- Page II -
Friday, June 19, 2009
Google Street View Imagery Leads to Arrest
Dutch twin brothers who mugged a teenager in the northern town of Groningen were arrested after being caught on camera by a car gathering images for Google’s online photo map service, police said.
The pair stole the 14-year-old boy’s mobile phone and 165 euros ($230) in cash last September.
“The picture was taken just a moment before the crime,” a police spokesman said.
In March, the victim recognized himself and the two robbers while surfing Google Maps, which has a “Street View” feature allowing users to see images of buildings.
[Thanks Juha-Matti!]
Update: Andries in the forum writes about how apparently Google had the faces in question blurred like usual, but then handed out the unblurred raw material upon being asked by authorities:
I live in the Netherlands, and this story made it to the 20:00-journal on all channels. The story is true, but the crime itself isn’t captured by Google, only the few minutes before. But the images were important, because they didn’t know who committed the crime.
The guy who was robbed only saw the censored version of the pictures on Google Maps. He contacted the police. The Dutch police contacted Google USA, to get the original, uncensored, versions of this pictures.
One of the investigating agents recognised one of the robber’s faces, because he committed a crime before, and the other robber was a brother of the first robber.
I asked Google about a statement and will update should they reply. Also, according to some sources pointed to in the comments, the location is Merwedestraat (I don’t know if this is true and if the screen shows the actual scene).
[Thanks Luca, Andries and Scjm!]
Google’s Gay Pride Results
Searching Google.com for keywords like gay, lesbian or sf pride currently shows a rainbow-colored stripe at the top of the results. While well-intended, I wonder if editorial engagement in supposedly neutral-and-automated-as-possible result pages is the right way to go. Put differently, while I personally agree with the notion of promoting gay pride (much needed still in many to most countries of the world where there’s no full acceptance or equality), here I wonder if it’s expressed in the right place (i.e. on result pages instead of a company blog post, a homepage doodle and so on).
Also see talking gayglers with Google’s Bennet Marks.
[Thanks Caleb, D. and M.!]
Google Product Ads Beta?
Someone who’d like to remain anonymous forwarded an email from Google, confidentially referring to a product called “Google product ads beta”. From the mail:
We are pleased to invite you to participate in an exciting beta program with the Google Affiliate Network to show product ads on Google. Product ads are paid product listings that appear when users search for products on Google. Through participation in the Google Affiliate Network product ads beta program, you can promote your products to users actively searching for your products and pay only when users make a purchase on your site.
This opportunity is a very tangible benefit of the integration of our affiliate platform into Google. Google is constantly experimenting with new features and visuals to improve ad effectiveness and advance the end user experience.
Google product ads will feature product specific information directly in the ad such as price and product image. During the beta program, Google will be testing to identify the most effective ad formats. Google product ads will complement standard text ads on Google.com and will run independently during the beta. All reports on the performance of Google product ads can be seen through your Google Affiliate Network account.
The Google product ads beta program relies on your Google Base feed. Minimum pricing is your standard publisher rate, plus the network fee. To maximize your competitiveness among advertisers participating in product ads, you are encouraged to increase the CPA rate for your product ads relationship.
The FAQ goes on to say (my embolding):
Where would my product appear on Google?
Your product would appear as Sponsored Links on Google when users search for products that match the items in your feed. Beta testing is for US / English.How much does it cost?
You specify the commission rate for conversions that take place via clicks on Google product ads. Minimum pricing is your standard publisher rate, plus the network fee. To maximize your competitiveness among advertisers participating in product ads, we recommend a higher commission with the Google product ads relationship.How do I see reports during the beta testing?
Through your Google Affiliate Network account, you will see reports for Google similar to other tracking publisher reports. Please keep in mind that our initial experiments will be limited to a small percentage of traffic until our full launch.Do product feeds need to be submitted through Google Base?
Yes, these ads use information from your Google Base product feed. Google Affiliate Network is integrating with Google Base to create product ads with trackable product links.How do these ads compare with free product listings via Google Base?
Product ads appear as a Sponsored Links on the page and Google will use a variety of prominent placements and additional display options to optimize and drive targeted traffic to your product ads. You pay only for qualified conversions that take place through Google product ads. Product listings that appear in the search results will continue to be free.How do these ads different from AdWords text ads?
Product ads represent individual offers with product information directly in the ad itself. Unlike keyword targeted text ads, product ads appear when the user’s query matches information provided in your product feed. During the beta, product ads are shown independently from text ads.Can I add keywords to my ads?
No. Google categorizes and matches the items from your product feed to the user’s search query for you. Please make sure your product information is complete and accurate in your product feed to Google Base.How are these ads ranked?
Ad Rank = Commission × Quality Score. The quality score takes into account the relevance of your product to the user’s query, conversion rate of the query and the matched product ad on Google, your account history, and other relevant factors.
[Thanks Anonymous!]
Thursday, June 18, 2009
Microsoft’s Browser Comparison Chart
Microsoft has a “Get the facts” page for Internet Explorer 8. As you can see, Internet Explorer has security, privacy, and ease of use, whereas Firefox and Chrome don’t have those. Gotta love some old-fashioned propaganda. [Via Reddit.]
Google Germany to Erase Internal Unblurred Street View Data On Request
Associated Press writes:
Google Inc. is willing to concede to German demands the company erase photos for its panoramic mapping service after they have been processed, a data protection official said Wednesday.
Johannes Caspar, the head of the Hamburg regional office for data protection, said Google had agreed to erase the raw footage of faces, house numbers, license plates and individuals in Germany who have told authorities they do not want their information used in the service.
In other countries, like the US, Google does blur the faces and license plates and so on in the output seen on the website – but they’re still keeping that unblurred photo internally. The crucial difference here is that in that case, the government (if laws permit it etc.) could ask Google for certain content, and Google may be handing it out to them. (Whether that would be that bad after all – it’s usually stuff taken in perfectly public spaces anyway – is yet another discussion.)
Should this catch on then there’s a couple of downsides to this from Google’s perspective. For one thing, Google can’t fully reprocess the blurred parts, e.g. should they find a better way to blur faces or license plates. Plus, in the future they can’t fully check on aggregate data of blurred parts, say to calculate the amount of Dutch cars driving around in Hamburg. However, the original photos are only going to be erased on user request, as the AP says, so that probably won’t make a big difference overall.
Tuesday, June 16, 2009
Interestingly Shaped Communities on Google Maps
The SEO Company shows off a couple of intriguingly shaped communities as spotted on Google Maps. [Thanks Dave!]
Make a Celebrity Odd
At the new MakeaCelebrityOdd.com, you can, well, make a celebrity odd. Please give it a try and comment with your feedback, it’s just a little experiment I did.
On a side-note, this runs completely on the client-side, as JavaScript. The image search engine used in the background of this is the Bing (JSON) API, as it offers a neat way to restrict to just faces without showing shoulders, whereas Google’s REST API only lets you restrict to faces. (When not enough Bing results are found, the Google API is used, though.) For the image effect, there’s no Flash, but Canvas for some browsers, and a flip filter for Internet Explorer.
Trying Out Google Wave (Videos)
By Philipp Lenssen & Tony Ruscoe
We are trying out the developer preview of Google Wave. Please keep in mind that this is a very early version of Google Wave with known bugs... this is not a final product yet.
Each Wave document consists of several sub-parts, called "Blips". Tony and I can chat along inside a Wave, and we're seeing what the other types letter by letter. (PS: Oops, I miscorrected Tony's sentence at around 3:30!)
