Customize YouTube's Player

YouTube has recently updated the video page's design. Besides some obvious changes, like a slightly bigger player that lets you jump anywhere in the video, YouTube now lets you customize the player when you copy the embedding code. The custom players were already available for playlists, but now you can use them for individual videos as well. Just click on "customize embed" next to the generated HTML code and choose a theme. You could also add a border around the video and choose not to show related videos.


The code will probably include an URL that looks like this:
http://www.youtube.com/v/FI76sKLMkMU&color1=0x006699&color2=0x54abd6

You can change the two colors from the URL (color1 and color2) to any value you want: the 216 web safe colors are a good start, but don't forget to replace "#" with "0x".

Here's a list of other YouTube parameters that could be useful if you want to manually edit the code (the default values are in bold):

* border (values: 0, 1) - if the value is 1, shows a border around the video
* rel (values: 0, 1) - if the value is 0, the player doesn't show related videos
* autoplay (values: 0, 1) - if the value is 1, the video starts to play automatically

Google's Marketing Dashboard

MediaPost reports that Google wants to integrate the reporting features from all of its ad products to provide a "fully functional marketing dashboard". Google intends to connect the three pieces: search, display and offline advertising. "The more measurement you can put on this type of functionality the better, said [Google's Tim] Armstrong - noting that once the DoubleClick acquisition closed, its display metrics would add yet another layer of functionality."

This week, Google partnered with Nielsen "to bring demographic data to the Google TV Ads advertising platform. (...) A key benefit of Google TV Ads is the ability to report second-by-second set-top box data so advertisers can evaluate the reach of an ad and only pay for actual set-top box impressions. Advertisers can better understand exactly how their ad is performing and make near real time changes to their TV advertising campaigns to deliver better ads to viewers. Data derived from Nielsen's representative television ratings panels will provide Google TV Ads advertisers with the demographic composition of the audience."

Cross-correlating data across different mediums could make Google AdWords even more powerful. Google knows the users' intentions (from search queries), what web pages they visit (from Web History or AdSense/Doubleclick's cookies) and now has information about TV preferences. While Google can't connect all this data to a Google Account, it could still use it to create profiles and cluster them, irrespective of the medium. Your actions could infer things about yourself, if you connect them to aggregated demographic data. Google also collects personal data through sites like YouTube, orkut, Blogger.


At the moment, Google uses demographic targeting only for offline ads (print, radio, TV) and for AdSense site targeting, but all the data comes from external sources.

Related:
Google Analyst Day 2007 - Ads and Enterprise Overview

The AdSense Loop

I didn't want to post about this, but Google disappointed me enough to do it. Last week, I received a message from AdSense that informed me I was "displaying Google ads in a manner that is not compliant with [their] policies". More specifically, the problem was the usage of "Google Brand Features", such as the "Google" from this blog's URL.

This is the second time I receive this message from Google. Last year, a post from Philipp Lenssen's Google Blogoscoped made Google change its mind. Thanks to Matt Cutts and other Googlers, I got this reply:

"While we do require that publishers obtain permission before running AdSense ads on Blogger sites that contain Google Trademarks in the URL, we've re-reviewed your site and are happy to grant you permission to continue running ads on googlesystem.blogspot.com."

This week, Google disabled the ads again, for the same reasons as before, but this time I didn't receive any reply from Google. It's strange to see this if you think that Google Operating System has never mislead people into thinking it's an official Google blog and the usage of "Google" in a subdomain should be allowed: "Don't register Google trademarks as second-level domain names," says one of the many Google guidelines. There are many important sites that use "Google" as part of the subdomain or domain name and Google still serves ads for these sites. What's more, I received permission to run ads on this blog.

unhappyGoogleTMUsers++;

Update (13 hours later).

unhappyGoogleTMUsers--;

Interesting response from Google: "As you know, we value automation, and sometimes, that automation is unable to parse nuances that separate illegal use from legal use." Hopefully, Google will do a better job at detecting the AdSense policy violations in the future and will adopt a clearer set of rules.

Thank you, Matt Cutts and Vic Gundotra.

SearchMash, Now in Flash

SearchMash, Google's experimental site launched last year to test new user interfaces for search results, has a Flash version (requires IE/Firefox and Flash 9). The site uses tabs to let you seamlessly switch between different search engines: web search, image search, Google Maps, Wikipedia search and more. When you click on a search result, you won't open the page as you might expect, instead you'll see a sidebar that shows a snapshot of the page, the snippet and three interesting options: see more results from the site, go to the homepage and read the page from Google Cache.

