News for hackers. Stuff that matters.
Re:GET, POST, and safely surfacing more of the web by franze on Wednesday November 02, 2011 @05:14AM (#3185915)
ok, i'm highlighting two awesome pieces of information in that specific blogpost. both of them aren't new but this is the first time they were ever officially confirmed by google.

  Google begins indexing yummy-sundae.html and, as a part of 
  this process, decides to attempt to render the page to  
  better understand its content and/or generate the Instant 
  Preview.
this is the first time ever that google confirms that they are internally rendering the page - not only for preview purposes - but also for indexing purposes.

  >Remember to include important content (i.e., the content you’d like 
  >indexed) as text, visible directly on the page and without 
  >requiring user-action to display. Most search engines are text-
  >based and generally work best with text-based content. We’re always 
  >improving our ability to crawl and index content published in a 
  >variety of ways, but it remains a good practice to use text for 
  >important information.
first time ever they advise - directly - against "user action dependent hidden content"
Re:GET, POST, and safely surfacing more of the web by antifuchs on Wednesday November 02, 2011 @12:22PM (#3187514)
One more nice piece of infomration (I'm sure it was out there before, but the consequences are now more interesting): If .js files referenced in your page are disallowed by robots.txt, googlebot won't load them.
Re:GET, POST, and safely surfacing more of the web by andybak on Wednesday November 02, 2011 @06:17AM (#3186039)
This strongly supports this: http://news.ycombinator.com/item?id=3182579

I wonder if another implication of Google actually rendering pages is that SEO is going to have to focus much more heavily on visible page content rather than hidden or quietly shuffled out-of-sight content.

Re:GET, POST, and safely surfacing more of the web by franze on Wednesday November 02, 2011 @06:50AM (#3186104)
well, there is the implication that we will see HN posts a la "Google deleted my content!"

google says they will only post POST request that are automatically triggered onpageload and will have no other side effects other than (fetching & ) rendering content.

ok, lets say google does 1000 000 000 POST requests and gets it 99.999% right. this means that there will be 1 000 000 occurrences where google actually will trigger some unintended side-effects.

"At scale everything breaks" (c) Urs Hölzle http://www.zdnet.co.uk/news/cloud/2011/06/22/google-at-scale...

Re:GET, POST, and safely surfacing more of the web by rwmj on Wednesday November 02, 2011 @08:00AM (#3186269)
If you've got a website and look at the logs, you'll notice dozens of badly behaved bots, spam farms etc that follow any and every get or post link, fill in forms with spam and submit them, and ignore robots.txt. If your site is deleting comments based on an unauthenticated POST, you've already got a broken site.
Re:GET, POST, and safely surfacing more of the web by davedx on Wednesday November 02, 2011 @07:24AM (#3186173)
If you have POST actions that are not behind any kind of authorization or validation step then you've already shot yourself in the foot.
Re:GET, POST, and safely surfacing more of the web by martian on Wednesday November 02, 2011 @07:24AM (#3186172)
This seems relevant (from HN a couple weeks ago):

http://www.thumbtack.com/engineering/googlebot-makes-post-re...