In Google Wave, you can invite a robot to a document (a Wave) like you would invite another person. This robot is a program that will do something to your Wave in real-time; Rosy, for instance, auto-translates things being typed into the Wave ("Rosy" as in "Rosetta Stone"). Another robot can auto-complete sentences for you. Yet another robot can convert text smileys into images.
You can use Wave as a kind of real-time teaching tool if you like. For instance, in this video Tony tells me about Sheffield, while embedding map gadgets and more.
Technically, Wave is sometimes a bit slow to load, but otherwise very impressive. On more of a non-technical side, we were wondering whether Wave trying to be everything at once is its upside or its downside.
On the upside, the idea may be that there's no need to switch to other tools, as a Wave is a wiki, chat, email, translator, gadget playground and more all in one! Less choice could theoretically mean less thinking about which app to switch to.
On the downside, an interface that's meant to handle everything at once can risk being not particularly suited for anything specifically. The app may additionally become socially ambiguous ("does the other person want to chat with me in real-time, or is it OK if I answer tomorrow?" ... "Do I 'hang around' in this document waiting for the other person?"), and not very discoverable in terms of use patterns ("what am I supposed to do here?", or as Tony and I wondered, "Are we using this thing the right way?").
But it seems too early to tell for Wave as Google says it's in its early stages, so we'll have to see when it goes live. Already, several of the things available in Wave might be inspiring for other web apps and spawn and push forward new ideas elsewhere.
[Thanks to Google for the invites, and thanks to Jens Rasmussen!]
Sunday, June 14, 2009
Google Calendar Gadgets Appearing
Andrew Pariser emails in a surprise finding in Google Calendar (if by any chance you’re able to reproduce this – I’m not – please comment):
Opened up gcal today, to find an interesting new “Gadgets” link under the google logo. Clicking it opens a right side pane that introduces a gadgets menu and a selection of gadgets to choose from:
- Add Tasks
- Add Googler Search
- Add Jump to Date
- Add Next Meetings
- Add Time Zones
- Add Where Are My Friends?
I also have a nice “INTERNAL ONLY” tag at the top of this menu, even though I’m not and have never been a google employee. (...)
Most of the gadgets (tasks, jump to date, time zone) are pretty obvious. Next meetings shows the next event scheduled in your calendar. Googler search presumably allows me to search the employee database, although I think it’s disabled for me. Where are my friends allows you to add friends and it displays their availability information as decided by their google calendar (provided they keep this information publicly viewable).
[Thanks Andrew and Tony!]
Update: As noted in the comments, you can try disable all CSS on the page (e.g. in Firefox, hit View → Page Style → No Style), then click “Gadgets” to the right, and scroll to the bottom of the page. Now you may see the text “INTERNAL ONLY”. Clicking in these parts of the page will lead to (supposedly internal but accessible) feedback forms, or (inaccessible from the outside) Google intranet pages.
The source of the Google employee search widget, “Googler Search” (facewall.xml), is also available, albeit it probably only runs from within Google. For instance, thumbnails for the search are stored at https://moma-api.corp.google.com/[prefix]/thumb (where prefix is an employees name as appearing in their email). The server search returns a JSON profile of an employee, and the codename “Woodstock” appears (perhaps denoting the Google intranet info API, I’m not sure).
[Thanks WebSonic!]
Friday, June 12, 2009
If Google Were Your Roommate... (Video)
[Thanks WebSonic, via Digital Inspiration!]
Dear Websites, Please Follow These 10 Rules.
Dear websites! Here are some general rules I’d like you to follow. Please, just to make me happy. It’s not that hard!
- Open pop-unders. Pop-ups were those additional advice windows containing helpful information or contests allowing you to grab music players for free. Since a while, evil browser vendors hide them from my view. (Firefox does not want you to have a free iPhone.) However, clever programmers out there found a technology that still gets me the information I want when I want it: pop-unders that are triggered on click. Use them!
- Come on, bundle it with a little extra! You know those boring programs you can download that don’t exceed your expectations? They just do what you believed them to do. One word: boring. When I download an app, I want it to come with a bonus that shows customer’s king and all. Programs like Java, which by default comes spiced up with the really helpful and neat Yahoo Toolbar, are reasonable distribution patterns your app should follow too.
- Long click paths help us relax. Don’t let me get to your product or offering too fast – make it a multi-part route. Have you ever heard of mandalas? That’s the ancient art of sandpainting that helps us humans get into trance. Well, if you put your download or order form right on that first page where I’m looking for it, how do you think I’m going to reach the next spiritual level? It’s like producing a mandala using a laser color printer: fast, yeah. Trance, meditation, finding the meanings of life – no!
- Use smallish, protected images when we zoom in. We all hate to click on a thumbnail of a product, or a pic on some blog, only to discover there’s a really hi-res photo of that pic opening on us. Why? Because most of the time, that squelches our imagination and disillusions us. Instead, as a webmaster, ensure you’re opening a picture that’s basically the size of the thumbnail plus around 6 - 8 pixels. Extra tip: Put a fat watermark over the zoomed pic, and when we right-click it, show the message “Don’t steal this pic, ’cause if you do it will be gone and no one else can look at it.” (And sheesh, if by any chance you have a Creative Commons license on any of your content, get rid of it... that stuff is worse than communism.)
- Make that Captcha harder. When you solve a picture to submit a form, you’re expecting a riddle worth your attention. No patronizing please, we can handle it. Rule of thumb for captchas: If you can read it, it’s probably too easy. Websites in the last years made great progress in this area – follow their lead.
- Spread the article across several pages. Have you ever come across a news article or blog post that had all of its content on a lengthy, scroll-intensive, single web page? That sucks, for obvious reasons: the mouse scroll wheel hurts our fingers; the page takes longer to load; we only get exposed to one blinking banner. So, rule number 6: please put your article on around 5 - 7 pages, depending on length. You’ll probably wanna shoot for around 2 paragraphs per page. (Note the loading time between the individual pages should give people time to reflect on what they just read... around 10 - 15 seconds should be fine.)
- Inform me via email! When you have a website, please email me about it – even if I didn’t opt in to that. After all, how should I have opted in at your site? I didn’t even know it existed! As a general guideline, people are most interested in steel abs, buying cialis, replica design watches, and dubious business offers from other countries, so if your website is about any of those, bring it on.
- Use PDF files. HTML is a lousy format to transmit information. It does not properly fix the layout on the receiving end, allowing us to mess it up in all the wrong ways. Instead of freezing the browser for a bit it loads too quickly, ruining all excitement. It contains ugly brackets, like this one > and this one <. The PDF format solves all these problems, so please, use it wisely, i.e. wherever you can.
- Be the mystery guy. This one is for all the blogs out there who have their Contact and About info just a click away from every page: please. Stop it. No one cares about you, and by revealing who you are, you’ll spoil the mystery. Have you ever read a Sherlock Holmes book that would start by showing us who’s behind the killings? Have you ever watched a Columbo episode that starts by revealing who the murderer is?
- Lie. That last one may be obvious, but I’m listing it here anyway, for completeness: lie. Don’t tell me all the nasty details about your service. Don’t let me know about the secret program running in the background of the program I installed. Don’t disclose who’s paying you. Don’t keep to your privacy policy. (Note: if you can’t lie straightaway, then at least stretch the truth a little, bend reality, be a tad dishonest, or keep quiet about the stuff I would like to know. It’s not ideal, but better than no lying at all.) Without lies, life would be boring indeed: a game of poker where all cards are on the table. Who wants to play like that?
