Good Google search tips

Via Sarah Marshall. And she has a link to even more "site" operator tricks.

AP v. Meltwater - I'm not betting for AP on the appeal

AP won a "big" victory against news aggregator Meltwater yesterday.

But I'm not betting against Meltwater on appeal when it comes to the judge's ruling that showing the lead from a story is not fair use.

While the ruling specifically couches it in the frame that Meltwater is not like a search engine, driving traffic to other sites, I expect the EFF and other groups to really pile on in the appeals court to gut the AP's "heart of the work" argument. I just have this sneaking suspicion the appeals court will agree.

What's clear, however, is that the next few years will see sustained battle in the courts - both legal and of public opinion - over the new equilibrium to be established in the digital age. Google already is battling on the European front.

The courts are usually about a decade behind technology in having the law catch up. We're about due.

(For some detailed commentary on all this, see Mike Masnick on TechDirt, who finds numerous flaws in the ruling, and Jeff John Roberts on Paid Content.)

Google's power searching course

I am having fun this week taking Google's online power searching course.

Some of it is very familiar for anyone who does a lot of searching, but there is some mind-widening stuff in here, too. For instance, I hadn't thought about using the color filter on images to narrow down specific types of things (the black and white option, for instance, will pull out charts and graphs that tend to be in B&W).

There's also the ability to find KML data using the "filetype" filter and feed that link back into Google maps to quickly create a useful map.

I'd recommend that most journalists and journalism students take this. It's six classes with five or so lessons per class. All are on video, but you can download the documentation from Google Docs to have a hard copy.

Editing lesson - why Goog is not enough

Here's a lesson for your editing classes - or for any journalist, for that matter - as to why relying on Google (or any other single source, especially in these days of ubiquitous data that also contains easily propagated errors).

I gave my editing class a simple car-train accident story the other day. It happened on Bonhomme Richard Drive in Lexington County. But many students went into a tizzy because it was listed differently on Google:

Now, those of us who've been around awhile probably have some sense that Goog was in error. If you have a sense of French, you know it's Bonhomme Richard, or maybe stuck back in the corners of the brain is the factoid that several U.S. warships have had that name.

But these days it is easy - and imperative - to check multiple sources. In the editing room, we have paper maps on the wall (I know, how quaint, except Google does not list county boundaries or subdivisions, both important for a local journalist.)

A quick run over to Mapquest shows this (which is also on that paper map):

And just running "Bonhomme Richard Lexington" through a Google and Bing search pulls up numerous real estate listings with the correct name.

Of course, preponderance of the evidence is not good enough in journalism, so my students should have checked with us, which some did. But the disappointment was that they were relying only on Google. What if I had put "Richard Bonhomme" in the copy? They most likely never would have asked.

Anyhow, victory is ours! OK, too much caffeine there so early in the morning. But Goog did confirm the error once I pointed it out.

I hope you'll find this useful as an example you can use in class and elsewhere.

Google removes plus operator from searches

And the hoi polloi ain't happy about it.

Apparently Google+ has the great masses of unwashed misusing the "+" operator that allowed you to specify that a term had to appear in the search.

Now, Google says it has expanded the functionality of the quote marks so that if you put a single word in quotes, it will also require that word to appear in the search.

Wonder how the research librarian community feels about this.

Some new insight into where traffic is coming from

Outbrain, an outfit that provides content recommendation and sponsored links on sites that install their widget, says it's gone through 100 million sessions representing 100 publishers that use its service to see how people are finding content.

Keeping in mind all studies like this are limited and not generalizable, it still offers some insight that can be added to the mix with others:

• Interestingly, across the Outbrain sites, most traffic (67%) is coming from people typing the URL into their browser bar, bookmarks or the publisher's home page.
• Of the remaining third:
• 41% came from search
• 31% came from another content site (linking, Outbrain referrals, etc.)
• A portal (17%)
• Social media (11%), though Outbrain says social media appears to be "gaining" share (not sure how the company concludes that, since this is supposed to be an inaugural study). Social media is not just Twitter and Facebook, but things like Digg, Fark and StumbleUpon.
• Traffic from social media sites has the biggest bounce rate - in other words, they aren't sticking around to see what else you've got to offer.

