So I use ISO8601 dates whenever I'm given a choice.
]]>A quick search just now produced this "curated list" (first in the results): https://github.com/kdeldycke/awesome-falsehood
Here's a thread discussing that list: https://news.ycombinator.com/item?id=24405941
TL;DR Nightmare fuel for programmers.
]]>[*] I could so easily have checked, but I'd want to do it by looking at something other than Wikipedia. Fortunately, 'founder of mormonism' was a useful search term. (This time.) Shame I didn't use it when I should have.
]]>Giving up and trying another search engine also happens. E.g. I've found many computer science papers via CiteSeerX, but sometimes there's a more direct path to the paper I want: the author's own online archive. For any history, I don't rely on Wikipedia to help me.
E.g. the article on John Smith looks like it was written by Mormons. I.e. it differs from what some historians may tell you. The BBC R4 In Our Time episode on John Smith differs regarding his wife. So the Wikepedia article looks like a case of DARVO, but who knows? Who should we trust, historians or Wikipedia editors?
]]>I've also chatted with alien cultists. That used to be fun, but every one of those folks has been sucked into Q-Anon. I've seen what they've talking about on their Discord server.
I've also seen a chatroom I used on IRC taken over by neo-nazi Trump supporters, at least one of them complaining about the violence of anti-fascists. I made it clear how I disagreed with them (two words) and left, never to return.
In an earlier post I said "Standard disclaimers apply." That's how I refer to a line from an Edgar Allan Poe story. "Believe nothing you hear, and only one half that you see."
So my default position is to only "half believe" any statement not backed by evidence or at least one citation I can verify. Even then, this "half belief" may have very short duration. E.g. the duration of a conversation with a friend or relation. Just out of politeness.
If they were to begin spouting, say, Q-Anon conspiracies, I might express some disbelief. If/when they don't take the hint, I might make my point less politely.
However, here I'm talking solely about how search terms may relate to search results. This is an area of active research by many people, including Google's former ethics people.
BTW, the Start the Week episode Greg recommended was better than I expected. I feared they would make the same category errors other media people make. This was a welcome exception. I can also recommend an episode of another BBC radio show, Word of Mouth, about chatbots. The guest that week was Emily M Bender.
]]>BTW, there were other sources at the time that had nothing to do with Bellingcat. Totally independant. One of them was the UK gov. Including 'Bellingcat' in the search term may have biased the results. Substituting 'BBC' may produce different results. I got a load of BBC coverage. Only one story cited Bellingcat. More frequent sources were from Theresa May, Ben Wallace, and others in gov.
This matches my own memory of the news coverage in 2018. I only learned later - much later - about the Bellingcat coverage.
Please understand I'm still talking about search results. Not journalistic bias, gov bias, or any other bias.
]]>https://www.bellingcat.com/tag/skripal/
I'll come back to these stories later.
It's hard to see how these stories could lead to your conclusions, so perhaps you got different results, or used a different search engine entirely. This is the problem with using search results as a guide to anything - search engines can learn from our search queries and adjust the results they feed us.
This is a known problem and has been discussed a lot over the years. A related problem is "search bias", where the bias comes not from the user submitting search terms, but in the shaping and selecting of the data in the engine itself.
https://en.wikipedia.org/wiki/Search_neutrality
https://en.wikipedia.org/wiki/Search_engine_privacy
These topics come up from time to time in InfoSec (information security) news. Nevermind the echo chambers in the infosec community itself.
This provides me with a constant source of amusement. See the three links I started this reply with. Why these three links? Why these two sites? That'll depend on so many biases. Some may be from my own search queries over the last year or more. Other biases may be from saerch terms used by infosec folks. Some will come from the way a search engine crawls the web, indexes the results, and the ranking system (very proprietry, very secret) used.
All these biases are being gamed, all the time, by various interested parties. One of those parties had an unfortunate incident earlier this year. Their founder famously became suddenly deceased, perhaps arranging a fake funeral - postumously. Who knows what really happened? It's the nature of these things to be heavily disputed, expanded on, and generally confabulated.
Welcome to the 21st century. Please strap in, this ride may get bumpy. Standard disclaimers apply. Tickets are non-refundable.
]]>Nevermind ways of the system time. The W32.StuxnetDossier has a few examples.
Stuxnet will only attempt to use MS10-061 if the current date is before June 1, 2011.
Also:
Stuxnet will verify the following conditions before exploiting MS08-67:
- The current date must be before January 1, 2030
- Antivirus definitions for a variety of antivirus products dated before January 1, 2009
- Kernel32.dll and Netapi32.dll timestamps after October 12, 2008 (before patch day)
MS10-061 is described by Microsoft as "Vulnerability in Print Spooler Service Could Allow Remote Code Execution", while MS08-67 is "Vulnerability in Server Service Could Allow Remote Code Execution", so happy fun time.
]]>MRDA ("Mandy Rice-Davies Applies")
So I'm not sure how reassured I should feel. I still have the Stuxnet dossier in my computer science papers archive, filed under "security". Looking at it again, one line immediately leaps out.
Victims attempting to verify the issue would not see any rogue PLC code as Stuxnet hides its modifications.
I may have missed something in the Guardian "report". Does this rebuttal say anything about lessons learned from Stuxnet? Is it too soon to expect any action from a bureaucracy the size of the UK government? Could this just be another bad case of clickbait journalism?
]]>