Google’s Cloud Contest... With an Odd Twist
Google has a US contest going on in partnership with airline Virgin America that’s aimed to promote “the cloud” (that is, moving your stuff online and then working on it using web apps). Ironically, by the way the prize is set up, it almost promotes the risks of that cloud. Here’s what happens when you win: Along with a notebook and some other stuff, you get 1 Terabyte of data! Whoopee? Not quite – after 1 year, you have to give the prize back (i.e. the 1 TB offer ceases), and if you want to keep the extra storage, you’ll need to pay for it. As Drtimofey in the forum suggests... you may end up spending 6 months to upload your data, and the other 6 months downloading it just to keep it.
In the cloud, things aren’t fully yours... not even contest prizes?
[Thanks Mrrix32, Drtimofey and Jason!]
How Facebook Uses Your "Skipped" Webmail Contacts
By Tony Ruscoe
Yesterday, Facebook suggested two people to me through its “Suggestions” feature which usually includes friends of friends, co-workers and people I used to go to school with. The odd thing about these two suggestions was that although I knew both of them – I had made contact with them years ago because they are my third or fourth cousins – we had no friends in common, we had never worked at the same place, we even lived in different parts of the world. So how did Facebook know that we knew each other?
I’m sure all you Facebook users are already aware that you can enter your Gmail (or other webmail) username and password to import a list of your contacts into Facebook to see if any of them are already registered based on their email address. This is something I have never done as I don’t like to enter my Google Account password on third-party websites. Even if I had done this, I knew for a fact that I had never used my Gmail account to email these two people.
But what if Facebook had used my friends’ imported contact lists to suggest their profile to me even though they didn’t add me as a friend? I am now pretty sure that’s what happened here. Here’s how I proved it:
My friend added my email address to his Contacts in Gmail.
My friend signed in to his Facebook account and imported his Contacts from his Gmail account using the “Find People You Email” feature.
My friend chose to skip the friend suggestion it was making based on my Gmail address.
I signed in to my Facebook account and saw that my friend’s Facebook account was being suggested to me.
In summary, it seems that even if you choose to skip the contacts you have imported, Facebook will still store your relationship with those contacts. Not only will it continue to include them in your suggestions, but it will also alert them to the fact that you previously imported their email address and that you are registered on Facebook. Facebook clearly states that it will not store your password, but it doesn’t tell you that it will store all your contacts even if you chose to skip them.
Admittedly, your account will only be suggested to others if your privacy settings allow your profile to be returned in search results, so anyone could search for your profile themselves, but is it right for Facebook to suggest you to the people that you have chosen to skip? Also, does this mean it’s possible to force yourself into someone else’s suggestions list by simply adding their email address to your contacts?
Update: Canna points out in the forum that you can now remove this information from Facebook using the Remove Contacts Imported using the Friend Finder page, usually accessible via: Friends > Find Friends > Learn More. (Perhaps this is a new page as I don’t remember seeing that link before...) [Thanks Canna!]
XKCD on Google Latitude
[Cartoon by Randall Munroe, CC-licensed. Also see Google Latitude. Thanks TomHTML!]
YouTube Increased Privacy of White House Videos
The Electronic Frontier Foundation yesterday writes:
When the Whitehouse.gov website launched in January, including embedded videos from YouTube, privacy advocates raised concerns that without extra privacy measures, YouTube would be improperly tracking visitors to the government website, including recording which videos were watched and combining that information with the ever-growing amount of information that Google and YouTube have about internet users, through YouTube’s use of cookies.
In response, The White House first made sure that YouTube’s cookies were not served merely upon visiting the website (...)
Now YouTube says that they have taken a second step urged by EFF: essentially ignoring their account cookies for videos viewed on Whitehouse.gov. Ordinarily, YouTube maintains a record of every YouTube video you’ve ever viewed, associated with your YouTube account, through use of the YouTube cookie. Now, they’ve agreed to exempt videos embedded on Whitehouse.gov from this logging.
The EFF says this is a good step. Now, they say, they’d like to see similar approaches being taken beyond just the White House website. “Human rights videos, politically sensitive videos, or even ordinary videos where viewers may want privacy should all be available without tracking.”
If you like what the EFF is doing, you can donate to them.
[Thanks Juha-Matti!]
Thursday, June 11, 2009
Picasa With Creative Commons Search
Clicking “Show options” in the upper left expands settings like sizes, aspect ratios, or licensing.
Google’s photo album service, Picasa Web Albums, now allows you to show options during your search. As Ionut noticed, as part of these options you can tick the “Creative Commons” link, which will only return shareable pics.
The amount of images is not all too bad either, at least for some queries: a CC-only search for the keyword google shows 276,529 pics, according to Picasa. A search for obama returns 43,510 pics right now. For comparison, the same CC-only obama search yields 127,858 results on Flickr. (Naturally, someone claiming that a pic is CC-licensed does not necessarily make it so, as the original person may not have the copyright to begin with.)
[Via Ionut!]
Tell Google When You Moved a Site
If you’re changing from one domain to another for your website, it’s good practice to use a permanent redirect header for all the old pages. Added to that, you can now let Google know of the move in a new Webmaster Tools section called “Change of address”. “This will help us update our index faster and smooth the transition for your users,” Google says.
A theoretical problem with configuring settings like these in the Google Webmaster Tools is that other search engines are locked out of your preferences, unless you’d mundanely repeat them for all search engines which would want to support them. Formats like robots.txt or Sitemaps.xml are more open in these regards – though probably also tougher to agree on, if you’re looking for standards to apply across all major search engines. (For this particular setting, permanent redirect headers may be the standard here already; however, even unauthorized sites could point a redirect to your site.)
Back in 2007, I moved this domain from blog.outer-court.com to blogoscoped.com... after a couple of months, I were able to check in on some of the results.
[Thanks WebSonic!]
Wednesday, June 10, 2009
New From Google: Fusion Tables
The tables overview...
A single table is displayed, 100 rows at a time.
Map visualization of data. (Google Spreadsheets has a related feature via Insert → Gadget → Maps.)
The Google Labs have released a new tool called Fusion Tables. It lets you view, visualize, merge & discuss large tables of data. In the tables gallery right now, there’s data like baby names, or Academy Awards by country.
Somewhat confusingly, this product seems like a set of features that already is, or seemingly could become, integrated in the existing Google Spreadsheets. Creating a new product just to manage more data seems like the opposite of consolidation efforts – merging products into a larger one instead of creating multiple related ones at separate locations. Google’s Marissa Mayer said in 2006 “It is hard for people to remember more than 5 or 10 products from a particular company. If we can take each of the products we have and make them even larger and more meaningful to people, I think there’s a lot of benefit that could be had by both the users, because they don’t have to remember quite as much”.
In the Fusion Tables FAQ, Google addresses the issue of how this product is supposed to be different from Google Spreadsheets:
The goal of Fusion Tables, as with other database systems, is to manage larger amounts of data than spreadsheets typically do. This size difference leads to a focus on a different set of functionalities. For example, Fusion Tables focuses more on bulk operations on the data (filtering, aggregation, merges). These operations are typically not necessary for smaller collections of data stored in spreadsheets. In contrast, spreadsheets preserve complete flexibility in managing data – you can put any value you want in any cell and work carefully to format your spreadsheet to make it look nice. Fusion Tables requires that the data be tabular with column headers and types for each column.