Some other interesting observations:

• The traffic from social sources is mostly to news, entertainment and lifestyle material, with news getting 42% of the referrals (makes eminent sense to me when you include Digg, etc., which tend to feature lots of news stories)
• Readers going from one content site to another are more engaged (which makes sense, as the report observes, "presumably because they already are in content consumption mode")
• Social media falls way below search and traffic from other content sites when looking at "hyper-engaged" users - those viewing five or more pages per session. Makes sense to me, especially the traffic from other content sites. In other words, if you aren't linking to other sites and getting them to link to you, you still don't really get it.

The top five sources of traffic: By far Google, followed by AOL, Yahoo, Facebook and Drudge.

There's a PDF of the report available too.

That journalism degree can be pretty handy

Support from Media Post columnist Derek Gordon that in the world of optimizing things for search, journalists have the natural advantage.

Google explains news SEO

Want to know how Google performs search engine optimization on news sources. It's explained (sort of - not all the secret sauce is revealed) in this video:




Google's "webmaster channel" has other good things, too.

And specifically for publishers http://www.google.com/support/news_pub/

Google Wave and Bing

If you've kept up with any of the tech buzz in the past week, you might have heard about Google Wave. It's Goog's soon-to-come (as in sometime this year) product that combines features of e-mail, messaging, collaboration and content management into one kind of super platform.

It's being built with a whole set of APIs designed not only to make it easy to create features within it but also to integrate it into other content, such as blogs and Web sites.

If you are in journalism, you need to pay attention to this -- heck, if for nothing more than the contextually smart spell check (about 45 minutes into the video). But consider a breaking news situation, say the plane that was forced to land in the Hudson River. The tech and journalism world was all a-twitter about Twitter and how it allowed eyewitnesses to post photos well ahead of mainstream media.

Imagine what happens if a collaborative group witnessing such an event has a tool as powerful as Wave. (One thing unclear to me is how a "wave" might be made generally public if you were not publishing to another site, but I suspect I just overlooked that.)

Here's Jeff Jarvis' take on it. Meanwhile, make time to look at the hour and 20-minute video.

Microsoft, meanwhile, has come out with its latest search-engine foray, Bing. I tried it for a bit tonight, and it is a worthy tool to add to the kit. I still think Google gives me more relevant results faster, but Bing gave me some results on my name, for instance, that I'd forgotten about and almost never see that high up (first five pages) in Google. The contextual box that comes up when you mouse over a link also is nice.

I have to give it more of a try. But my suggestion is to start playing with it.

How online changes the AP style game

Just something to consider from Robert Niles discussing how to search engine optimize your site:

Keyword repetition and density on the page still play a role in where you end up in the SERPs (though not nearly as much as in the pre-Google era.) You can help yourself, therefore, by moving away from rigid AP style rules on second references and place names to more SEO-friendly use of full names on some (but not all) subsequent references within a story.


So I find myself coming back to the average college student I teach who has been brought up in an elementary and high school system that, more than likely, encourages rules, standardized testing and the like. Those students struggle enough trying to navigate the "often you do, but sometimes you don't" vagaries of current news styles.

Niles is correct in his suggestion, but I can hardly wait for the fun.

25 advanced search engines

OK, if you have the better part of a day to blow (cumulatively, of course), check out this post from the Online Education Database -- 25 search engines trying to harness advanced technology to dig deeper into searches or make them easier to understand. It's an oldie (February 2007) but a goodie that resurfaced on my radar thanks to Dave Dillard and his Net Gold site on Yahoo.

Bottom line: If you still are just using Google, you are soooo Web 1.0.

About those search terms lists

The Wall Street Journal's "Numbers Guy," Carl Bialik, lends some leavening to the over-hyped most popular search terms lists that come out from the various search engines at this time of year.

A must read for copy editors handling such stories.

Quick links

Some quick links for a Monday morning:

A NEW STUDY COMMISSIONED BY the Newspaper National Network and performed by Scarborough Research has found a high degree of overlap in the use of online and print newspaper products, with 81% of respondents saying they regularly consumed both. It also contained some encouraging findings about readers' Web behaviors, which suggest that newspapers are well-positioned to expand their online footprint. MediaPost story.

Google, the world leader in Web search services, is the focus of mounting paranoia over the scope of its powers as it expands into new advertising formats from online video to radio and TV, while creating dozens of new Internet services. Reuters.

