> Got bored, did some digging in their binary, they built using some of our our svn libraries (I know it’s our libs since the paths embedded for some of the source files are local to my computer :slight_smile: ) which makes it easy to track down
> they ship openimageio 1.7.15 which means the code is likely from somewhere in between 2017-06-04 and 2018-08-27 which is indeed around the 2.79 time frame.
> It would have been nice if they kept the code opensource, so it could get all the improvements blender makes easily, but they seemingly chose not to go that way. which is somewhat strange since they clearly lifted the GPL licensed bf_blenlib [1] so the closed source nature of eyesight is odd to say the least.
> [1] the executable contains strings like <inline>Error! Could not get the Windows Directory - Defaulting to Blender installation Dir! , Error! Could not get the Windows Directory - Defaulting to first valid drive! Path might be invalid!</inline> and <inline>BLI_dynstr_append text too long or format error.</inline> that can only could have come from bf_blenlib
UPDATE: Additional context from Orion Pobursky[0], the LDraw.org's webmaster:
> The https://Stud.io has been historically bad with attribution of their sources. LDraw had to contact them directly to get them to acknowledge use in their about page and most average users still don't know that the Stud.io's library is back by LDraw.[1]
Is there precedent for that holding in court? Saying something like "I put this unusual string in my software & it is also in theirs" seems pretty useless from a legal perspective.
>In cartography, a trap street is a fictitious entry in the form of a misrepresented street on a map, often outside the area the map nominally covers, for the purpose of "trapping" potential plagiarists of the map
I thought the same thing prior to seeing your comment -- yes, these are basically the first version of watermarks... so cartographers could silently 'sign' their work and see who steals it.
That would be nice, but the share of successfully litigated copyright cases that have video evidence of the act of copying, or something similar, is essentially zero.
Heck, criminal jury convictions, despite the much higher standard of proof, very often don’t have “video of the act being committed” kind of evidence.
IIRC there was a Mac program that proved it was being infringed on because the knockoff software implemented the same Easter egg when a certain series of hot keys was pressed
On the contrary, that seems more likely to be understood by a non-technical audience than any amount of decompiler output. Doesn't replace the latter, but it certainly removes a reasonable doubt.
Yeah if the string is something like "I sidewndr46 am the copyright holder for this software", I can't really see how it would be possible for a vendor to say they just happened to put that in their own independently created product.
Don't some map makers put ghost towns or roads that don't actually exist in their own maps as a sort of fingerprinting method? Unless you actually visit that particular spot, one might not know and inadvertently copy the proprietary map data,
"Unusual string" seems a lot more innocuous than unusual string referring to the product as another copywrited work by name which contains the same unusual string.
I'm not sure of the actual prevalence of police using fingerprints to track down theft. It is however accepted as a piece of evidence in various criminal courts, so one would expect it to hold up in a non-criminal court as well.
The difference is I didn't choose my fingerprints nor did the police. It's also really, really hard to get rid of them OR to have someone else with the same fingerprints as mine.
What I meant by faking fingerprints was creating another living human being with my same fingerprints. It's obviously trivial to create fingerprint evidence.
It is not just that you can create fingerprint evidence.
It is that you can create fingerprint evidence in normal interactions with fingerprint systems in ways that leave no other evidence, and are unlikely to draw scrutiny from nearby humans.
From comment #3 in this discussion:
> Got bored, did some digging in their binary, they built using some of our our svn libraries (I know it’s our libs since the paths embedded for some of the source files are local to my computer :slight_smile: ) which makes it easy to track down
> they ship openimageio 1.7.15 which means the code is likely from somewhere in between 2017-06-04 and 2018-08-27 which is indeed around the 2.79 time frame.
> It would have been nice if they kept the code opensource, so it could get all the improvements blender makes easily, but they seemingly chose not to go that way. which is somewhat strange since they clearly lifted the GPL licensed bf_blenlib [1] so the closed source nature of eyesight is odd to say the least.
> [1] the executable contains strings like <inline>Error! Could not get the Windows Directory - Defaulting to Blender installation Dir! , Error! Could not get the Windows Directory - Defaulting to first valid drive! Path might be invalid!</inline> and <inline>BLI_dynstr_append text too long or format error.</inline> that can only could have come from bf_blenlib