Tuesday, January 22, 2008<\/p>\n
\n Last month, several of us with Webmaster Central \n Listening to the Q&A, I was pleased to hear the major search engines agreeing on best \n Several webmasters asked, “Is it necessary to keep the verification \n What’s the difference between using subdomains and subdirectories? When it comes to Google, there \n Webmasters asked about the difference between how Google interprets underscores and dashes in \n We were also asked if it is useful to have relevant keywords in URLs. It’s always a good idea to \n We got a question about changing a site’s IP address, and provided a few steps you can take as a \n A few webmasters were concerned that proxy services are being indexed with copies of their \n It was great talking with webmasters at the conference—we hope those of you unable to join \nUpdate<\/b>:
\n raked in the “good times” jackpot<\/a>
\n at PubCon Vegas 2007. We realize not all of you could join us, so instead of returning home with
\n fuzzy dice for everyone, we’ve got souvenir
\n conference notes<\/a>.\n<\/p>\n
\n practices for many webmaster issues. In fact, the presentations in the duplicate content session
\n were mostly, well, duplicate. When I wasn’t sitting in on one of the many valuable sessions, I was
\n chatting with webmasters either at the Google booth, or at Google’s “Meet the Engineers” event. It
\n was exciting to hear from so many different webmasters, and to help them with Google-related
\n issues. Here are a few things that were on the minds of webmasters, along with our responses:\n<\/p>\nSite Verification Files and Meta Tags<\/h2>\n
meta<\/code> tag or HTML file in place
\n to remain a verified owner in Webmaster Tools?” The answer is yes, you should keep your
\n verification file or meta<\/code> tag live to maintain your status as a verified owner. These
\n verification codes are used to control who has access to the owner-specific tools for your site
\n in Webmaster Tools. To ensure that only current owners of a site are verified, we periodically
\n re-check to see if the verification code is in place, and if it is not, you will get unverified
\n for that site. While we’re on the topic:\n<\/p>\nSite Verification Best Practices<\/h3>\n
\n
\n have each person verify the site with their own account, rather than using a shared login.
\n That way, as people come and go, you can control the access appropriately by adding or
\n removing verification files or meta<\/code> tags for each account.\n <\/li>\n
\n so you can easily control access later. If you lose track, you can always use the “Manage
\n site verification” option in Webmaster Tools, which allows you to force all site owners to
\n reverify their accounts.\n <\/li>\n<\/ul>\nSubdomains vs. Subdirectories<\/h3>\n
\n aren’t major differences between the two, so when you’re making that decision, do what works for
\n you and your visitors. Following PubCon, our very own Matt Cutts
\n outlined many of the key issues<\/a>
\n in a post on his personal blog. In addition to those considerations, if you use Webmaster Tools
\n (which we hope you do!), keep in mind that you’ll automatically be verified for deeper
\n subdirectories of any sites you’ve verified, but subdomains need to be verified separately.\n<\/p>\nUnderscores vs. Dashes<\/h3>\n
\n URLs. In general, we break words on punctuation, so if you use punctuation as separators, you’re
\n providing Google a useful signal for parsing your URLs. Currently, dashes in URLs are
\n consistently treated as separators while underscores are not. Keep in mind our technology is
\n constantly improving, so this distinction between underscores and dashes may decrease over time.
\n Even without punctuation, there’s a good chance we’ll be able to figure out that
\n bigleopard.html<\/code> is about a “big leopard<\/span>” and not a
\n “bigle opard<\/span>“. While using separators is a good practice, it’s
\n likely unnecessary to place a high priority on changing your existing URLs just to convert
\n underscores to dashes.\n<\/p>\nKeywords in URLs<\/h3>\n
\n be descriptive across your site, with titles,
\n ALT attributes<\/a>, and yes, even
\n URLs, as they can be useful signals for users and search engines. This can be especially true
\n with image files, which otherwise may not have any text for a search engine to consider. Imagine
\n you’ve taken a picture of your cat asleep on the sofa. Your digital camera will likely name it
\n something like IMG_2937.jpg<\/code>. Not exactly the most descriptive name. So unless your
\n cat really looks like an IMG_2937<\/code>, consider changing the filename to something more
\n relevant, like adorable-kitten.jpg. And, if you have a post about your favorite cat names, it’s
\n much easier to guess that a URL ending in my-favorite-cat-names would be the relevant page,
\n rather than a URL ending in postid=8652<\/code>. For more information regarding issues with
\n how Google understands your content, check out our new
\n content analysis<\/a>
\n feature in Webmaster Tools, as well as
\n our post on the URL suggestions feature<\/a>
\n of the new Google Toolbar.\n<\/p>\nMoving to a new IP address<\/h3>\n
\n webmaster to make sure things go smoothly. Here’s what you can do:\n<\/p>\n\n
\n minutes (300 seconds). This will tell web browsers to re-check the IP address for your site
\n every five minutes.\n <\/li>\n
\n address.\n <\/li>\n
\n sure it’s really Googlebot who’s visiting, you can verify Googlebot by following
\n these instructions<\/a>. You can then log
\n into Webmaster Tools and monitor any crawl errors. Once Googlebot is happily crawling on the
\n new IP address, you should be all set as far as Google is concerned.\n <\/li>\n
\n to your old IP address before shutting it down.\n <\/li>\n<\/ol>\nProxies<\/h3>\n
\n content. While it’s often possible to find duplicate copies of your content in our results if you
\n look hard enough, the original source is most likely going to be ranked higher than a proxy copy.
\n However, if you find this not to be the case, please drop us some URLs in the Webmaster Help
\n Group. There are many Googlers including myself who monitor this group and escalate issues
\n appropriately.\n<\/p>\n
\n us found this post useful. If you want to continue to talk shop with me, other Googlers, and your
\n fellow webmasters, join the follow-up conversation in the
\n Webmaster Help Group<\/a>.\n<\/p>\n
\n Additional PubCon notes<\/a>
\n from Jonathan Simon are available in our discussion group.\n<\/p>\n