As far as size limits are concerned, Google in their help elsewhere says that each Google Spreadsheet “can be up to 256 columns, 200,000 cells, or 100 sheets – whichever is reached first. There’s no limit on rows.”
[Hat tip to Hebbet and Ionut!]
Tuesday, June 9, 2009
Google Translator Toolkit
By Tony Ruscoe
Google Translator Toolkit is a new tool being launched today to help translators organize their work and benefit from shared translations, glossaries and translation memories, the Google China Blog reports (English translation by Google).
Evidence that Google was working on a service like this originally surfaced in August 2008 when references to Google Translation Center appeared in Google’s robots.txt file. At the time, the service was only available to Trusted Testers and most of the pages and screenshots were quickly taken offline. Since those screenshots were produced, it's clear that a lot of changes have been made to the tool.
The Translation Process
The Google Translator Toolkit Workbench, showing side-by-side editing of Wikipedia's Google article.
For those not familiar with standard translation processes, a professional translator is likely to use a Computer-aided translation (CAT) tool to help identify and extract snippets of text for translation from various file types.
Google Translator Toolkit currently only allows users to upload HTML, Microsoft Word, OpenDocument Text, Rich Text and Plain Text documents up to 1MB for translation. Alternatively, it's possible to enter the URL of a file on the web, select a Wikipedia article or a Knol for translation.
Once uploaded or selected, files can be translated using the Workbench interface which shows the source text and the target language translations either side-by-side or above and below each other.
Previously translated segments from the translation memory are suggested and can be rated by yourself and others.
One good reason to share translations with others is so that they can be reviewed for consistency and style. Google allows users to rate translated segments, presumably for style and accuracy. Comments can also be added to the target document, which is especially useful when collaborating with other users.
Translation Memories
In addition to the global translation memory, users can also create and share their own TMs.
Many CAT tools allow the translator to store their human translations in a database called a translation memory. The memory can then be used to help with future translation projects by checking to see whether a certain word, phrase, sentence or segment has been translated before. Even if it's not exactly the same phrase, the translation memory can be used to suggest what's called a fuzzy match, often indicated by a percentage to reflect how similar the text is.
When translating Wikipedia articles and Knols, the translations are stored in a global, shared translation memory that's available to everyone by default. That means previously translated phrases from these articles are stored and available for use by other translators using the service, so if they ever find themselves translating the same piece of text, Google will automatically populate the interface with the previous translations to help save time.
Google's support article explains the process:
Pretranslating your documents
When you upload a document into Google Translator Toolkit, we automatically 'pretranslate' your document as follows:
- We divide your document into segments, usually sentences, headers, or bullets.
- We search all available translation databases for previous human translations of each segment.
- If any previous human translations of the segment exist, we pick the highest-ranked search result and 'pretranslate' the segment with that translation.
- If no previous human translation of the segment exists, we use machine translation to produce an 'automatic translation' for the segment, without intervention from human translators.
We realize for some translators, pre-filling with machine translation may actually slow, not speed up, the translation process. In such cases, you can change your settings to pre-fill the segment with the source text, so you can type over the source text instead of making corrections to automatic translation.
Uploaded documents can benefit from using this global TM too, but if users don't want to share their translations with everyone, they can create their own translation memories and control exactly which users can make additions and rate translations.
Translators already using CAT tools may have translation memories stored in the Translation Memory eXchange (.tmx) open standard XML format. Google allows translations contained in those TMs to be uploaded and added to existing Google Translator Toolkit TMs, providing they're no larger than 50MB and confirm to TMX 1.0 or higher.
TMs other than the global TM can also be searched for previously translated segments which can then be rated without opening a translation document.
Glossaries
Glossaries are collections of words and phrases with definitions and notes associate with them. They are often used in the translation process to help choose which phrase is most appropriate and to maintain consistency between translations of technical or specialty subjects. Google Translator Toolkit requires CSV format glossaries to be uploaded (it's not possible to create one from scratch) which will then be automatically searched for terminology in the segments that are currently being translated.
Learn More
For a really quick overview of some of these features in action, you can watch this YouTube video:
How could this be useful to Google?
A machine translation of the Google China Blog explains, "Google's mission is to organize the world's information and make it universally accessible and useful. Translation of information, in our view is the key to access to information."
Google has been working on a statistical machine translation system for a few years now, which it started to use for Google Translate instead of Systran in October 2007. Since then it's been slowly integrating translation into many of its services, including Google Toolbar, Google Talk, Google Reader, Gmail, and YouTube. There's even an AJAX Language API which anyone can use to build upon.
In my opinion, this latest tool has clearly been designed to help improve Google's translation offerings. One thing on which statistical machine translation relies is aligned translations. In very simple terms, to help train a statistical machine translation system, text in one language is fed into the system alongside the same text in another language. Will enough text, the system can start to learn how certain phrases should be translated. Without aligned translations, there's no easy way to know exactly which sentence in the source document relates to the translated version. That's where translation memories are very useful; they contain aligned translations.
There are literally thousands of Wikipedia articles being translated all the time, but the translations aren't usually maintained in a translation memory. Through using Google Translator Toolkit, translators could benefit from seeing previously translated text from the global translation memory and, in return, Google could clearly benefit from translators using its interface to translate any content that's then stored as aligned translations in their global TM, which it can ultimately use to enhance its statistical machine translation system and improve the translations that are provided to end-users of any service using Google Translate.
And as the global TM grows, it might even be possible for end-users to get near-to-human-quality for translations of their documents, websites, blog posts, emails and tweets instantly.
[Thanks TOMHTML!]
Disclaimer: I am an employee of SDL, a translation company that provides translation services and software.
Android Scripting Environment
Google announced the Android Scripting Environment. You can use Python and other languages on phones running the Android operating system to “Make phone calls ... Send text messages ... Scan bar codes ... Poll location and sensor data ... Use text-to-speech ... And more.” Personally I wasn’t interested in using the G1 phone partly due to its hardware keyboard, but things are starting to get very interesting here.
[Image by Google, via their post. Thanks Tony!]
Monday, June 8, 2009
Bing API
Microsoft’s Bing search engine has a neat REST API. You need to register an AppId, and then you can grab the XML, JSON and more like this:
http://api.search.live.net/json.aspx?Appid=YOUR-APP-ID& query=hello&sources=web&market=en-us
Here’s a couple of things you can do: get web results; get video results; get image results; integrate ads; use a spellchecker; grab news results, and more.
[Via DeWitt.]
Apply to Try Google Apps Script
If you want to test Google Apps Script but don’t have a Google Apps account, you can now apply for a test login. In the form, enter your name, your email address, and the domain tryscript.com. The Google Docs product manager Jonathan Rochelle says, “I will try to respond to each valid respondent on Monday - 8 June - for as many people as I can allow on that domain...”
Once you do have access you’ll get an email. For your first script, open a Google Docs spreadsheet and click Tools -> Scripts -> Edit Scripts, and e.g. follow the further guidelines from the introductory tutorial.
[Thanks Ahab!]
Update: The limit for how many invites will be handed out has been reached, so this approach won’t work anymore, Ahab says.