Google uses snapshots provided by Snap, who also uses them to power its search engine. You may remember Snap from the annoying little bubbles that pop up every time you hover over a link in many blogs, hoping to "enhance links with visual previews of the destination site".

It's easy to move between search results by using keyboard shortcuts (up and down arrows) or the mouse wheel. To go next set of results, you can press Page Down. Unfortunately, it's not that easy to actually visit a search result, which is the main purpose of a search activity. You'll need to either double-click on a result or click on the snapshot from the sidebar. Google will open the result in a frame, so you can't see the URL in your browser's address bar or link to the page unless you remove the frame. Google had to use frames to save the context of your search, but the result is very bad.


For images, Google shows in the sidebar a snapshot of the page that embeds the image, a link to the site and information about the image size. Now it's much more difficult to find the image and you need more clicks to find your way. The video section lets you play videos in the sidebar, but fails to properly indicate the source of external videos and Google continues to host third-party content. The frames from SearchMash team up with Google Video's frames to make your life miserable.


The flashy SearchMash has a list of recent searches, but it's not that easy to find it.

All in all, the Flash interface doesn't bring enough value to compensate for the many usability problems. While it's easy to switch between results using keyboard shortcuts or the mouse wheel, that doesn't justify using Flash, especially if you realize that most people won't know about that. It's also very difficult to link to a search result, the thumbnails are too small to be useful and only slow down the page.

Some things I'd like to see in Google's search interface from this experiment: link to the homepage of a site, search inside a site, keyboard shortcuts for switching between results, a list of recent searches.

{ via TechCrunch }

Nested Folders in Gmail

If you like Google's hack to bring folder support in Gmail's IMAP implementation, there's a way to better visualize the nested folders. Basically Gmail converts any folder into a label: if you create the folder Comments as a subfolder of Blog in a mail client, Gmail will convert it to the Blog/Comments label.

To see the folder structure, you can use the Folders4Gmail Greasemonkey script (requires Firefox and the Greasemonkey extension). Instead of Gmail's geeky labels, you'll see the hierarchy, but don't expect that by clicking on a folder's name Gmail will show all the messages from the subfolders. Only the folders that don't have any subfolder are actually useful.

The script can be used even if you don't like Gmail's IMAP, but in this case you'll create parent folders instead of subfolders. If you already have Comments and Tips as two labels in your Gmail account, you need to create the parent folder (the label Blog) and add Blog/ to the name of each of the two labels (Comments becomes Blog/Comments).

Many people want folders or hierarchical labels in Gmail, but this is as close as you can get. Those who really want support for folders will feel more comfortable using an email client instead of Gmail's web interface: you can easily create new folders and use drag&drop to move messages to a folder, even if the folder is actually a label in Gmail's internal representation.

Decomposing the Web and Rearranging its Fragments

Gadgets (or widgets) are usually compressed web pages that do one thing. They're powerful and interesting because you can use them to aggregate in a single page the things that are important to you. But each gadget has its own existence and it's not influenced by other gadgets, the page doesn't have the necessary cohesion and you often need to have to enter the same settings more than once.

Imagine you want to visit London and you create an iGoogle tab about London. You'll probably add a gadget that displays maps, a list of interesting places, news about London, weather. But the map gadget can display any map, so you need to enter "London" in the settings or as a query. Each gadget has its own settings and you can't click on a place and see it one the map or change the location from the map and see the weather for your new place. The gadgets don't talk to each other and don't share their settings.

The problem was solved by PubSub, a system that "allows multiple gadgets on the same page to send and receive data from each other". Of course, the gadgets need to be updated with the types of accepted messages. Probably the first interesting collection of PubSub gadgets is the Google Finance tab that basically recreates most of the Google Finance functionality from pieces. The first three gadgets show data from the homepage (your portfolio, top movers, sector summary), while the other five show information about a specific company (stock value, charts, competitors), but you only need to enter it once.


Google Finance also announced an API that allows gadget developers to integrate financial information. "The gadget API for market data provides a framework so developers can display stock market information from the American, Nasdaq and New York stock exchanges within a gadget on Google properties. Unfortunately, due to data licensing restrictions, we couldn't open this API for display on any platform," informs one of the many official Google blogs.

I think we'll see more and more full-featured collections of gadgets that deliver a lot of functionality by aggregating data from different places, while still providing a unified experience.

To see what is all about, add the Finance tab to your personalized Google homepage and try to imagine other creative uses of the synchronization system.

Email Notifications for Blogger Comments

Here's a good tip for those who leave comments on an old post from a Blogger blog: now you can get email notifications if someone posts a new comment. Just check "Email follow-up comments to my Google account address".

