outsize influence on verification (and computer science in general), was a very kind and supportive person. It was very sad to hear he's no longer with us.

https://cacm.acm.org/news/in-memoriam-c-a-r-hoare/

cacm.acm.org
3
0
1
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 5mo ago
Replying to
@regehr yes, but it is also important that people understand what verification means (and yes, I know you do), also to manage expectations. You can only verify code against some specification or requirements. If that specification is incomplete, then so are your guarantees. Formally verified doesn't mean bug free, unless you have fully formalized what bug free means (which is next to impossible, if you think of security vulnerabilities such as side channel attacks). Formal verification still improves the situation, so we are better off with "formally verified" code than without.
2
8
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 5mo ago
Replying to
@regehr you state that someone takes formally verified code and finds real problems. However, the blog post states that the problems were found in the *non-verified* code: the parser and the Lean runtime. I quote from the blog post: "The positive result here is actually the remarkable one. Across 105 million executions, the application code (that is, excluding the runtime) had zero heap buffer overflows, zero use-after-free, zero stack buffer overflows, zero undefined behaviour (UBSan clean), and zero out-of-bounds array reads in the Lean-generated C code." Arguably, the denial-of-service was a problem in the verified code, but: "[t]he denial-of-service was a missing specification". Which, as I posted elsewhere, is a validation problem and not a verification problem.
2
10
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 6mo ago

This might be #ww3, but at least it wasn’t an Austrian starting it this time.

https://www.nytimes.com/interactive/2026/03/21/world/middleeast/iran-war-gravity.html?unlocked_article_code=1.U1A.g7kU.mBOHCqPlVjcZ&smid=nytcore-ios-share #nytimes

fediscience.org
2
0
1
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 13mo ago
Replying to
@Gargron@mastodon.social in principle, yes, but on the other hand… https://en.m.wikipedia.org/wiki/BonziBuddy
en.m.wikipedia.org
5
1
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 8mo ago
Replying to
@mmorschel@mastodon.social the best solution is probably an open architecture like web apps combined with secure 2FA (like #yubikey). That way all you need is a browser and you are not tied to a (mobile) operating system.
2
3
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 6mo ago

The European Open Access platform is growing, and Austria is one of the 11 countries backing it.

https://research-and-innovation.ec.europa.eu/news/all-research-and-innovation-news/new-era-open-research-europe-2026-03-26_en

#openaccess #austria #fwf

A new era for Open Research Europe
Research and innovation

A new era for Open Research Europe

European Commission to co-fund new phase of operation for open access publishing platform.

1
0
1
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 7mo ago
Replying to
@davefischer @petko remember that this was on DVDs and paying customers could not skip it.
1
1
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 8mo ago
Replying to on mastodon.social
@miskolin@mastodon.social I guess the mushrooms are a reference to LLM hallucinations?
1
0
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 9mo ago
Replying to
@sqwabb@mstdn.ca @cstross@wandering.shop Warp also already had speech recognition! “Move window 5 pixels up.”
1
0
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 10mo ago
Replying to
@joey Attendance: Not mandatory (unless you want to pass) 😂
1
0
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 13mo ago
Replying to
@aral@mastodon.ar.al @writingslowly@aus.social There’s a problem with point 1 - who decides what “can be built”? For instance: Many legislators want companies to implement encrypted communication in a way such that they - and only they - can listen in. Numerous experts believe such a system can’t be built (at least not securely). If I’d run a company I’d rather not end up in court where a lawyer explains to me what can be built and what not.
1
1
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 6mo ago

Isn't AI search great? I asked Google why bangs (Qwick search) on #Qwant doesn't work (the way I'm used to it from #DDG )

What would you expect the AI summary to link to? Well, a page which explains why bangs don't work, right?

Indeed, that's what the link (https://www.instagram.com/p/DJuMQSVxn74/) is about - kind of.

instagram.com
0
0
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 7mo ago

@GrapheneOS@grapheneos.social @navi@social.vlhl.dev @alexia@shrimp.starlightnet.work @lumi@snug.moe I don't remember writing anywhere that GrapheneOS is awful or a stopgap.

0
0
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 7mo ago

@GrapheneOS@grapheneos.social @lumi@snug.moe @alexia@shrimp.starlightnet.work well, to be fair, you wrote "it's a lot of work which is meant to be done by the OEM and their hardware partners" (referring to providing mainline branches of the kernel)

@GrapheneOS@grapheneos.social

grapheneos.social
0
1
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 5mo ago
Replying to
@dan @lindsey paper copy is in the mail!
0
0
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 7mo ago
Replying to
@pax0707 @marcoarment the cheapest 1password subscription is EUR 33 a year. For a ridiculously small amount of storage space. I don’t understand why password managers are so expensive.
0
0
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 2mo ago
Replying to
@krypt3ia@infosec.exchange @briankrebs@infosec.exchange have you updated your Linux kernel recently?
0
0
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 13mo ago
Replying to
@aral@mastodon.ar.al @writingslowly@aus.social I understand that there are regulations, and of course I'm all for that. But they need to be specific. Building codes, for instance, are very clear (at least where I live). They specify the minimum height of rooms, size of the windows (for appropriate lighting), ventilation requirements, etc.
0
0
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 7mo ago
Replying to
@Cdespinosa@mastodon.social yeah, right. Did anyone ever get their $125 from #equifax ? Even the #FTC advised consumers in the end to not even ask for it. https://www.ftc.gov/news-events/news/press-releases/2019/07/ftc-encourages-consumers-opt-free-credit-monitoring-part-equifax-settlement
FTC Encourages Consumers to Opt for Free Credit Monitoring, as part of Equifax Settlement
Federal Trade Commission

FTC Encourages Consumers to Opt for Free Credit Monitoring, as part of Equifax Settlement

The Federal Trade Commission is urging consumers affected by Equifax’s 2017 data breach to consider signing up for the free credit monitoring offered as part of the settlement.

0
0
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 8mo ago
Replying to
@mmorschel@mastodon.social technology has changed since then - e.g., now there's WebAssembly (introduced 2017, 2.0 is a W3C standard since 2024).
0
2
0
0
Open post
Georg Weissenbacher @GeorgWeissenbacher@fediscience.org
· 6mo ago
Replying to
@skaphle@social.tchncs.de @Becovich@mastodon.social as @michael_w_busch@mastodon.online writes, the problem is not necessarily bandwidth - satellites can provide a very high bandwidth - it is latency. Satellites in geostationary orbit, for instance, have a latency of 1/4 of a second. That’s why Musk’s satellites need to fly low. But that reduces the area they can cover.
0
0
1
0
Back
313k7r1n3
Elektrine

Tor hidden service

elekhj7afj4qnrr4yd3bkzslsyo5jgfxw3orgjkhlcxifueodybyiiad.onion

I2P eepsite

j6b6cyk6gjmepjih7jjadxgxvvf3lzzujljuu2v4biemzpg3naya.b32.i2p

Platform

  • Email
  • Chat
  • Timeline
  • VPN
  • DNS

Company

  • About
  • Contact
  • FAQ
  • Lite (no JS)

Legal

  • Terms of Service
  • Privacy Policy
  • Transparency Report
  • Report Abuse
  • Warrant Canary
  • VPN Policy

Support

  • support@elektrine.com
  • Report Security Issue
Mail client setup IMAP mail.elektrine.com:993 POP3 mail.elektrine.com:995 SMTP mail.elektrine.com:465
© 2026 Elektrine. All rights reserved. Server: 21:54:57 UTC