Comment by __MatrixMan__
7 hours ago
This seems largely to be about security by obscurity.
I was hoping it was gonna be about how our modern practices are making things less secure.
For instance, we claim we need to be able to rapidly update clients so that we can patch security vulnerabilities as they are discovered (often without involving the user at all). And there are a lot of companies that have an incentive to push this narrative because they have products which facilitate this whack-a-mole approach to security. But there's no reason to believe that new software is more secure than old software. Old software is just more likely to be known to be insecure. So anything written before it became trendy to update without your user's consent is more secure in at least one way because it is not configured to automatically update to whatever comes down the pipe from "the vendor".
Right, the sub-headline "security through antiquity" is a nod to both the oft-repeated "security through obscurity," while pointing out the fact that antiquity is sometimes underutilized or undervalued as a security mechanism (even if not tamperproof). Cyber defenses are thus best multi-pronged, which offers the best protection- by maximizing the amount of time an intruder would need to spend to determine what system it is using.
About your second comment- newer systems CAN be less secure, but not always. But even if they are, falling back on things like eLoran are important.
Sure, it can go either way, and so which version to use is a decision to be made by the user, not on their behalf by a vendor.
Consider the CrowdStrike debacle. It wouldn't have been a big deal if they could just boot to yesterday's config because today's config is broken, but neither vendor involved trusts their users enough for that kind of thing, which turned it into a disaster.
Forgot about that. I intially thought you were referring to CloudFlare Captcha's, which Google and Firefox are partnering with: https://www.techtimes.com/articles/318891/20260623/cloudflar...
It's remarkable how much more time one has to wait just to access the same level of information (Cookie Policy, EULAs, etc).
Even with "lighter" weight HTTP Firewall utilities such as Anubis (https://github.com/techaroHQ/anubis), the average user has to pay for other's (DDoSers) misdeeds. Now no one can visit a site without a firewall unless it's static and rate-limited by IP address to avoid crashing a small home server. Some impressive LMDBs might be able to serve up to 100,000 requests per second on a lightweight PC, but then again they could still be knocked offline from a super resourceful organization. I like not putting all my eggs in one basket.
A similar bug could happen with something like Cloudflare- the computer serving up the firewall could have a bug, and it's not programmed to fall back on serving the site without the gatekeeping. If the site is prominent enough, it might routinely face DoS attacks which prevents it from being used. But if it's an uneventful day, it could still manage a lower-tech firewall and still be functional.