Toluu

Google, the World, and the World Wide Web, Weblogged

Google, the World, and the World Wide Web, Weblogged

On Google Disallowing Crawling of Their LIFE HostingToday

Isn’t it great how Google makes millions of photos available to the world in their LIFE photo archive? Well – with one exception: they disallow other search engines to access these photos. The same access rights that make Google Image search crawl other photo collections are not given by Google to competing image search engines for their photo hosting. Here’s the respective part in Google’s robots.txt that says so:

<<Disallow: /hosted/images/
Disallow: /hosted/life/>>

In crawler speak, this means: stay out of our images directory! Indeed if there’s any site in the world where the owners don’t need to think about SEO in traditional terms, it’s Google.

LIFE images still do appear in Google Image Search, ever since the service was released in November last year. Google’s internal programs can internally access whatever Google stores without looking at any robots.txt.

Asking Google for their reasoning behind barring external image search engines from accessing their site, I received a reply, but it didn’t actually provide a reason:

<<While Google allows crawling of many of its own properties from Blogger to Knol, the LIFE photo archive is not available for crawling at this time. To learn more about the licensing or m

Gapminder World Chart Maps Health, IncomeToday

Google-acquired Gapminder published a different kind of world map in November last year. While there’s a compass and a sea monster on the map, the directions are not North and South, but Healthy and Sick, and not West and East, but Rich and Poor. The data the visualization is based on is from 2006. You can access the PDF of the full image from Gapminder’s download page, and I’ve copied it here as PNG image.

[Image Creative Commons licensed by Gapminder.]

[By Philipp Lenssen | Origin: Gapminder World Chart Maps Health, Income | Comments]

[Advertisement] Google Apps Hacks (book)

Checkout Icon Makes Ads 10% More Clickable, Google SaysToday

An ad with a Checkout badge to the right, with non-Checkout ads below it.

Cross-integrating their products can help Google bring customers from their more successful products to their less successful ones. Google Checkout is a payment service, and when you use it and buy an AdWords ad, you’ll not only get the Checkout sale processed for free – you’ll also get extra visibility in Google search results due to a Checkout shopping cart icon. But how much does this add in clicks, anyway? According to Google’s Checkout page: “users click on ads 10% more when the ad displays the Checkout badge”.*

Furthermore, Google quotes one advertiser, Fred Lerner of Ritz Interactive, saying the Checkout icon has given them “a 23 percent lift in clickthrough rate.”

Good for Google, perhaps less good for competition like EBay-owned PayPal.

Different Google Checkout icons used over time (is the middle one still currently used? Above image is older, and I’ve not spotted one for a while)

*I’m not sure how Google came up with these stats, e.g. whether they directly compared the same ad with and without an icon,

Google Code Hosting Abused for SpamToday

Spammers have found a way to try abuse the Google Code Hosting project. Originally, Code Hosting was and is meant as a platform to store and manage your open source projects. Including an issues tracker, a download section, and source access. Plus a descriptive frontpage which allows things like text, links, and pictures. Now look at what one user of the service made with that front page:

A screenshot from code.google.com/p/freegaysexpicturemoviex/ (if you want to visit that page, best to not click the links on it) [Update: that page has now been removed by Google.]

Below the headline “Free Gay [and so on] Movie”, an abundance of 509 searcher-luring adult keywords are pasted, with several phrases throughout printed in bold. The “embedded movie” with an animated loading symbol isn’t one at all: it’s a GIF image merely looking like a video. Click it, and you’ll be taken to another site (imp-porntube.net), which my browser warns me from using due to it being an “attack site” which may “steal private information, use your computer to attack others, or damage your system.”

Whether this Code Hosting approach works for the spammer is hard to tell. The links from the spammy page in question are nofollowed (meaning they don’t pas

Blogora, Blog + Forum Platform, Released as AlphaYesterday

David Mulder just Alpha-released Blogora, “a project to create a blogging platform tightly integrated with a forum”... and a project spun off by looking at how the Blogoscoped blog and forum functions.

A bit of a back story: You might have noticed that Blogoscoped is both a blog and a forum, and that there’s some integration between the two parts. For instance, sometimes a forum thread will later on also be blogged, in the case of which you’ll see something like “12 + 8” comments (meaning that 12 comments were made before, and 8 after this was blogged). Sometimes there were questions on whether the whole Blogoscoped code could be released as open source, and while I have many open source projects around there was one obstacle with this – I would basically have to rewrite large parts to turn it into a neat module manageable for anyone (managing not by jumping into a specific SQL table or code line like I do when I want to tweak X or Y to get a certain result, knowing where to look), including the task of locating and removing those parts which are too specific to this site to be of general use in other contexts.*

Now David took on him the job to start craft the core platform ideas of Blogoscoped, and a bunch of features not available here, into a completely new, open source framework called Blogora. Using CodeIgniter as web app development framework, David reached fast early results. He is looking f