Update changelog for 0.14.1
(cherry picked from commit aaf8f870cb242750e842d21ac7ae2f72876f6346)
Update readme for 0.14.1
(cherry picked from commit c2dde8df0b365f341416201d1794b0c5e96d1951)
0.14.1 blog post
(cherry picked from commit 6e2450e8f29f336ffe9c80d067c8df816957a349)
Update website for 0.14.1
(cherry picked from commit f0782e2b2ec79af37987eead2464ac9f303183f5)
Just seen on Stack Overflow a question that suggests people may be
confused if they run across some of the old syntax in one of our old
blog posts:
http://stackoverflow.com/questions/33261751/why-does-relay-ql-result-in-unterminated-string-constant-error/
Ideally people would consult the (up-to-date) documentation instead, and
we don't want to have to maintain old blog posts, but in this case the
cost of updating the syntax was low so I just went ahead and did it.
Or the alternative:
`REST is an acronym for Representational State Transfer, which is an architectural style rather than a formal protocol`.
Sorry , I'am not a native speaker:)
Update links to use https:// where it is supported. There's probably a lot
more that could be fixed, but these are the core ones I found (especially
the download links in order to prevent MITM attacks). Note that there are
some fb.me links that will redirect to http:// even while accessed over
https://, but this seemed like the best way to fix those for now.
NOTE: Only non-third-party files were modified. There are references to
http:// URLs in vendored/third-party files, but seems appropriate to fix
upstream for those rather than editing the files.
Also, copy one image locally to the blog, as it was hotlinking to a site
that did not support https://.
Last, use youtube-nocookie.com instead of youtube.com for video embeds,
as the former doesn't try to set a cookie on load (privacy enhancement).