3 ms·

I can tell you that the comment section on FT.com is positively trolled by at least a few accounts. They seem to have allocated some of the smarter ones to the

by sbt 12y ago

I can tell you that the comment section on FT.com is positively trolled by at least a few accounts. They seem to have allocated some of the smarter ones to the site, but they are easily identified because.. * They sometimes drop determiners in English. (A pattern I've noticed). * They post a lot of comments. * They fill up the comment section early in the morning (Europe) and then nothing. * Rampant quoting (especially zero hedge), like someone spent their whole day researching the issue. * Assumptions in arguments are official Russian talking points. Now, how do I distinguish these guys from just gold buying Max Keyser followers? Mainly the time that seems to have gone into it. If I were to post as prolifically I would need to dedicate at least several hours to this every day.

Procedural 12y ago
Huh. So the situation is worse than I expected. The sites like the one you mentioned should close their comment section when it's morning in Europe and open it at night when their work day is over. I hope people won't judge other Russians just because of these trolls...
- sbt 12y ago
  Yes, I have counted at least five accounts on FT.com that I am confident are professional trolls. The upshot of all this trolling is that all legitimate support for Russia is effectively ignored. The Kremlin troll phenomenon is now so widespread that it is working against whatever intentions they could have had. Basically, anyone who holds a pro-Russian opinion is now either dismissed as a troll or a victim of propaganda. Unfortunately, this is just as it has to be. They have destroyed any sensible debate and have rendered themselves and their followers labelled.
Polunocnica 12y ago
This comment is actually quite interesting. For a while, I've stuck to the Wikileaks suggested formula of "Scientific Journalism" [https://en.wikipedia.org/wiki/Scientific_journalism https://en.wikipedia.org/wiki/Scientific_journalism] and I've seen the progression of State Actors (both automated and manned) across the www. What's telling that I did a CTRL+F for the term - and it's not mentioned. Added to that, even mature news sources (such as the Guardian or Telegraph in the UK) have only ever dabbled at it - for instance, the Guardian provides links within its own articles, but the links are entirely self referential. i.e. recursive back into their own site. Even with actual science articles, actually getting a link to the paper [pay-walled or otherwise] is rare. I will say that certain writers for the Guardian (Monbiot recently on soil), however much their bias, actually source links to multiple first / second tier sources, which is a good thing. The upshot of this is: there [b]is[/b] a solution, and you can often spot the genuine comments by virtue of first and second tier links (wikipedia being second tier). The actual real issue (again, not mentioned in the comments here) is to break the media consumption mold whereby people read comments as personal anecdotes / opinions. It's a form of conditioning around since forever [the TV turn to the camera, eye contact, personal little look]. What might be a fun project for HN viewers would be an independent (eff?) project that pattern matched text and caught similarities. Probably using a modified code base from a commercial plagiarism detectors. [https://en.wikipedia.org/wiki/Plagiarism_detection https://en.wikipedia.org/wiki/Plagiarism_detection] Note: the US at least has spent a lot of money on their own versions. As probably have the Russians / Chinese: you can be fairly sure comments are chucked through an automated detector to see the (likely) source: yours, theirs, randoms, bots or the 'Other'. [Let's see if this comment gets flagged by evul bots!]
- Polunocnica 12y ago
  Edit is apparently not working: Data, Text, and Image Mining—Analysis of data stream in real time as well as cluster analysis and their applications to data, text, and image mining are important tools for anomaly detections in the global war against terrorism. New and unifying methodologies are needed in order to provide efficient search for patterns or meaning from the analysis of usually huge data sets that consist of multivariate measurements. Developments of mathematical theory for data, text, and image mining techniques are also highly desirable. http://www.arl.army.mil/www/default.cfm?page=185 http://www.arl.army.mil/www/default.cfm?page=185

by sbt 12y ago

Procedural 12y ago

Huh. So the situation is worse than I expected. The sites like the one you mentioned should close their comment section when it's morning in Europe and open it at night when their work day is over. I hope people won't judge other Russians just because of these trolls...

sbt 12y ago
Yes, I have counted at least five accounts on FT.com that I am confident are professional trolls. The upshot of all this trolling is that all legitimate support for Russia is effectively ignored. The Kremlin troll phenomenon is now so widespread that it is working against whatever intentions they could have had. Basically, anyone who holds a pro-Russian opinion is now either dismissed as a troll or a victim of propaganda. Unfortunately, this is just as it has to be. They have destroyed any sensible debate and have rendered themselves and their followers labelled.

Polunocnica 12y ago

This comment is actually quite interesting. For a while, I've stuck to the Wikileaks suggested formula of "Scientific Journalism" [https://en.wikipedia.org/wiki/Scientific_journalism https://en.wikipedia.org/wiki/Scientific_journalism] and I've seen the progression of State Actors (both automated and manned) across the www. What's telling that I did a CTRL+F for the term - and it's not mentioned. Added to that, even mature news sources (such as the Guardian or Telegraph in the UK) have only ever dabbled at it - for instance, the Guardian provides links within its own articles, but the links are entirely self referential. i.e. recursive back into their own site. Even with actual science articles, actually getting a link to the paper [pay-walled or otherwise] is rare. I will say that certain writers for the Guardian (Monbiot recently on soil), however much their bias, actually source links to multiple first / second tier sources, which is a good thing. The upshot of this is: there [b]is[/b] a solution, and you can often spot the genuine comments by virtue of first and second tier links (wikipedia being second tier). The actual real issue (again, not mentioned in the comments here) is to break the media consumption mold whereby people read comments as personal anecdotes / opinions. It's a form of conditioning around since forever [the TV turn to the camera, eye contact, personal little look]. What might be a fun project for HN viewers would be an independent (eff?) project that pattern matched text and caught similarities. Probably using a modified code base from a commercial plagiarism detectors. [https://en.wikipedia.org/wiki/Plagiarism_detection https://en.wikipedia.org/wiki/Plagiarism_detection] Note: the US at least has spent a lot of money on their own versions. As probably have the Russians / Chinese: you can be fairly sure comments are chucked through an automated detector to see the (likely) source: yours, theirs, randoms, bots or the 'Other'. [Let's see if this comment gets flagged by evul bots!]

Polunocnica 12y ago
Edit is apparently not working: Data, Text, and Image Mining—Analysis of data stream in real time as well as cluster analysis and their applications to data, text, and image mining are important tools for anomaly detections in the global war against terrorism. New and unifying methodologies are needed in order to provide efficient search for patterns or meaning from the analysis of usually huge data sets that consist of multivariate measurements. Developments of mathematical theory for data, text, and image mining techniques are also highly desirable. http://www.arl.army.mil/www/default.cfm?page=185 http://www.arl.army.mil/www/default.cfm?page=185