• 2 Posts
  • 437 Comments
Joined 1 year ago
cake
Cake day: June 24th, 2024

help-circle




  • Yeah. It’s a fucking son of a bitch cancer. Together with Pancreastumors the ones I am most afraid of.

    When I had my paramedic traning one of my instructors got one. He waited till he had the first seizure - and then drove into the next town so the ambulance crew finding him would not be “one of his own”, went into the woods, called the cops to tell them what he way about to do and, well, “self removed” the whold thing with a 9mm to the head.

    …little did he know I was send over that day to cover for someond who got sick. …Nevertheless I would do the same in his situation.






  • No. Patriot act had provisions to make it basically impossible to go through the regular law system. This is not the case here - the whole stuff needs to be approved by a judge and they usually handle these things fairly restricted due the high constitutional burden. Additionally you need to be involved in a pretty specific subset of crimes to be even a possible target.

    This is explicitly not the case and unlike the US the evidence obtained illegally can basically never been used with a red-hering, etc.

    And there are provisions in the law that actually make the cops already be angry about it, make DAs cry and defending lawyers happy: They must prove that there is no other,less invasive, way to achieve the control of the possible danger - and that can be fairly hard and they risk of the evidence not being admissible in court and their own legal consequences for it. Additionally the approval is time limited, etc. Don’t get me wrong,I am not happy about it either, but it’s a necessary evil,imho - it’s the modern way of a phone tap, which has been a measure used by the cops since 1920ies and it’s sadly one of the few ways to fight organised crime. And it’s the far better alternative to what a lot of other countries want to use and currently push: Backdoor in all messengers.

    The true issue with that law is NOT that. The AI bullshit, numberplate recognition, population data use,etc. are the actual issues.


  • Heise is generally one of the most reliable tech source news outlets in German. (E.g. Netzpolitik.org

    And yes,this has been reported on various other news outlets as well

    Sadly this is actually not the main issue with that law - the use of KI and population data is far more problematic and overstepping boundaries. The installation of the remote logging software (“govermental trojan”) was already possible before, but not by the state,only by the federal criminal investigation office (BKA) and it still has pretty high boundaries (a judge needs to approve,approval is fairly limited in it’s timespan, there are limits what crimes it can be used for and how data can be used) While I am not happy about it either, personally I must admit I have far less problems with it than with the other parts of the law. Observation on high risk people has always been part of police work and tbh, it needs to be done if you want to tackle organized crime, violent extremists, etc. Back in the 90ies they tapped the phone of the Mafia associates, now communications have shifted so from my point of view it’s acceptable IF “imminent danger” is not routinely assumed regularly (that reduces the limits) and the judges look at it critically - which at least some of them do. And: It’s far far better than the alternative that is being pushed: Backdoors in all chats - as pushed by some EU countries on a EU scale. Far worse.

    The AI, automatic number plate scanning,face recognition, etc. part of that law is the issue.

    If you speak German: Page 25 ff. https://www.parlament-berlin.de/ados/19/IIIPlen/vorgang/d19-2553.pdf


  • Just a theory: There is a good chance that your provider does CG-NAT and that was the issue with OpenVPN. These would persist with wireguard,sadly, unless you solve them properly. (Which can be tricky). But just for the book: Running an Wireguard Container behind your router and have a port forwarded to it is an option. (But still needs CG NAT adressed)

    Thaft leaves you with a few options:

    • Cloudflare: Imho a bad idea - it’s evil, it’s monopolistic and while it’s “an easy way” it has its technical downsides. As you said a domain is still required.

    • Use a small VPS and run a wireguard tunnel and maybe pangolin as a reverse proxy on it.It has the benefit of being very flexible and once configured is fairly stable and it puts the security part outside your network. But it costs money unless you maybe make it work on oracle’s free tier. I would still recommend using a cheap domain,though)

    • As others have mentioned: Tailscale/Zerotier/Netbird absolutely are an option if it’s just for you. But they get nasty if it’s for more people or larger deployments with tailscale and while netbird is far better it’s less common and does require a domain as well. (Which,again,is not a bad idea to have)





  • Yeah. LLMs are helpful IF you now what they are, what they can do, etc. and that you still have to check their solution.

    For some things they are good and safe time.