Blogger also provides feeds for each post's comments, but not many people use feed readers and those who use one don't want to increase the number of subscriptions with "weird" feeds that are useful for a limited time.

Gmail Supports IMAP

One of the most requested Gmail features was the addition of IMAP support. POP is nice, but IMAP is a much better option. Among the advantages, you're always connected to the server, more clients can connect to the same account, you can obtain the text from a message without the attachments and the state information is synchronized (you can add labels from the client, read or delete a message and Gmail will synchronize).

Gmail added IMAP support, but you'll have to enable it by going to Settings > Forwarding and POP/IMAP. Because the new feature is slowly rolled-out, you may not see it, but rest assured it will be available in the next days. Gmail's help provides more information about configuring your client, whether it's Microsoft Outlook, Thunderbird, Blackberry, iPhone, Windows Mobile or Symbian device. You should also make sure you use these recommended settings.

"As some of you know, IMAP is the best way to access your email from multiple devices (e.g. phone or desktop). It keeps the same information synced across all devices so that whatever you do in one place shows up everywhere else you might access your email," explains the Gmail blog. That means you can read a message by using Microsoft Outlook and it will appear as read if you go to Gmail's web interface. Gmail's labels become IMAP folders, but if you try to add subfolders in your mail client, they'll become labels that look like this: Folder/Subfolder.

IMAP means you can experience most Gmail features using your favorite mail client. I said "most" because you'll still miss Gmail's conversations, the integration with Google Docs, Google Calendar, Google Maps or the chat feature.

Gmail becomes the most important webmail service that offers free IMAP, joining AOL Mail. The new feature, which should be added to Google Apps accounts in the near future, will make Gmail a better option for businesses.



{ via Download Squad }

Google Switches to Its Own Translation System

Google switched the translation system from Systran to its own machine translation system for all the 25 language pairs available on the site. Until now, Google used its own system only for Arabic, Chinese, and Russian.

"Most state-of-the-art commercial machine translation systems in use today have been developed using a rules-based approach and require a lot of work by linguists to define vocabularies and grammars. Several research systems, including ours, take a different approach: we feed the computer with billions of words of text, both monolingual text in the target language, and aligned text consisting of examples of human translations between the languages. We then apply statistical learning techniques to build a translation model," explains Franz Och.

You can compare the new Google Translate with Babel Fish, a site that uses Systran to provide translations. The switch is a sign that Google's system has improved a lot and could soon be ready for expanding its coverage.



{ Thanks, Steve Rubel. }

Two

2004 posts, 11302 comments, 4400000 unique visitors, two years of Google Operating System. When this blog started, Google didn't have many products that get a lot of attention today. The list didn't include Google Docs, Google Calendar, Picasa Web Albums, Google Apps, YouTube. Gmail was still invite-only, iGoogle was launched in May 2005 and Google expanded in the geo-space with the acquisition of Keyhole (Google Earth) and the launch of Google Maps.

"Another day, another rocking Google service. Google Maps is now launched and is easily a better mouse trap vs. the withering no innovation competition of Mapquest and Yahoo Maps. It has a better user interface, layout, and features. They include real-time zooming, ability to click and drag maps, ability to use the keyboard to move around, driving directions, traffic conditions, and instant location of businesses in the area." (FirstAdopter.com, February 2005)

"Last Thursday, Google launched a personalized homepage. (...) For many years, Yahoo! has had the industry-leading portal (and still does, in almost everyone's opinion). (...) So why would Google launch a portal that can't compete content-wise with Yahoo? The Fool contemplates that the company is going to leverage personalization to sharpen its paid-search machine." (Rick Whittington, May 2005)

From 4,989 employees in September 2005 to 15,916 two years later, from $343.37 in October 2005 to $654.56 last week for GOOG, from acquiring Blogger to buying YouTube, what has changed except for the size? We expect more from Google and the mistakes are more visible. The secrecy is still there, even if Google made efforts to become more open, there are less new products and more new features, the products integrate with each other, the user interface is more polished and Google's steps seem less random. People find less reasons to leave the browser as Google's applications become mature, but there's still no community, no social layer that connects these apps in a consistent way. Google is now a verb and, for many, it's synonymous with search and finding things online.

"Google is not a conventional company. We do not intend to become one. (...) Sergey and I founded Google because we believed we could provide a great service to the world—instantly delivering relevant information on any topic. Serving our end users is at the heart of what we do and remains our number one priority. (...) We want Google to become an important and significant institution. That takes time, stability and independence. We bridge the media and technology industries, both of which have experienced considerable consolidation and attempted hostile takeovers." ("An owner's manual" for Google's shareholders, 2004)

One is blogging about Google and its immediacy, two is about looking back and noticing the leaps. What has changed and what's unchanged?


