Posts by Pascal 346 publicly visible posts • joined 25 Dec 2009
What does "destroying an NFT", aka the ultimate vaporware, actually involve?
Deleting the bookmark to the URL that points to a JPEG of stoner cat from their company browser?
Deleting the "wallet" that's listed as the owner of that bookmark in whatever blockchain they used to pawn off their URLs that point to JPEG of stoner cats so that it can't ever be pawned off?
Deleting the actual JPEG of a stoner cat from the web server that's hosting it?
The Gen2 satellites are not only 4x heavier but also much bigger than Gen1.5 that they currently launch.
The largest batch with Falcon 9 was 54 satellites but it's generally in the 45-50 range depending on target orbit.
For Gen2, it would be single-digit counts. Not economically (and timely) viable.
Only way they can get 1000s in orbit is with the ~100+ tons to LEO (and massive volume) of Starship.
Not even a little bit, since they can still go ahead as long as the existing vulnerabilities don't affect security or there is a mitigation plan. At best it encourages disclosure --- and only of KNOWN vulnerabilities at that. If the biggest flaw imaginable is discovered the next day, it "was there", but it wasn't known.
Once starlink v2 satellites (the ones with satellite-to-satellite comms) are widespread, the obvious US Military application would be to strike a deal with Starlink to piggyback on the network without using Starlink ground station. Plane-to-Carrier, Boat-to-Boat, Drone-to-Drone or whatever they fancy.
What Michael Roe got was a big cease-and-desist letter from Microsoft for what clearly looked like trademark infringement. When it came out that it was a kid that probably did that because he thought it was funny, they actually traded him training, an xbox, and all sorts of goodies for the domain. So in that other, legal-defense-of-their-brand case, MS actually treated the guy pretty well in the end.
Nobody should have entered in "free, infinite storage" agreement with Google and not know this would come to an end at some point.
The fact that there are entire universities that rely on that for their data storage and are at a loss as to what to do with their petabytes of data all of a sudden is however absolutely insane. Then on top of it to have this huge blindspot where this deadline announced a year in advance is causing a stir *now*. Think of all the potentially insanely critical research data that is just entrusted to Google, with no idea of what their backup plans / recovery procedures are.
90 days was chosen for a couple reasons.
One of which was to force automation.
With 1 year certificates, inertia would have caused LE certs to be managed by hand as a majority of certs from old CAs are. 90 days, nobody goes live with that before automating them.
The other one is stated as "They limit damage from key compromise and **mis-issuance**".
Which ties in real well with the current situation: why not let them expire?
It could easily work with an internal key in the CPU that intel needs to counter-sign to activate a feature, that can easily be done a) offline so "phone home" component, and b) with zero danger if the key leaks because it will only activate the feature for that one particular chip.
Of course Intel's private key that signs all these things would be prime target for leaks/espionnage!
"You'll be authenticated on any services which the user was logged in on when they committed the database"
Not to diminish the importance of the issue but the above statement assumes that "all authentication/session cookies for any and all web sites ever have no server-side expiration mechanism".
One one side of reality we carefully organize VMs to not cross numa node boundaries because the interconnect between 2 CPUs slow down memory access and performance too much... And on the other side, these lunatics think "off-system in the next rack" would be useful for any practical application.
Still:
Assuming the artist rendition in the article is accurate, the sail increases the footprint of the satellite by about 3 or 4 times. 3-4x is still 3-4x, no matter how negligible the initial risk is.
The added drag is expected to reduce deorbit from 25 days to 15 days, a 40% reduction.
So how much is the risk of collision reduced over those 15 days given the higher surface, compared to the risk of collision with the original size over 25 days?
It's not about "accepting", it's about "getting business done".
As a service provider you align with what your client provides, if LibreOffice fits that's fine, if it does not, MS Office's license is just part of the cost of doing business.
Adobe's licences are more less the same, your client provides photoshop files only compatible with the latest version that your free software can't handle right? Well, the Adobe license is part of your cost of doing business.
Sometimes you are in a position to recommend technology to a client, or help a startup pick what they will use, and you can then push as much as you like towards your preferred technology.
But if you try to take a stance and tell a client "yeah no, don't send me .docx files" or "don't send me psd files" and try to put it on them to align with your preferences instead of the other way around, you'll be "that weird guy they stopped doing business with".
"The same is true of emails - easy to spoof, but far more difficult to intercept (especially if using a private/company email address & server rather than a web-based public server)"
You are saying it's easier to steal email content from a specific email account hosted at Google than from the average, halfass-configured corpo mail server that is most likely something installed 10 years ago and left unimproved?
In my youth (1991-92) I worked a couple years as a technician in a local computer shop, so of course Ihave a lot of funny stories in that day and age of people being completely clueless about computers.
The one regarding floppies:
The DOS format command had a message stating "you can format multiple disk" when you started it (the exact form it took is lost to my memory).
What that meant of course is that after formatting a floppy it showed a prompt asking "Do you want to format another disk (Y/N)?".
One of the calls I got went: "No matter what I do, I can't put more than 2 in at the same time".
Was it?
I am legitimately curious.
One one side, you buy a licence and support contract for something, and that support is due.
Then other people install too many copies, unlicenced; you are at fault for sure, and the supplier can seek reparations.
But can they actually refuse to support the bits that you use legitimately, with your paid-for contracts?
Or can they only refuse to support the "extra" licences that are under dispute?
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
