That part made me kind of angry -- "so-called" makes the writer sound like they know something we don't, like "subversion" is some cool trick that smart people know.
"so-called" here is a way of distinguishing the use of "subversion" as a name/description distinction (somewhat similar to a use/mention distinction, but not the right way of making that distinction, IMO, as correctly using "Subversion" as a proper noun would be better, though) as the common noun "subversion" has meaning, but the word "subversion" was not being used for that meaning, only as the name of a thing.
Would it have? On my last day at work at Google, before I am slated to start working in the Bing group at Microsoft, I mail myself a copy of the PageRank source code. Does it matter what VCS I use?
Technically no. In a court description or in a press article, both addressed to people who are not technical, the name of the VCS is important.
Consider the extreme case where the VCS was called "theft-assistance" for example.
How does "He copied the code from a theft-assistance repository" would sound to a jury that doesn't know what a VCS is and that theft-assistance is just a name?
If it was something generic and inoffensive, like Git, it would be OK, but "subversion" has a ...subversive undertone.
In a previous video posted on HN, a police officer explains how they would use a concession made by a honest humble person during an interview to boost their conviction rate: "Sure, I never like the guy. But I would never do anything malicious against anyone or even animals, especially not theft or murder which is totally against my conscience." would become big uppercase red letters on a videoprojector: "I NEVER LIKE THE GUY". Be sure you nail every emotional aspect to convince a jury.
Most people have no idea what VCS is. To a lay person, "a so called 'mercurial repository'" would sound far less sinister than "a so called 'subversion repository'". The reality is that the repo probably had little to do with subverting anything, and was just a convenient way to transfer code.
I'm referring to the fact that no matter how damning the evidence is on its own, it probably seems even more damning to a layperson who doesn't know the name "subversion" is a pun and not something subversive.
Except the code copied by Aleynikov wasn't anything like PageRank. It was like some Go library that fetches the pages, and one that was primarily open source code to begin with.
This was some initial (mis?)-information about the case, but at least according to the Second Circuit:
> In addition to proprietary source code, Aleynikov
also transferred some open source software licensed for use
by the public that was mixed in with Goldman’s proprietary
code. However, a substantially greater number of the
uploaded files contained proprietary code than had open
source software.
Were the files source code or were they some sort of transaction logs, generated records, and what have you? Or were they various forks/branches of mostly the same thing intended for testing? If you don't know then you can't assess their relative "value".
One wonders whether this would have seemed less nefarious if he'd used a VCS that wasn't called "subversion." :)