Hey, hey, a little more time for teh wit and teh banter… unfortunately my brain is melted so you still don’t get any. SUCKA! I’m in the coldest room in the world which will only get more cold as the search engine reps stonewall the audience with non-answers. I’m kidding. Only not really. I don’t know anymore. What day is it? Mom?
My mom’s not here but our godfather Danny Sullivan is. He’s moderating Arnab Bhattacharjee, Matt Cutts, and Sasi Parthasarathy. Matt just tossed me a package of peanut butter M&Ms from the stage. Hmm, does that mean that link to Google up there is a paid link? Free M&Ms are almost like free Google phones, right? Questions.
Okay, let’s see what they’re talking about. I’m gonna nom quietly on my bribe.
Is Google going to do something with the “you are sending automated” message when you send a lot of queries? We’re doing research.
Matt: You can always stop by the Webmaster Central forum. The motivation behind that is that there are some worms who will spider queries looking for vulnerabilities. They try to throttle those if they look like automated requests. You may have someone in your company doing scraping, etc. Check on what queries are going to Google. They try to balance it so that most people can do all the queries that they need.
What’s the future of Yahoo Site Explorer?
Arnab: They’re going through everything they do for the ecosystem. They’re in deep discussion in how to move that forward.
Sasi: The users comes first. What’s more important for you is what we’re gonna do.
Danny: …so we don’t know. [Ha!]
Is it a problem if my site is in secure https:// format?
Matt: No. The query [paypal] shows they happily show them in the search results. Just pick one or the other.
Google is showing a date for my WordPress blog. How do I get the date out of the results? I don’t like it there.
Matt: Users do like seeing the dates. If you’re showing something from 2006 and they want new info, they’re not going to click on your page. We can infer the content age based on when we first saw it. He’s not sure of any way to NOT make the date show up.
Matt, you mention in a video you’re showing position now. Is that totally rolled out?
Matt: It’s not totally rolled out. The idea is you want to show really fast search results. It turns out you can use AJAX to not just do the search, but to do the search and use that same URL, fetch what the results should be, and fold them into the search results. The problem is not everything browser supports that functionality that well. It’s about speed.
There’s been discussion about PR sculpting – should we be sculpting links?
Matt: The idea is that people want to change the attributes on links to shunt PageRank. They don’t think it’s a good use of your time, but they made a change where that link juice doesn’t get pushed to other pages, it dissipates into the link graph. What’s better is to work on the architecture of your Web site. You can take your best performing products, move them closer to your root, and then they’ll get more juice naturally. You want to surface your best content near the source of your PageRank. In the time you’re spending trying to PR sculpt your site, you could have created some great content. It’s more useful to not think about PR sculpting, but to focus on good architecture and content.
Arnab: All these micro optimizations pretty much don’t help. He says he agrees with Matt.
Does blocking the links cause the remaining links to get more credit?
Matt: It doesn’t. If you have 100 links, your PR on each of those links is roughly 1/100.
Danny: So I can do whatever I want and Google’s going to make it’s own assumptions anyway? [Matt says, ‘right’.]
Arnab: Page importance is just one of many factors the engines look at. It’s really about if you’re relevant to that specific query.
Matt: Don’t get obsessed over PageRank. They don’t take it out of the toolbar because a ton of regular people enjoy looking at PR to see how reputable page is [I would like names of these so imaginary people]. If he found out it was just SEOs obsessing over it, he’d get rid of it.
Cross domain canonicals – we had the canonical tag released last year. What’s the status now?
Sasi: They support it for intra-linking. They are not supporting cross domains for now. Maybe they will in the future if the usage is better.
Arnab: They do support it within a domain. They are exploring algorithmically if it makes sense to do it cross domain.
Matt: He’s been watching sites shoot themselves in the foot with it and accidentally 404 their main pages. They keep the right to say you probably didn’t mean to do what you’re doing.
Arnab: Please do not use rel=canonical as a bandaid. Its better if you don’t have to use it.
What’s up with Google and domain authority. A blank Wikipedia page ranks higher than my content.
Matt: Google typically ranks lower Wikipedia lower than other search engines. [the audience basically laughs. because we don’t believe him.] Wikipedia has done better at not showing stub pages.
Danny: Big sites can fart and it ranks.
Matt: I disagree that big sites can fart and they rank. A lot of the small sites can move faster and do things well. Wikipedia does a good job on SEO. They have good linkage. A typical Wikipedia page is useful and will be returned. We don’t think Wikipedia pages have to be ranked.
Arnab: We don’t do anything to make Wikipedia rank. It’s natural. If you have content and links, you’ll rank.
Sasi: It’s that person’s opinion that they have better content. We may think Wikipedia has better content. [FIGHT!]
We created workarounds because you couldn’t understand AJAX. Now you can. Should we take them down?
Matt: No. If it’s working and white hat, leave it alone. Go make great content.
Danny gives a request. If you’re gonna spider the AJAX content, I want to see that in the cache page.
Arnab (Yahoo rep), do you need to come next year? [OH MY GOD!]
Arnab: I hope that you would invite me next year. A representative from Yahoo search will be here next year. Yahoo will continue to focus on the search experience. Yahoo search is super important. We’re continuing to innovate. Our abstracts will look different. We’ll pull in Yahoo-specific content in the results. Text, image and video will come from Yahoo. Yahoo will keep all of its vertical properties.
Is it beneficial to use CSS to position content higher on the page? Does your important content have to be higher on the page?
Matt: It’s not a huge thing. It you think about how regular people make Web pages (not SEOs), they don’t worry about it. They’re just gonna write their content and have it generate it however it does. You probably don’t want to bury your lead 5 pages out where NO ONE will find it, but it’s not a big issue. A lot of people spend a lot of time worrying about table tricks – if the content is compelling, that’s what’s important.
Arnab: We do understand there are sections of the page that are purely navigational. They can recognize that. They also consider PDF documents.
Is the real-time content part of the main index or are they from different indexes?
Matt: He’s not sure but it’s an excellent question.
Sasi: It depends how fresh the data is.
Why don’t you ban Mahalo?
Matt: Mahalo is an interesting site. It’s either a search engine or a content site. Jason Calacanis has been working on it for awhile. It uses Wiki software so they build individual pages. He thinks of it as a content site. Aaron Wall railed against Mahalo. He showed some bad screen shots, but that’s not all the content on Mahalo. He told Jason that Aaron had made some good points and if you have auto-generated or stub pages and don’t make sure there’s a high bar of quality there, we will take action on our side. All the pages Aaron pointed out now have noindex on them. They try to be zen about sites. If it’s good for their users, they’ll return it. If it’s not, they won’t.
Is it better to have everything on the same domain or should I use subdomains?
Sometimes you dont want a subdomain because you have to think about latency. There are pros and cons.
Is it suspicious to block JS files?
Matt: No, but he’d recommend letting them crawl it because sometimes they can discover new pages.