    Example: We currently work with a company who has understood this verx well: They provide doctors reports and also validate manually written ones. Their workflow basically is: The doctor orders a doctors report for patient A. The system checks witht the doctor that it has understood the major bullet points from the clinical information system and the order given by the doctor. (Basically a “yes” “no” answer system like “Patient was admited for chestpain?”) and then creates the report with a color coding in the version the doctor needs to check. (E.g. "white background for “we are 100% sure about the data as it comes directly from the clinical information system” aka names, dates, lab results, green for “very likely to be correct”, yellow for less than 99% certainity, orange for less than 90%,etc.) The system also has a mandatory “minimum verification time”, e.g. for a longer letter you will need more time to read it and can’t simply click okay without reading it. Nevertheless the doctor needs around 10-15% of their time for a report and overall satisfaction by the facilities receiving them has massively improved as they are more “standardized” in it’s format and they find the same information in the same parts every time.

    For quality assurance a cetain degree of reports will be send “upwards” and the QA manager of the facility reads them as well, additionally some departments have implemented that if doctor A writes a report e.g. 5% goes to doctor B to check and vise versa.

    Additionally the system also validates manually written reports (for training reasons med students and interns are not allowed to use it), e.g. “are you sure this patient had hypotension when admitted? He was given large amounts of a anti-hypertension drug within 5 minutes after being admitted”. This has significantly reduced errors in manual reports (we check a four digit number by now).

    Whole thing runs locally (*) and does not need any outside connection at all and the model is open to the clients (they are actually encouraged to train it with their own data and to let their data security officers check it).

    (*: This is actually an issue,sadly. As doctors reports are basically written all at the same time during the day and of course the model needs a fair share of ressources it needs a somewhat beefy appliance for it. While that is not an issue for a larger hospital it is not feasible for a smaller community hospital who barely manages it’s own IT infrastructure or even a small doctors office - but we found these to be the ones who would benefit the most from it. At the moment we don’t have a good answer for that beside hosting it elsewhere which would defeat the purpose and make it a privacy nightmare. We will see if these guys solve that before they fully introduce it into the market)

    That saves time, money (even if it’s only for the larger facilities atm) and benefits patient care directly.



  • What part do you mean exactly? That we train technical rescue with the firies every year? That comes with the job - and it’s more “their part” - even the smaller volunteer fire departments do that at least once a year to keep up their skills with the heavy tools and we get train to work with them. It’s usually not that expensive either,as they use an old car that wouls be wrecked either way. Often they get them for free (as disposing a car can be expensive otherwise).

    The total submersion training? That is much rarer and I only did that once, but it’s part of the training of the more specialist water rescue crews, afaik. The issue here are not the cost,but finding a suitable location - you can’t just use your community pool or nearvx lake so you either have a quarry with a lake or something like that (we did it in a army tank training ground, they have a pool to drive tanks through. Nowadays it’s almost impossible to get a permit to train there due to the hybrid warfare the fucking russkies do) We have a specialist training side that enables indoor training of helicopter based winch rescue from flood water/flooded buildings,etc. though. (Mainly focuses on mountain rescue,though and has a cold chamber,etc. as well) (https://bw-zsa.org/) (https://youtu.be/2qWJNgKVo18)

    Similar training of an automobile club: https://youtu.be/T5l1ayTryhc

    HUET for helicopter is mandatory for everyone who works with maritime helicopters, e.g. oil rig workers, maritime pilots,etc. Therefore they are fairly common. https://youtu.be/YyPzzLwpzvw


  • It’s very very very unlikely/next to impossiblr that you are able to destroy the front window in a modern car - even with tools that takes time and skill

    With a back window you might have more luck,depending on the car,at least with a center punch tool, you might have a better chance - but that requires you to be able to a) be alert and uninjured enough to do so b) find your way there in total darkness, wrong orientation,etc. c) manage to apply the right amount of force.

    Source: Am a paramedic,have to destroy windows once a year for our training with the firies, additionally have done “total submersion” training once. (Basically the same as what happend here. You get into a car,this car get spun on the roof, then slid down nto a pool/lake - with the difference that you have space on all sides, there is a rescue diver with you in the car and another two are next to it. It is still a fucking nightmare and MUCH worse than what I used to do to work on the helicopter - their HUET was much easier.


  • Netbox,especially when combinded with Plugins is so incredibly good and might,that’s it’s almost funny how good it is. What I do Plugin wise:

    • Documents: not implemented yet by me,but one could store manuals,etc. directly within netbox.

    • Lifecycle and Inventory: While it’s not as good as snipe-it (tbh, inventory is imho one of the worst plugins) it does the job for my small deployment

    • Slurp it to scan automatically

    • QR Code for obvious reasons

    • Floorplan as well

    Of course that sounds overkill for a small deployment, but I simply forget too many things after a few months otherwise and it’s something my family (wife is in IT and far more qualified than me) would need if something happens to me,so a proper documentation would be essential for that as well.