Wasted PageRank Sculpting, and JavaScript Links Needing Nofollow
Barry Schwartz over at Search Engine Roundtable a couple of days ago highlighted two interesting points from the SMX Advanced conference. Based on information Google’s Matt Cutts offered, Barry says (but please take this with a grain of salt until Google clarifies the issues officially):
- that when you have say 10 links and you nofollow 5 of them, you won’t gain anything – PageRank-wise – for the remaining 5 links (but basically, you have thrown away PageRank). Webmasters trying to “sculpt” the PageRank of their sub-pages by carefully adjusting nofollows on their stronger-PR homepages and the like may be surprised that their tactic may not be successful.
- that people are now required to use the nofollow attribute even on links which execute JavaScript to forward to the text ad target page. The reasoning behind this is that as Google gets smarter about following links, including JS-based ones that are somehow contained within onclick events, there is need to adopt nofollow even for these special cases (if they’re advertisement links, that is).
I’ll be updating this post if there will be an official word from Matt/ Google, like an adjustment to some of the details of above. Also worth noting that even if point #1 is correct as it is, Google may update their algos in the future, so it might make more sense to simply output your site with mostly the user (and the general HTML specs) in mind.
[Thanks WebSonic!]
Update: Matt Cutts has now written a post on PageRank sculpting. He confirms that yes, you can’t increase the PR of remaining links if you make say half of them nofollow. He also mentions how it might not be a good idea to nofollow most outgoing links, because just as Google would punish links to a bad neighborhood, they may reward links to good sites.
Microsoft Advertising Bing Through Google AdSense
Daniel Jarratt sent in the following:
I saw Microsoft ads online today as part of their hundred-million-dollar advertising campaign for Bing. What I couldn’t understand is why MS took out ads on Google AdSense. Sure, it’s the best way to distribute ads Internet-wide, but that’s a huge amount of money to send to one’s arch-competitor. I navigated over to your site to email you, and saw the same ad on blogoscoped.com [screenshot]. See attached images. Advertising on hulu and other popular sites makes sense, but they must think they’re going to get a huge return on investment from AdSense to use with a competitor’s product. That also says a lot about Microsoft’s respect for Google’s AdSense efficiency. (...)
P.S. My wife thinks Microsoft is going for irony
[Thanks Daniel!]
Sunday, June 7, 2009
Blind Search: Compare Search Engines and Cast Your Vote
Blind Search is the name of a site that lets you enter a search query, and then shows you three result columns where you can vote on the best one. Because you don’t see where the particular results are coming from until after you vote, you will not be directly influenced by the brand. Part of the test of Blind Search are Microsoft’s Bing, Yahoo, and Google.
I find this an interesting test. My first, completely non-representative trial was entering blogoscoped about. Of the three results, two showed the Blogoscoped homepage, but only one showed the actual, more relevant Blogoscoped About page. Guess which search engine it was?
Some caveats: Even when the logo will not be visible, it’s entirely possible that using any particular search engine for a long time will subconsciously makes you adjust how you formulate queries – namely, you might train to phrase searches in such a way that your particular (imperfect) search engine of choice returns its most relevant results. So if you’ve used Google for years now, then it’s worth noting that your queries themselves might be more “Google-friendly”, thus give Google a slight edge over competition. Another thing to keep in mind is that certain features, like automated spelling corrections for top results (as Google offers it for some queries), are not correctly pushed to the Blind Search interface; this may skew voting results.
Right now, among the 793 votes casted already, Bing is in the lead with 38%, followed by Google with 36%, then Yahoo with 25%. (There may be further caveats to note for these poll results, like which groups are starting to enter queries at the site, and also, what types of queries people enter when testing vs when in real-life search situations.)
[Thanks WebSonic!]
Update: Another potential caveat: the creator of this test, Michael Kordahi, works for Microsoft. [Hat tip to Matt Cutts.]
Update: Google was in the lead for most parts of the day as far as I watched it, but during the last few minutes Yahoo is rising suspiciously fast (“Google: 31%, Bing: 24%, Yahoo: 45% | 41,304 votes”) – perhaps the poll is being hacked, or there are some other problems skewing this.
Update: Blind Search’s creator Michael comments: “Indeed the poll has been compromised probably equally due to my lack of leet skills and a malicious person gaming it ... Will eventually get around to fixing the data.” [Thanks Michael!]
European Elections
Today are the European elections in several countries, including in Germany. One of the parties on the list here is the Pirate Party. On the date of the election, BBC in their election FAQ writes: “The date of the election will vary from 4 June to 7 June 2009 according to local custom in the 27 member states. The UK and the Netherlands are voting on 4 June, while most of the other member states, including Belgium, Denmark, Estonia, Finland and Germany, will hold the election on 7 June. First results are expected to be announced at 2000GMT ... on 7 June.”
In regards to the Pirate Party, Wikipedia writes: “The Pirate Party (Swedish: Piratpartiet) is a political party in Sweden. Its sudden popularity has given rise to parties with the same name and similar goals in Europe and worldwide. The Party strives to reform laws regarding copyright and patents. The agenda also includes support for a strengthening of the right to privacy, both on the Internet and in everyday life, and the transparency of state administration.”
Is Sun Indirectly Paying for Blog Post Reviews?
Just saw this Google ad above – is that JavaFX contest by Sun a kind of paid blog posts scheme? (And if those were to contain non-nofollowed backlinks to JavaFX.com, would that be against Google’s webmaster guidelines?)
The following bit is from Sun’s contest guidelines:
Complete the submission form and provide a URL to your blog describing your experience using JavaFX 1.2 SDK. The blog may be about your experience using JavaFX 1.2 SDK, its reliability, performance, manageability, ease of use, new features, a tutorial, tech tip, code sample, requests for enhancements or anything related to JavaFX 1.2 SDK. (...) By participating you give us the right to use the story on Sun’s web site and/or in other marketing promotions related to JavaFX. (...) Entries that are lewd, obscene, pornographic, disparaging of the Sponsor or otherwise contain objectionable material may be disqualified in the Sponsor’s sole and unfettered discretion.
Saturday, June 6, 2009
Printer Sounds Revealing the Words Printed
German Heise writes (my translation):
Combining automatic speech recognition and machine learning, Michael Backes – professor for cryptography and information security of Saarland university – and his research team were able to reconstruct nearly three quarters of texts printed with a needle printer by the printing sounds it made. To do so, the researchers originally printed a dictionary, recording the printing sounds and associating words with their characteristic sound in a database. The resulting program then aimed to analyze further sound recordings, filtering through the noise to automatically recognize the words. 70% of the printed health records and bank account statements were successfully filtered – including PINs of bank accounts.
Heise goes on to report how the researches posed as patients, and – equipped with microphones – were able to spy & crack certain prescriptions printed in a doctor’s office. You can find out more about this project (in English) at the University’s homepage.
Just last year, Backes and his team were able to show that it’s possible to see what’s on your monitor by looking at the monitor’s reflection on things like a tea pot, coffee mug, or your glasses – from 10 meters away, using a special telescope.
[Via Rainer.]
StackOverflow’s Karma Bounty, Data Dump, and Sister Sites
Karma Bounties
Programming question & answers site StackOverflow.com has a system based around karma and badges, and karma acts as a kind of in-site currency. One nice extra feature are bounties. When you don’t get a good answer to your question after some time, you can start a karma bounty. You will assign 50 to 500 points from your own hardly earned karma, which will then be assigned to the question – up for grabs to anyone who can offer an answer that will become the top-voted answer.
Furthermore, the question will then show a special icon, and be shown on the homepage’s “Featured” tab for some time. To use the bounty system yourself, click the “start a bounty” link on certain questions of yours, or check out the “Featured” tab, pictured at the top.
