Dupe of a dupe from yesterday: <a href="https://news.ycombinator.com/item?id=14285045" rel="nofollow">https://news.ycombinator.com/item?id=14285045</a>
You'd think after 11 years of development, HN would be able to at least work out that:<p>a) <a href="http://googleresearch.blogspot.com/2016/05/announcing-syntaxnet-worlds-most.html?m=1" rel="nofollow">http://googleresearch.blogspot.com/2016/05/announcing-syntax...</a><p>b) <a href="https://research.googleblog.com/2016/05/announcing-syntaxnet-worlds-most.html?m=1" rel="nofollow">https://research.googleblog.com/2016/05/announcing-syntaxnet...</a><p>c) <a href="http://googleresearch.blogspot.com/2016/05/announcing-syntaxnet-worlds-most.html?m=1" rel="nofollow">http://googleresearch.blogspot.com/2016/05/announcing-syntax...</a><p>... the URL and title for a==c AND the titles for a) == b) == c) are the same or similar. Make the ^dupe^ post irrelevant.
Déjà vu…this is broadly the same as an idea I had around 2010, and even mentioned to a Google Translate engineer when interviewing there in 2011. Nice to see they finally did something along these lines…I can fool myself into feeling smug about it, and never had to do any of the real work!