A setback for Google against Oracle

Posted May 16, 2014 15:17 UTC (Fri) by dberkholz (guest, #23346)
In reply to: A setback for Google against Oracle by dlang
Parent article: A setback for Google against Oracle

You might like to read this post of David Pollak's on the topic. In it he points out that this decision didn't actually prohibit reverse engineering to create a compatible API. He's got a few others exploring a variety of implications of the ruling.

A setback for Google against Oracle

Posted May 18, 2014 1:03 UTC (Sun) by dlang (guest, #313) [Link] (2 responses)

I suggest you read the rest of his posts, they go on about how horrible this is.

Besides, how can you possible prove that you reverse engineered the function names and parameters of a library?

A setback for Google against Oracle

Posted May 18, 2014 2:06 UTC (Sun) by dberkholz (guest, #23346) [Link] (1 responses)

Presumably things like not copying comments and ordering within a header.

A setback for Google against Oracle

Posted May 18, 2014 2:27 UTC (Sun) by dlang (guest, #313) [Link]

from what I saw they didn't copy comments, the new comments were very similar to the old ones, but that's because they're describing the same functionality.

Google wasn't accused of literally copying the comments, but of trying to file the serial numbers off by rewriting them. (well, early on there was accusation of direct copying of comments, but that was eliminated during the trial

As for ordering within a header. My understanding is that most of them are in alphabetical order, possibly grouped by function family (although with Java's long naming schemes, the two are pretty much the same)

If simply re-ordering the functions within the header was enough to avoid it being copying, that's just a few minutes of scripting, running software won't care

But it will be much harder to be sure that you haven't missed anything if you re-order things that way.