Data Dump
StackOverflow is noteworthy not just for its community features on the site, but also for its openness. Recently, they announced that they will offer all the Q&A data generated on the site in the form of a Creative Commons Attribute-Share Alike license. In other words, if you do decide to republish it, you need to credit the source, but also license your own work generated from it under the CC license.
Jeff Atwood at the site’s blog writes, “The community has selflessly provided all this content in the spirit of sharing and helping each other. In that very same spirit, we are happy to return the favor by providing a database dump of public data.” Jeff adds, “Our plan is to create a new data dump every month, reflecting all data in the system up to that month.” The current dump is over 200 MB large and available as a torrent.
A sister site
Also, StackOverflow now has a sister site: ServerFault.com. Where StackOverflow is targeted at programmers, the new site aims to help “system administrators and IT professions”. The makers suggest that if this were a trilogy, then this sequel is a “darker, more serious movie. Some say the best in the series.” A trilogy? Indeed, as a third site called SuperUser.com is planned – for “computer enthusiasts and power users”.
As good as StackOverflow is, I was recently reminded how free, as good as it is, sometimes only gets you what you paid for. I asked a similar question (about htaccess speed optimization approaches) on both StackOverflow, and paid Q&A site Uclue.com (the unofficial successor to the canceled Google Answers site). At StackOverflow I received two very short answers, one of them helpful. But at Uclue, a researcher there got involved and took me through a multiple step process with many details, offering lots of help on the way.
Video of Android OS Running On Laptop
The video shows a Compal Qualcomm powered Smartbook, it says, from the Computex 2009 tech show.
In other news, according to the Financial Times, “Acer ... said it would begin shipping the world’s first mainstream notebook computer using Google’s Android operating system in the third quarter of this year ... The marriage of Acer’s low-end Aspire One netbook with the Android operating system, originally designed for mobile phones, also reflects a continued blurring of the lines between smartphones and computers.”
[Hat tip to Reddit/ Business2Press.]
Friday, June 5, 2009
Google’s Tetris Logo
Google Australia, China and some other places where it’s Saturday already are displaying a special logo doodle celebrating Tetris. The alt text reads “Celebrating 25 Years of The Tetris Effect - courtesy of Tetris Holding, LLC”, and the logo is linked to a search for tetris. Wikipedia knows: “Tetris ... is a puzzle video game originally designed and programmed by Alexey Pajitnov. It was released on June 6, 1984”.
And did you know? One of the original makers of Tetris is now working at Google – Google Australia, specifically, on the Google Wave project. Sydney Morning Herald reports: “Vadim Gerasimov was just 15 when he was taken under the wing of two Russian computer engineers – Alexey Pajitnov and Dmitry Pavlovsky – at the Moscow Academy of Sciences and helped them create Tetris ... Since around 2003 Gerasimov has lived in Australia, where he worked in complex systems research with the CSIRO [Australian Commonwealth Scientific and Research Organization] before joining Google about a year and a half ago.”
[Thanks Hebbet, Tom & Tony!]
Retro Google
Retroogle lets you google like it’s 1999. Sort of... the design is from those years, but the results aren’t. For some actual older results, I’ve mirrored parts of Google from 2001 (from the archive which Google later took down).
Also see Google.com 1997-2007.
[Thanks Ben!]
Thursday, June 4, 2009
Google’s Page Speed Optimization Add-on
Google has released a Firefox add-on called Page Speed. It integrates with another add-on, Firebug, and is aimed at web developers trying to make their pages faster. “Page Speed is a tool we’ve been using internally to improve the performance of our web pages,” Google writes.
To use this, once you installed Page Speed and restarted Firefox, expand Firebug and switch to the Page Speed tool tab. Move to your page, then click the Analyze Performance button. After a short loading time you’ll be presented with a handy list of things you did right, and things Google thinks you did wrong. The latter can be expanded so you can read up on the help provided in regards to issues like “Leverage browser caching”, “Remove unused CSS”, “Combine external JavaScript” and so on (you can also click on the entry to be taken to a longer explanation). Neat!
Also see YSlow, another Firebug add-on (this time by Yahoo) that “analyzes web pages and suggests ways to improve their performance”.
Google Squared Declares the Living Dead
Google Squared, one of Google’s fun but more problem infested products, has the macabre habit of declaring random living people dead. Take the query actors, for instance, which will automatically include a column titled “Date of Death” telling you that Clint Eastwood died in October 2008. According to Squared, Robert De Niro died on August 17, 1943, preceded by Al Pacino, who died in 1940. Dustin Hoffman died on January 28, 2003.
Enter a different query, like presidents of the russian federation, and – as Svetlana found out – there will be a colum titled died which lets you (correctly) know that Yeltsin passed away in 2007. Putin, on the other hand, is not among the living since 1952, Google suggests. Current president Medvedev supposedly died a little over two weeks ago.
If a table of choice doesn’t have a Date of Death column, you can add it yourself. When I did so for the query us presidents, I was informed that Barack Obama lived from 1961 to 1982. Google Squared’s source for this? Flickr.com, the cell says, with “Low confidence”. (When a value is given low confidence, it’s apparently printed in a lighter color than other cells, but this is easy to miss.) The Flickr source is actually a screenshot showing Google Squared’s mistake – the error is starting to feed on itself – while other sources Google shows for the 1982 value are from sites like Freebase.com, but referring to Barack Obama Sr.
Milivella yesterday in regards to Google Squared going live commented, “Quick impression: it has great prospectives, but in this moment it’s far from Google standard (even if you consider Google products at their launch date)”. Milivella added, “Yes, I know that it’s in Labs.” Or perhaps that’s what a laboratory should be for?
[Thanks Svetlana!]
Google Street View Offers an Additional Navigation Option
Another Street View car spotted in the location above
Recently Google Street View saw the addition of user photos for certain locations. Now, Google’s impressive photographic mapping of the world got a little more impressive, thanks to a new navigation approach. Your mouse cursor will now carry a 3D shape that floats over the ground and building walls. “We have been able to accomplish this by making a compact representation of the building facade and road geometry for all the Street View panoramas using laser point clouds and differences between consecutive pictures,” Google says. Double-click the shape – which Google calls “pancake” – and you’ll jump straight to where it was positioned, allowing you to get a quicker impression on points of interest, but also to travel forward much, much faster.
While the new navigation improves the app I did not find it completely intuitive yet. For instance, I expected that using the scroll-wheel would allow me to zoom out of locations, which didn’t work. (And sometimes, when you use the “-” button to the left to zoom out instead, you’ll be kicked out of the Street View mode and back to the map.)
[Hat tip to WebSonic, Hebbet and Reto.]
Differences in Google Image Results for Tiananmen Square Protests
The first page of the around 44,000 results for tiananmen massacre on the international Chinese version at Google.com:
The 11 results in China on Google.cn for the same query, along with a censorship notice at the bottom:
The results of a Google search may vary greatly in different locations around the globe. Above is a search for tiananmen massacre – once on Google international Chinese, and once on Google China – in reference to the Chinese protests culminating in violence 20 years ago today. You will get similar differences when searching for Chinese phrases like 六四事件, June Fourth Incident. This is due to Google having agreed to censor the China results when they decided to further move in the country with their technology in the years 2004 (starting with Google News) and 2006 (organic web search and other services), a compromise which they suggested they hope to be for the better, overall, but to which they do not provide many details to the outside world. “All history was a palimpsest, scraped clean and reinscribed exactly as often as was necessary.”
