4 ms·

Israel (Mossad?) can hack something in Russia, see tools and recognise those tools as top secret NSA gear. Do you wonder how they made that recognition? Were th

by harry8 9y ago

Israel (Mossad?) can hack something in Russia, see tools and recognise those tools as top secret NSA gear. Do you wonder how they made that recognition? Were they shared with Israel so they knew, in which case the source could have been Israel being hacked, right? Or they knew because hacking the NSA is something multiple nation states have done. I'd be completely amazed if the NSA wasn't absolutely full of spies acting for foreign powers and organised crime. At this point should you just fire everybody in the NSA and start again? If not, why not? I'm struggling to see genuine competence in improving the security of Americans amongst the constitutional attacks on the citizenry, attacks which most definitely have the opposite effect.

petre 9y ago
Governments never fire spies. They move them to reserve and pay them good pensions.
threeseed 9y ago
Israel is listed as an "observer" in the Five Eyes alliance. So yes it's quite likely that the US et al have shared assets with them. https://en.wikipedia.org/wiki/Five_Eyes https://en.wikipedia.org/wiki/Five_Eyes
dotancohen 9y ago
I can look at a Git commit and tell you exactly which of my coworkers wrote it without looking at %cn. Code has style, like spoken language has accents. One could argue that e.g. German spy tools copy the American style so that those decompiling it will think it is American. I argue that is a lot harder that it sounds. Code style is much deeper than whether or not to use braces around lone if clauses. The whole way of thinking, layout of data structures, use of getters/setters or properties, breakdown of what goes where and into which classes, breakup of large methods, etc etc etc. These signatures and many more give one a feel for the software's origin. Not proof, but a very solid foundation for suspicion.
- shagie 9y ago
  You might find De-anonymizing Programmers via Code Stylometry ( http://www.princeton.edu/~aylinc/papers/caliskan-islam_deanonymizing.pdf http://www.princeton.edu/~aylinc/papers/caliskan-islam_deano... ) an interesting read. I suspect that coding style guides are detectable in compiled output too. As an aside, a bit that caught my eye here: > This material is based on work supported by the ARO (U.S. Army Research Office) Grant W911NF-14-1- 0444, the DFG (German Research Foundation) under the project DEVIL (RI 2469/1-1), and AWS in Education Research Grant award.
  - jaclaz 9y ago
    >I suspect that coding style guides are detectable in compiled output too. I strongly doubt that (while I concur that source coding style is often recognizable). More or less a decompiler (when it works properly) attempts to interpret the machine code and translate it into the source. In order to do so, it must have some "templates" corresponding to regognizable "patterns" in the code, so the source derived from the decompilation will reflect these templates and not the "original".

by harry8 9y ago

petre 9y ago

Governments never fire spies. They move them to reserve and pay them good pensions.

threeseed 9y ago

Israel is listed as an "observer" in the Five Eyes alliance. So yes it's quite likely that the US et al have shared assets with them. https://en.wikipedia.org/wiki/Five_Eyes https://en.wikipedia.org/wiki/Five_Eyes

dotancohen 9y ago

I can look at a Git commit and tell you exactly which of my coworkers wrote it without looking at %cn. Code has style, like spoken language has accents. One could argue that e.g. German spy tools copy the American style so that those decompiling it will think it is American. I argue that is a lot harder that it sounds. Code style is much deeper than whether or not to use braces around lone if clauses. The whole way of thinking, layout of data structures, use of getters/setters or properties, breakdown of what goes where and into which classes, breakup of large methods, etc etc etc. These signatures and many more give one a feel for the software's origin. Not proof, but a very solid foundation for suspicion.

shagie 9y ago
You might find De-anonymizing Programmers via Code Stylometry ( http://www.princeton.edu/~aylinc/papers/caliskan-islam_deanonymizing.pdf http://www.princeton.edu/~aylinc/papers/caliskan-islam_deano... ) an interesting read. I suspect that coding style guides are detectable in compiled output too. As an aside, a bit that caught my eye here: > This material is based on work supported by the ARO (U.S. Army Research Office) Grant W911NF-14-1- 0444, the DFG (German Research Foundation) under the project DEVIL (RI 2469/1-1), and AWS in Education Research Grant award.
- jaclaz 9y ago
  >I suspect that coding style guides are detectable in compiled output too. I strongly doubt that (while I concur that source coding style is often recognizable). More or less a decompiler (when it works properly) attempts to interpret the machine code and translate it into the source. In order to do so, it must have some "templates" corresponding to regognizable "patterns" in the code, so the source derived from the decompilation will reflect these templates and not the "original".