Think local TV is headed for the dumps, as some commentators have suggested? Apparently the private equity firms don't think so; they're snatchin up the properties even as the ad market appears soft. As MediaPost's David Goetzl writes, "This may be one reason that private-equity firms are so interested. Part of their modus operandi is to buy companies that are potentially overlooked and undervalued--then use shrewd management to build their worth, without the pressures from Wall Street to show impressive results every three months." And we all know what "shrewd management" is code for, don't we?

Social networks pose threat to newspapers. OK, nothing really particularly new here, just echoing what I and others have said for some time. The story is about a World Association of Newspapers study. More novel, however, was the finding that "the importance of the social network as a disseminator of news and information is on the rise." The survey elaborated: "Many participants in this phase listed 'discussion with friends' as a top source for news and information, sometimes ranking higher than TV or newspapers."
Well duh. Just ask around any college classroom where students get their news -- Yahoo News is up there, but so are referrals from friends, MySpace and Facebook buddies, and just plain old mom. Some of these things are so viral, I can't figure out if there even is a "patient zero."

Mark Bowden, author of "Black Hawk Down," which was backed by a pioneering multimedia Web site, says things still haven't changed much online, but the pressure is mounting for a breakout that will change the way online news is presented. (One of his descriptions sounds a lot like what we do at Newsplex: The old idea of reporters covering a beat might well be replaced by an online reporter/editor who oversees a subject area driven by the entire community - a constantly updating police blotter or transit map, for instance.) And his advice to journalism students -- learn video, learn audio, learn digital.

Tuesday Quick Hits

• Over at the Community Journalism Interest Group blog, Bill Reader has an interesting rundown of what's happening with some of the latest J-Lab funded New Voices citizen journalism projects.
• Looking for the right archive picture? UPI has put 300,000 pictures online for licensing. It includes up-to-date stuff, too, and has both royalty free and rights-managed images. You need to create a membership to find out the prices. (From Research Buzz)
• If you aren't paying attention to search engine developments as a journalist, then you're likely to get blindsided sometime in the future (sorry, but as things move online, search engines become a huge gorilla you have to learn to deal with). Pandia, the metasearch engine, now has a search site dedicated to news about search engines. (Thanks again, Research Buzz). Pandia also has a very good tutorial on how to conduct searches and another on search engine optimization (hate to say it, but that's something probably every budding copy editor should be familiar with).
• And a third from Research Buzz. If you're into movies, Flixfind is compiling a list of all the movie-related sites on the Internet.
  

'Computational Journalism'

Now this sounds like a cool course at Georgia Tech, Computational Journalism:

In this class we will explore themes such as (a) storytelling in the context of news, (b) sense-making from diverse news information sources, (c) the impact of more and cheaper networked sensors (d) collaborative human models for information aggregation and sense-making, (e) mashups and the use of programming in journalism, (f) the impact of mobile computing and data gathering, (g) computational approaches to information quality, (h) data mining for personalization and aggregation, (i) authoring and broadcasting, and (j) citizen journalism.

And here's the course blog.

Google loses again in Belgium

The AP reports that Google has lost again at a Belgian court in a suit by 18 mostly French-language papers that Google News violates their copyright rights.

The AP says the newspapers, part of Copipresse, argued that Google's cache allows access to older stories that the papers normally sell out of their archives. But Google News does not show a cache link, unlike its main search, so I wonder if the wire service is interpreting that correctly. (For a good, earlier article on the ins and outs, see Danny Sullivan's on the Search Engine Watch blog. Sullivan pretty much concludes this isn't about caching but about forcing Google to pay for any kind of link to a publisher's material.)

The same court had ruled against Google last summer, but the search-engine giant did not appear at that hearing and asked the court to reconsider so that it could present its case. Google says its news service is "entirely legal" and that it will appeal.

The Belgian court ruled that Google's technology violates Belgium's data storage laws. Unclear is whether anything similar exists in U.S. code (I'm not an expert; feel free to chime in on this). It did cut the potential fines to about $33,000 a day from a possible more than $1 million.

As search engine expert John Battelle told E-Commerce Times after the latest ruling: "The honeymoon period is over for Google when it comes to content owners."

Of course the other side of the argument has been that Google drives users to Web sites, providing untold revenue opportunities