Wednesday, June 3, 2009
Google Squared Is Live
Google Squared is live, and it’s a lot of fun... albeit the info presented is far from always being accurate. Here’s how you use it: just enter the name of a group of things, say, I entered comic book character. Google Squared then calculates for a bit, and presents you with a table of attributes. In the case of comic book characters, the automatically found columns were name, image, description, publisher, first appearance, and author.
The cells shown are all very fitting in this case, but the values are sometimes wrong. For instance, while Wolverine’s publisher is indeed “Marvel Comics” and his first appearance is indeed “The Incredible Hulk #180” as Google presented, his author is not “NineInchNail”. In the row for Spider-Man, a Superman cover is wrongly displayed, and the publisher is shown to be “Activision” (well, “Activision” is not wrong per se – they may publish Spider-Man games – but it is intuitively wrong in the context of the other values of the column, which refer to the comic book publisher).
Still, I’m really having fun. Especially because you can add your own columns on the fly. For my character table I added “Real Name”, and Google Squared got a couple right, and a couple wrong. Note you can change your query and some values may change along; when I named my square “superhero” instead of “comic book character”, some of the values ended up different. The same when I used the plural of the original query, comic book characters (does Squared prefer the plural form?).
For each created cell, you can expand the source URL of the data bit by clicking in the cell. A source may be a site like Wikipedia, or another third-party website (like ComicVine.com, in my sample). As usual, Google prefers automatic web-wide gathering & mining of data instead of having human editors create non-scalable content, so Google Squared is another step on their knowledge mining road (the old Google Q&A onebox is kind of related, as is the Google Spreadsheets GoogleLookup function).
Once you are happy with the table, you can log-in to save it. It will then show under the “Saved Squared” list. You can also share it by copying the URL to send it around – try this one. One thing I was really missing from this current version (or just didn’t find) was a way to export the data, though, like to some CSV structure or so, or have it be saved as a Google Spreadsheet. Another limiting factor is the number of items you can create; I was only able to go up to 50 for my sample, and had to grow the table in steps of 10. I’m curious how and if Google will continue to develop this little app, and I’m also curious if they’ll find ways to push the data towards more accuracy in the future.
I’m interested to read what tables you create and how they work for you. Here are some of the samples provided by Google on the Squared homepage: US presidents, British poets, digital cameras, dog breeds, arctic explorers, African countries, and romantic movies.
[Thanks WebSonic and Franta H.!]
Update: For an apparent Easter Egg, try entering magpie team. So perhaps “Magpie” was the codename for Google Squared, and these are the creators, as you’ll see names like Tony Aiuto, Randy Brown, Daniel Dulitz along with descriptions, thanks, and favorite queries. [Thanks John Hartley!]
Project Natal for XBox 360 (Videos)
Tuesday, June 2, 2009
From SketchUp to First Person Shooter
Last month, the authoring tools to zombie shooter Left 4 Dead were released. Part of the package (if you own the game) is a pair of plug-ins for Google’s SketchUp. “You may use these plug-ins to create building blocks and props that can be imported into Hammer for use in game levels,” the makers write at their blog. [Via Reddit.]
Google "Holodeck" Video
Brian Ussery made this video of the “Google Holodeck” for Street View while at the recent Google I/O conference. [Thanks Brian!]
How Well Does Bing Handle New Stuff?
Microsoft’s search engine Bing is doing pretty stable on certain types of queries. But one thing Google has been doing exceedingly well in the last years is proper indexing and ranking of very new stuff. (Not too long ago, it was surprising and noteworthy when Google added a blog post below an hour or so... these days, it’s almost expected!) For a comparison, I wanted to see how well Bing does on more fresh topics. The Google & Bing comparison tool comes in handy for tests like these.
What did I find? Note my tests were non-representative; I simply mostly took a couple of items from the Waxy.org link blog, as it contains new stuff. In the end, quite a few queries resulted in Google and Bing doing equally good while quite a few other queries resulted in Google doing substantially better. There was no instance during my tests in which Bing was beating Google when tested on these fresh items, though.
Below is a list of where Bing failed, specifically (admittedly, “failure” is somewhat subjective in search engine rankings, and my search queries may be “biased” in some way due to mostly coming from Waxy). Disclaimer: the queries were performed yesterday and today, so things might have changed in the meantime; also, I only took into account web search results on both search engines, and not news oneboxes:
- [dj & the fro] - on Tuesday, this worked on Google (bringing up the new MTV series), while it failed on Bing (bringing up a MySpace DJ unrelated to the series, same for auto-completion; MTV.com is not found on the full first SERPs).
- [project natal] - Google shows the Xbox Natal homepage first, an Engadget review second, and CNet third. Bing does not show xbox.com in the first page of SERPs at all, but shows a potentially spammy blog – latestknews.blogspot.com which says “CLICK HERE: www.loss-weigh.com”, though at least the page also contains an apparent copy of a Natal review.
- [Tales of Monkey Island] - Google shows the official actual Telltale homepage with info about this (including a trailer). Bing shows a page merely embedding a trailer; that’s not wrong, but also not quite as good.
- [Zoho CEO on Google Wave] - Google at the top shows the actual recent post where the Zoho CEO comments on Google Wave. Bing shows an old post (from January 2008) where the words “Google Wave” merely appear in a “latest comments” widget.
- [tim oʼreilly google wave] - Google finds the actual blog post by Tim on the subject of Google Wave, at radar.oreilly.com. Bing puts a profile page about Tim at spock.com first – radar.oreilly.com appears only once in the SERPs, but with an unrelated post from 2007.
- [us government data] - Google shows the new data.gov among the first page of SERPs, Bing doesn’t have data.gov on the first page at all.
- [kevin fox nilla] - Google shows most relevant blog post as first, Bing as second entry.
- [gmail automatic message translation] - Google ranks the official post on this recent feature first. Bing ranks the official post 4th, and puts someone merely copying the original source (without credit, on a blog titled “Latest SEO news - Trick n Tips”) first.
- [Classic Movie Quotes, as Translated to Japanese and Back Using Google] - this was the title of a post from Blogoscoped, posted half an hour ago at the time I searched. Google had it already and presented it at #1, whereas Bing didn’t have it at all on the first page the time of searching (Bing lists mostly general sites instead, like iGoogle, YouTube, or Answers.com).
Here are a couple of queries where I found both Google and Bing doing equally good, if differently at times – note these were not all the queries where they both did well, I’m just showing a limited set of samples:
- [Mozilla Jetpack] - both different, both good.
- [Goodbye GM by Michael Moore]
- [Twitcaps]
- [Star Wars first draft script]
- ...
But fresh queries isn’t all there is to a search engine! It might be Bing is doing better on another type of queries (comments on this welcome!). It’s also worth noting that the difference between best search engine and second or third contenders is growing increasingly small these days... Microsoft generally did a very solid job, and they find the “correct” page on many queries.
Getting search as right as Bing must take an impressively huge effort, though that also raises the question of “why” – which problem are they trying to solve? Which deficiencies of Google do they want to make up for, what niche are they trying to target, in which ways do they want to offer more to users than Google already does? And if for the sake of argument we assume Microsoft has solved the problems Google tackled several years ago, then how many years will it take for them to solve the problems that today’s Google handles well, like ranking fresh stuff? And if they do ever catch up, isn’t there a good chance Google already reached the next level by then?
[Hat tip to James!]
