What about linking a single sign-on strategy across Gmail, Google and Blogger?
That would be great. Weve had migration issues, of course, since we have a user base that dates back long before we came to Google. Its something we definitely will look at.
Theres been some feedback regarding the Blogger redirect policy. Can you illuminate?
People have noticed—and it isnt new to this release at all—a lot of the links on Blogger.com go through redirects, either on Blooger.com or Google.com, before they send you to the destination. This happens on the list of recently updated blogs, Blogs of Note, and even the Help articles. It also happens on comments: If you put a link on a comment, it gets redirected through Blogger.com.
And the reason for that?
There are actually three reasons for it, which are fairly unusual. One, a lot of the content on Blogger is published with an internal install of Blogger—the intranet version—and that redirects URLs to not reveal the referrer, as a lot of intranet publishing systems would, so that people outside would not see the [internal] names of your sites and pages. We use that same system in the content that gets pushed out to Blogger.com. Its not necessarily on purpose, its just because its an intranet publishing system. That ones fixable, and were going to look at that.
Why would you want to fix this?
The reason it exists is for content that is published on the intranet. In this case were using the intranet publishing system to publish publicly, so theres no reason for it. It doesnt really do anything harmful, it just mainly confuses people, but in that case it does attempt to redirect and slows people down for a second.
But we wouldnt completely stop doing the redirect. The second reason we do redirects, which is fairly important, is that Blogger.com has a really high page rank, partially because of the site and partially because its linked from Google.com. Therefore anything we link on Blogger.com, we flow a lot of page rank to. We dont want there to be a possibility of Blogger blogs having an unfair advantage in the index, just because theyre Blogger blogs. So when people click on ones that are recently updated, we send them to a redirect that Google knows not to flow page rank through.
The third reason for redirects is when they show up in comments, as a comments spam protection. Its similar to the page rank issue, where people spam comments oftentimes just for page rank. That just wont work for Blogger blogs because they will not be assigned page rank by Google.
If you fix the first one, will that have an effect on the second two?
Fixing the first one would mostly mean changing the method so its the second one. The first one is the only one that really affects the user because it slows them down first. The others use a method thats much more transparent. The people will still notice it and were going to explain that—were definitely not doing it to collect data or anything. Its just the reasons I mentioned.
This wouldnt have an effect on applications or other data mining operations like Technorati that could be examining referrals?
It shouldnt. At least in Technoratis case, theyre focused on crawling the contents of the blogs themselves. They certainly count links, and thats a good point. It may affect them if they count links in comments, but thats something that I think they try and avoid. And Movable Type has a similar feature thats optional, on the comments side.