Google Factory Tour - May 19, 2005 (339 minutes, the slides)

More Google Sitelinks

Many people use Google to find a website they already know, so they type things like [Yahoo], [BMW] or [weather.com]. But some people search for [Yahoo], even though the real intention was to go to Yahoo Mail. To save a click and some precious time, Google introduced sitelinks: a list of links to popular places from a site. "Our systems analyze the link structure of your site to find shortcuts that will save users time and allow them to quickly find the information they're looking for," explains Google. It's a search refinement at the site level.

Now Google shows up to 8 sitelinks instead of 4 and allows you to block the ones you don't like, but only for your sites. While Google displays only a small number of links, it actually generates a site map and sorts it according to a quality factor. A patent filled in 2005 included some examples of factors: the number of page views, the amount of time spent on a page as determined from Google Toolbar's data, but it's not clear if Google actually sorts the links this way or it only uses data from Google searches. Google Analytics and Google Web History are other two possible sources of information.

Log data storage may include information indicating whether a typical client scrolled through the web pages identified in log data storage or linked out of the web pages without scrolling. In still further alternatives, or in addition to the information described above, log data storage may store information retrieval scores associated with each web page identified in log data storage, where the information retrieval score indicates how closely a particular search query matches information on the web page. In still other alternatives or in addition to the information described above, log data storage may store information identifying the likelihood that a typical client will make a purchase associated with an item displayed on a web page. The likelihood that a purchase will be made may be provided by an entity (e.g., a company) associated with the particular web page or may be provided from user logs.

It will be interesting if Google decides to expose the automatically generated site maps through Google Toolbar or other interface and standardize the way you navigate the web.

Traffic Analysis for Content Hosted by Google

Google Analytics is useful to find information about your site's traffic, but it would be even more useful to gather stats for other sites you use to post your content online. For example, you might post videos on YouTube, upload photos to Picasa Web Albums and create documents using Google Docs. It would be interesting to link these services with Google Analytics and get access to a wealth of information about the number of views for each item, the time spent on a page, the referrals and more.

Google already does this for Project Hosting: "To provide the most useful software to your users, you might want to know simply how many potential users have visited your project workspace, which countries they come from, which browsers they use, and which of your wiki pages they have viewed. Now all those questions can be answered. Project owners may simply sign up for Google Analytics and enter an analytics profile number into the project admin page. Tracking data can be viewed on the Google Analytics site about 24 hours later."

This could solve a lot of user requests and make these services more personal and more useful. Scribd, a document sharing service I mentioned earlier this year, shows a lot of traffic data publicly, but I'm not sure if that's the way to go.

Remove Spam from Google Blog Search

Even if Google Blog Search doesn't have too many interesting features, I still use it more often than Technorati because it's faster, it's not down for hours, it's much more comprehensive and it has features not available in any other important blog search engine. I still use Technorati for finding backlinks, because Google does a poor job in this area (compare Technorati with Google Blog Search). Unfortunately, Google Blog Search indexes a lot of spam posts that steal content and use it for lucrative purposes.

Google has two features that reduce the number of splogs (spam blogs) from search results. Like in web search, there's a duplicate filter that removes some of the posts that are almost identical. But it doesn't exclude all of them and it doesn't find posts that duplicate articles from news sites like Business Week.


The second feature is the option to sort results by relevancy, which is enabled by default. It may seem counterintuitive to sort blog search results by relevancy and not chronologically, but that's a great way to filter splogs or at least move them at the bottom of Google's search results. Google uses a lot of signals to rank blog posts, including PageRank, the number of feed subscriptions or the amount of duplicate content. But if you sort the results by relevancy, you'll find both recent and old posts and that's not always the optimal solution. A better way is to restrict the results to a recent period of time in the sidebar (to the last day or the last hour, depending on the volume of posts).


If you see a "References" link after the snippet, that's an indication that Google found (a significant number of) backlinks, so the result should be a little more reliable.

Many blogs use Google Alerts to pollute the web and make money, so you could also add [-"google alert"] to your query (a search for "google alert" returns more than 200,000 results). A lot spam blogs are hosted by Google's Blog*Spot, so removing the posts from blogspot.com could increase the quality of your results, but also remove non-spammy blogs like this one or Google's official blogs. I also noticed that many spam blogs use the .info TLD. A recent study showed that, when searching for commercial keywords, 75% of the results from blogspot.com and 68% of the results from .info sites are spam.

It's also a great idea to restrict the result to English (or another language) in "Advanced blog search".

