Internet Census 2012 Data Examined: Authentic, But Chaotic and Unethical

An anonymous reader writes "A team of researchers at the TU Berlin and RWTH Aachen presented an analysis of the Internet Census 2012 data set (here's the PDF) in the July edition of the ACM Sigcomm Computer Communication Review journal. After its release on March 17, 2013 by an anonymous author, the Internet Census data created an immediate media buzz, mainly due to its unethical data collection methodology that exploited default passwords to form the Carna botnet. The now published analysis suggests that the released data set is authentic and not faked, but also reveals a rather chaotic picture. The Census suffers from a number of methodological flaws and also lacks meta-data information, which renders the data unusable for many further analyses. As a result, the researchers have not been able to verify several claims that the anonymous author(s) made in the published Internet Census report. The researchers also point to similar but legal efforts measuring the Internet and remark that the illegally measured Internet Census 2012 is not only unethical but might have been overrated by the press."

Comments

Why not just get the metadata from the NSA?

[WillAffleckUW]

They illegally and unconstitutionally collect it anyway, especially on Americans, and give a copy of the feed illegally and unconstitutionally to the CIA and GCHQ.

Among others.

Unethical

[maevius]

Unethical? Whatever.
Having read the original "census", it was a cool hack and no harm was done, nothing more. I'm pretty sure he/they didn't go for vigorous scientific process when this was done.

I wonder

[NotInHere]

Why is using idle machines of other people (he's used only machines whose load was under a certain threshold), more unethic than to torment and kill mice in the name of science? I don't think that, when used responsible, latter is unethic, but I wonder why do they put things above biological life?

Biased, much ?

[aepervius]

We do not "torment" and kill mice gratuitiously, a choice of word which certainly show quite inherent bias here. Usually you have to go thru an ethical comitee for animal experimentation (although the barrier is lower for lab mouse). Furthermore most of those animal experimentation have a clear goal to help develop cure or model for the human health. If you can't differentiate that from people misusing the computer of others, then I can't help you.

OS Fingerprints!

[d33tah]

Apparently the researchers didn't analyze OS fingerprints at all. There is some metadata that the original researcher(s) forgot to remove (as well as a lot more mess). Service fingerprints are interesting as well. I did a lot of research on this data set and I have to say that while messy, this is also a really amazing data set. This article is IMHO biased.