Everything you wanted to know about crawling, but didn't know where to ask

Local Search

(Including Importance Metrics and Link Merging)

Everything you wanted to knowabout Crawling*

*But Didn't Know Where to Ask

Agile SEO Meetup – South JerseyMonday, September 10, 20127:00 PM to 9:00 PM

Bill SlawskiWebimax@bill_slawski

In the Early Days of the Web, there was an invasion

Robots

Spiders

Via Thomas Shahan - http://www.flickr.com/photos/opoterser/

Crawlers

Invaded pages across the World Wide Web

The Robots Mailing List was formed to solve the problem!

Led by a young Martijn Koster, they developed the Robots.txt protocol

Which Asked Robots to be Polite

And Not Melt Down Internet Servers

A student at Stanford named Lawrence Page went on to co-author a paper on how robots might Crawl web pages to index important pages first.

http://ilpubs.stanford.edu:8090/347/1/1998-51.pdf

<<Insert Subliminal Advertisement Here>>

Important Web Pages1. Contain words similar to a query that starts the crawl2. Have a high backlink count3. Have a high PageRank4. Have a high forward link count5. Are in or are close to the root directory for sites

Image via Fir0002/Flagstaffotos under http://commons.wikimedia.org/wiki/Commons:GNU_Free_Documentation_License_1.2

So most crawlers will not only be Polite, but they will also hunt down

important pages first

Search Engines filed patents on how they might crawl and collect content found on Web pages, including collectingURLs and Anchor Text associated with them.

<a href=“http://www.hungryrobots.com”>Feed Me</a>

http://patft1.uspto.gov/netacgi/nph-Parser?patentnumber=7308643

Also, in one embodiment, the robots are configured to not follow "permanent redirects". Thus, when a robot encounters a URL that is permanently redirected to another URL, the robot does not automatically retrieve the document at the target address of the permanent redirect.

“Use a text browser such as Lynx to examine your site, because most search engine spiders see your site much as Lynx would. If fancy features such as JavaScript, cookies, session IDs, frames, DHTML, or Flash keep you from seeing all of your site in a text browser, then search engine spiders may have trouble crawling your site.”*

*Google Webmaster Guidelines - http://support.google.com/webmasters/bin/answer.py?hl=en&answer=35769

Google’s Webmaster Guidelines make crawlers look prettyunsophisticated, and incapable of much more than the simpleLynx browser…

…But we have signs that crawlers can be smarter than that,and Microsoft introduced a Vision-based Page Segmentation Algorithm in 2003. Both Google and Yahoo have also publishedpatents and papers that describe smarter crawlers. IBM filed a patentfor a crawler in 2000 that is smarter than most browsers today.

VIPS: a Vision-based Page Segmentation Algorithm - http://research.microsoft.com/apps/pubs/default.aspx?id=70027

http://patft1.uspto.gov/netacgi/nph-Parser?patentnumber=7519902

Link Merging

Web Site Structure Analysis - http://patft1.uspto.gov/netacgi/nph-Parser?patentnumber=7861151

• S-nodes – Structural Link Blocks - organizational and navigational link blocks; Repeated across pages with the same layout and showing the organization of the site. They are often lists of links that don’t usually contain other content elements such as text.

• C-nodes – Content link blocks, grouped together by some kind of content association, such as relating to the same topic or sub-topic. These blocks usually point to information resources and aren’t likely to be repeated across more than one page.

• I-nodes – Isolated links, which are links on a page that aren’t part of a link group, may be only loosely related to each other, by virtue of something like their appearing together within the same paragraph of text. Each link on a page might be considered an individual i-node, or they might be grouped together by page as an i-node.

Crawling and Self Help

Canonical = Best!

There can be only one:http://example.comhttp://www.example.comhttp://example.com/http://www.example.com/https://example.comhttps://www.example.comhttps://example.com/https://www.example.com/http://example.com/index.htmhttp://www.example.com/index.htmhttps://example.com/index.htmhttps://www.example.com/index.htmhttp://example.com/INDEX.htmhttp://www.example.com/INDEX.htmhttps://example.com/INDEX.htmhttps://www.example.com/INDEX.htmhttp://example.com/Index.htmhttp://www.example.com/Index.htmhttps://example.com/Index.htmhttps://www.example.com/Index.htm

Canonical Link Element

<link rel="canonical" href="http://example.com/page.html"/>

Rel=“prev” & rel=“next”On the first page, http://www.example.com/article?story=abc&page=1,

<link rel="next" href="http://www.example.com/article?story=abc&page=2" />

On the second page, http://www.example.com/article?story=abc&page=2:

<link rel="prev" href="http://www.example.com/article?story=abc&page=1" /> <link rel="next" href="http://www.example.com/article?story=abc&page=3" />

On the third page, http://www.example.com/article?story=abc&page=3

<link rel="prev" href="http://www.example.com/article?story=abc&page=2" /> <link rel="next" href="http://www.example.com/article?story=abc&page=4" />

And on the last page, http://www.example.com/article?story=abc&page=4:

<link rel="prev" href="http://www.example.com/article?story=abc&page=3" />

Paginated Product Pages

Paginated Article Pages

View All Pages

Option 1

• Normal Prev/Next sequence • Self Referential Canonicals (point to their Own URL• Noindex meta element on View All page

Option 2

• Normal Prev/Next Sequence• Canonicals (all pages use the view-all page URL)

http://googlewebmastercentral.blogspot.com/2011/09/view-all-in-search-results.html

Rel=“hreflang”

Rel=“hreflang”HTML link element.

In the HTML <head> section of http://www.example.com/, add a link element pointing to the Spanish version of that webpage at http://es.example.com/, like this:

<link rel="alternate" hreflang="es" href="http://es.example.com/" />

HTTP header.

If you publish non-HTML files (like PDFs), you can use an HTTP header to indicate a different language version of a URL:

Link: <http://es.example.com/>; rel="alternate"; hreflang="es"

Sitemap.

Instead of using markup, you can submit language version information in a Sitemap.

Rel=“hreflang” XML Sitemap<?xml version="1.0" encoding="UTF-8"?><urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9" xmlns:xhtml="http://www.w3.org/1999/xhtml"> <url> <loc>http://www.example.com/english/</loc> <xhtml:link rel="alternate" hreflang="de" href="http://www.example.com/deutsch/" /> <xhtml:link rel="alternate" hreflang="de-ch" href="http://www.example.com/schweiz-deutsch/" /> <xhtml:link rel="alternate" hreflang="en" href="http://www.example.com/english/" /> </url>

XML Sitemap

XML Sitemap

• Use Canonical links• Remove 404s• Don’t set priority past 1 week• If more than 50,000 URLs, use multiple Sitemaps and a site index• Validate with an XML Sitemap Validator• Include a Sitemap statement in robots.txt

http://www.sitemaps.org/

Next, we study which of the two crawl systems, Sitemaps and Discovery, sees URLs first. We conduct this test over a dataset consisting of over five billion URLs that were seen by both systems.

According to the most recent statistics at the time of the writing, 78% of these URLs were seen by Sitemaps first, compared to 22% that were seen through Discovery first.

Crawling vs. XML

Sitemaps: Above and Beyond the Crawl of Duty – http://www.shuri.org/publications/www2009_sitemaps.pdf

Crawling Social Media

Ranking of Search Results based on Microblog data - http://appft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PG01&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.html&r=1&f=G&l=50&s1=%2220110246457%22.PGNR.&OS=DN/20110246457&RS=DN/20110246457

Questions?

Bill SlawskiWebimax

@bill_slawski