So here's a summary:

1. sort the results by relevancy
2. restrict the results to a recent period (last day)
3. restrict the results to English (or another language)
4. if you really have to sort the results by date, remove the posts that follow a spammy pattern (for example, add -"google alert" -site:blogspot.com -site:.info to your query), but make sure you don't remove important results
5. check the posts that contain "References"

Google should do a better job at detecting spam in Blog Search results and identifying results from sites that happen to have feeds, but they're not blogs. It should also make it more difficult for spammers to use sites like Blogger or Google Alerts to pollute the search results.

The Supercomputer that Connects Everything and Everyone

One of the best articles ever written about Google is "Google. Who's looking at you?" from the The Sunday Times. The article tries to understand Google's latest projects by looking at what Google really wants to do in the distant future. And while reading the plan, you'll realize that Google's power from today doesn't mean anything compared to the envisioned plans. Here are some interesting quotes, but the article is too good not to be read entirely (my emphasis):

[Craig] Silverstein, now director of technology, explains that, from the earliest days, Brin and Page envisaged a super-connected computer. "The vision of search has always been broader than has been portrayed in the press," he says. (...) He recalls one example that shows that Brin and Page imagined that one day even the smallest "stuff" would be online. "When we were doing the first research, we used to eat in Whole Foods [an organic supermarket chain]. We talked about using search to find out what aisle the salt is on. Instead of having to look at the big signs at the top of each aisle, you could use a search engine to tell you where in the store everything is, and maybe graph it out for you."

Brin and Page were obsessed with recording, categorising and indexing anything and everything, and then making it available to anyone with internet access because they genuinely believed - and still do - that it is a morally good thing to do. It may sound hopelessly hippie-ish and wildly hypocritical coming from a couple of guys worth £10 billion each, but Brin and Page insist they are not, and never have been, in it for the money. They see themselves as latter-day explorers, mapping human knowledge so that others can find trade routes in the new information economy.

Google's plans have three parts: bringing everything online and to make it searchable (universal search), indexing personal information about users (personalized search) and creating new things online using Google's web-based software (cloud computing). The idea is to bring everything online, with various levels of privacy, making it accessible using search and useful from Google's web apps.

Instead of worrying that they are going too far, Google's top team talk, with poker faces, about a "300-year mission" that will eventually see almost everything - including, perhaps, one day you and me - linked to the web and searchable online. (...) Google's techno-dream comes in three bytes. The first is loosely referred to as "universal search". Scribbling frantically on a whiteboard, Mayer, Google's head of search products and user experience, says the web is currently "very limited and primitive". It consists mainly of words, images and some music, mostly created in the last few years. There is much, much more that could - and should - be online. At its simplest level, this includes every film, TV show, video or radio broadcast ever made; every book, academic paper, pamphlet, government document, map, chart and blog ever published in any language anywhere; and any piece of music ever recorded. Google is currently developing new software that will scan millions of new sources of information to give richer search results. (...) Mayer and co argue that to be true to its mission statement of "organising all the world's information and making it universally accessible and useful", Google should be about more than searching for words, images and music; it should be about finding objects and, eventually, people. Any item that can be fitted with a radio-frequency identifier - an electronic tag called an RFID - can be linked to the internet over local or national WiFi networks.

But once you have everything online, how do you link the objects from Google's index with those who need to find them? How do you understand what they want to find if you don't know anything about them? In most cases, the same question shouldn't generate the same answer for everyone.

The second part of Google's techno-dream is "personalised search". (...) Google wants us to sign up for iGoogle on our PC, and also to install it, along with Gmail, Google Maps and Google Earth software, on our mobile phone, so that it knows not just who we are but where we are in the world, 24 hours a day, thanks to the satellite-positioning chips starting to be included in mobile phones. "Our goal is that you can, if you want, search for anything, anywhere, any time," says Douglas Merrill, 37, Google's chief information officer.

To make the information useful, you may need to use it in some context or share it with other people. Google's apps should make this happen from any computer or device connected to the Internet.

The final piece of the Google future is called "cloud computing". Instead of using the internet to search for information that we then copy and use to work on documents stored on the hard drives of our computers, using the software on those computers, Google wants us to create all our documents online, to work on them online using Google's web-based software, and to store them online on Google's vast global network of servers.

By looking at these placemarks from Google's future map, it's clear that search is the most important Google service and every new addition expands the index with more information. It's also obvious that search will propel Google's apps, which become some collectors of search results and channels for communicating them.
If this will happen and Google will become the universal ministry of search and communication, the future will tell. For now, Google seems the only company that has the resources and the desire to conquer the world at the informational level, but that's no guarantee for success.