Welcome to the Q&A Forum

There's a few ways to look at this.. but it entirely depends on the domains. If they are all brand domains, then you secure your brand names and redirect all traffic to the main domain. If you're aiming at SEO value, then you gain the authority (if there is any) of all the domains that you redirect to the primary. Why wouldn't you want to redirect 100 domains? Honestly I don't think Google would look at that like a red flag, but again it entirely depends on the domains. If you redirect domains with bad reputation, or former penalties, then you'll automatically inherit the bad reputation. So that can work either way - depends on the domains.

As per Google, depending upon the user-agent you should display the website. So, for Googlebot-mobile only mobile site remains visible and for Googlebot (desktop) only desktop site is shown. Google preaches adding a 301 redirect to mobile version depending upon user-agent detection. Find complete article here: http://googlewebmastercentral.blogspot.com/2011/02/making-websites-mobile-friendly.html This change should solve your current issues with the rankings

hi, In my case indexation took around 2 months after my manual submission to bing index. It was a long time of nothing and one day blow up almost all pages in bing index and on very good positions. Maybe it only need time.

Agree with Marek that you can certainly make URLs a lot firendlier for normal human beings. It can also help keep the urls shorter which can help when people are social sharing. One significant advantage of dropping the extension is that it makes it so much easier if you ever want/need to change the platform on which your site runs. Don't make changes to your URL structure on a whim though. You need to take a careful look at both your internal and  inbound links and plan and redirections you need to put into place. Remember that any change can be risky, so understand the risks and effort required before you start!

I would definitely use schema.org Google and other search are trying to move from rdfa and harmonizing everything to be under schema.org

I think some of the confusion may be due to Google's primarily using IP addresses assigned to their headquarters in Mountain View, California. Google has many (around 20) data centers located outside the US. I recall reading an article whereby at times they used their Mountain View IPs from centers around the world. For security reasons they do not wish the location of all their data centers to be known. I researched this topic before and I was unable to locate any official information from Google. It would only seem reasonable they crawl from all over the world. If they didn't, then a lot of sites which use geo-based targeting for site navigation would not have most of their content indexed. While it's true a sitemap could be used to overcome the issue, many sites do not use sitemaps and they still get indexed.

Artience Girl, the information shared by Shane, Aaron and Lewis is correct. Google wants to see the same page as it would be shown to a user under the same circumstances. If Google is crawling your page from San Jose California, then they want to see what a user from San Jose would see. If they decide to later crawl your site from their center in London, they want to see your site as it would be seen by a London user. The geo-targeting redirects you are presently doing are fine. If you were to write any code which says to always show the Google crawler the US version of your site, then that tactic would be defined as cloaking. Any time you write code to specifically identify a crawler and show it different content, then you are cloaking. It seems you are a bit uncomfortable with the answers so let me set you at ease by sharing a Matt Cutts response to your question: http://www.youtube.com/watch?v=GFf1gwr6HJw

I'm seeing another place where Google doesn't reference the Gdata in code.google.com. Check out http://code.google.com/apis/storage/docs/reference-headers.html#etag. In API changelog documentation for their calendar http://code.google.com/apis/calendar/changelog.html they say "If a client wishes to retain version 2.0 error behavior, then it can either send a request with the Gdata-Version: 2.0 HTTP header or append ?v=2.0 to request URIs. In this case, no error will be provided if an outdated updated-min parameter is provided or if a gCal:eid conflict occurs when inserting a recurrence exception. However, the server response docuement will still conform to version 2.1 and contain a Gdata-Version: 2.1 header." It seems like you're fine, as most of the references I'm seeing Google made don't include that GData info.

Thanks for sharing your tip.