Classic Movie Quotes, as Translated to Japanese and Back Using Google
“Kidd is here, I’m watching”
- Casablanca
“I will be in need of a big ship.”
- Jaws
“I denied him that he is not going to provide.”
-The Godfather
“yipe emphasis on form - Yay, motherfucker.”
- Die Hard
“And you, and cheese, or hit district of Paris called?”
- Pulp Fiction
“May you please force.”
- Star Wars
“Sisters, do not fight in here! This is the War Room.”
- Dr. Strangelove
“Honestly, my dear, I do not care”
- Gone With the Wind
“I am in Kansas is like no other.”
- The Wizard of Oz
“It means you can not say I’m sorry love.”
- Love Story
“And she always has to live like a box of chocolates. You’re going to make anything that is not known.”
- Forrest Gump
“One has to ask themselves two questions: ’I feel lucky?’ Well, what ya punk?”
- Dirty Harry
“HASUTARABISUTA, baby”
- Terminator
All Your Base Dialogue, As Translated by Google
Andy Baio recently pointed to the Wikipedia page which contains an actual proper translation into English of the old "all your base" dialogue from the game Zero Wing. The following is a translation based on the Japanese original also available in that wiki entry, courtesy of the Google Translator:
2101
The war has just begun.
Captain: I do say that one!
Engineer: someone who is so bomb.
Operator: captain! Communication got!
Captain: take nothing!
Operator: the vision comes to the main screen.
Captain: Wow you are! !
CATS: You look to the past, gentlemen.
CATS: the cooperation of the federal government,
you are our base, all you CATS.
CATS: We also ship you will end soon.
Captain: it fit my gosh!
CATS: You have to help us appreciate.
CATS:残RI少NAI life at best, catering to it.
CATS: HAHHAHHAHHAHHAッ
Operator: captain.
Captain: ZIG-marching orders to all! !
Captain: They have to leave now.
Captain: We hope for the future of
Captain: I beg. ZIG! !
Monday, June 1, 2009
Google AdWords Redesign
Google has launched a redesign of their ad campaign management tool, Google AdWords. Google lists the new features and tells you what’s changed.
Microsoft’s New Try At Search Is Live
You might have heard the news of search engine Bing – Microsoft’s rebranding effort after Live failed to take over the search crown from Google – going live. Expectedly, the web search results are often copying Google’s style but are of [edit]somewhat lower quality on a couple of queries I tried (but please add your comments in regards to search results quality)[/edit]. One interesting feature however is the video search, because it offers direct animated previews as soon as you hover over a given video thumbnail; combine this with the site operator, as in the search site:blogoscoped.com*, and you can get a good, uncluttered overview of videos shown or embedded on a given domain. The same search on Google Video returns... zero.
The image search of Bing – a word meaning “cold [mp3]” (as in fridge [mp3]) as well as “sick” in Chinese, among other meanings depending on the tones** – offers one noteworthy difference in comparison to Google Image Search. When you click on an image from the result thumbnail list, you’ll land on the target page with the image on top as with Google. But on the left side, you can also see other thumbs from the result page you were on before, so you can quickly click through the rest of the images. However, on the downside, you can’t use shift-click on a thumb to open a page in a new instance... this greatly lowers usability for people who like to control their browser windows or tabs.
A nice thing about Bing is that they automatically insert in-page anchor links below some snippets. A search for scientology wikipedia, for instance, has links like “Etymology”, “History”, “Organization” and more pointing straight to that part within the Wikipedia.org page. (There’s also an “enhanced view" for Wikipedia pages, which basically reprints the – GNU-licensed – wiki entry and adds an always-visible navigation box to the right.)
Just as with Google these days, Bing also has auto-completion. This picture was posted on Reddit under the headline “Seems like bing.com is indexing a different [Internet] than Google”.
*I’m not sure if the “site” operator is actually supported as that; while I did get different results when searching for site:example.com vs just example.com, there may be other factors in play here.
**Do a Pinyin Look-up for “bing” to find more meanings.
[Thanks WebSonic! Audio by Yinxue.]
Barcode-scanning Your Bookshelf for Google Books
Google’s Matt Cutts shows how he adds books to his Google Books “My library” feature... using a below-70 bucks barcode scanner which he USB-connected to his computer. Why would you want to do that? One reason is to make your book shelf searchable. If you don’t have a barcode scanner you could also search for the book title to add the book of course... I could imagine it’s not that much slower.
“My library” is a nice feature from Google Books (I copied it for one of my sites!). Perhaps for the future, Google should allow us to simply photograph our book shelf – for those who still have book shelves – and then apply some OCR magic or image comparison to save us even the job of barcode scanning.
In other Matt news, there’s an SEO related presentation he did and for which he put up the slides. There’s no video there yet but some of the slides are more or less self-explanatory. Slide 32 of 50, titled “Update often”, is a jab at Sergey Brin’s “blog” (called Too and opened in 2008, it consists of only two posts).
[Thanks WebSonic.nl!]
Friday, May 29, 2009
See Google Wave in Action (Video)
Google released a video from the Google I/O event showing their Google Wave program in use. I've jumped to 7:32 where the actual demo starts off. This video might be best enjoyed full screen.
Please comment in the existing thread.
Google Increases White Space Around Logo on Result Pages
It looks like Google increased the margin above and below their logo on the SERPs. Anyone dares speculate why?
I asked ex-Google employee and Gmail (and more) designer Kevin Fox to guess about why Google did the change, and what effect it might have on the user. Kevin says, “Greater breathing room. If Google could improve their rankings to the point where the ’right’ answer was just as likely to be in the first N results as it used to be in the first N+1 results, they can pass the savings on to the user by showing them less clutter. If Google was confident that the right answer was always in the top 3, they’d probably just show 3 and have a lot more whitespace. :-) But that’s all just speculation.”
One of the last design changes to the Google results was making the blue bar on top lighter in color, around half a year ago. Even though that could be considered a subtle change, Google actually moved the blue color from old to new in not one but two steps, meaning there was an intermediate blue used for a while... perhaps to make the change less noticeable to users:
On a related note, also recently, Google announced a change to their logos.
[Thanks Johan!]
Thursday, May 28, 2009
A Sneak Peak at Google Wave
Over the last years the web has moved towards more real-timeness and collaboration, with plenty of online communication forms to choose from. Google now is showing off a preview of a new product, framework and protocol called Google Wave. I’ve not seen this in action yet but judging from the screenshots and descriptions, Google Wave (the product) is an app running in modern browsers which is part wiki, part chat, part collaborative office and sharing tool, and part email.
Take a look at the screenshot below, via Google:
To the left side you’ll see a navigation pane offering you to view the Inbox, Active threads – or “waves”, as Google calls them – a history, spam waves and more. The inbox lists threads in which you participate, because (if I understand it right) you started them or were invited. A particular thread can consist of a mail-type conversation with a friend, like “BBQ on Sunday,” or a document sharing photo snapshots, which are drag-and-dropped into the document window.
That document window, to the right, offers the multi-user real time view onto the document, and you apparently can edit in different places within it. You can also play-back the history of the document to see how it evolved. Avatars of the people participating in a particular wave can be seen on top. If two people are online at the same time then “a wave acts just like an instant message – except that you see each character as it is typed,” Tim O’Reilly writes. “When you’re talking with someone, you know what someone is saying before they finish their sentence. You can respond, or even finish their sentence for them. So too with Wave.”
Search is provided at the top of the middle pane. It’s apparently real time, too, and searches can be saved and added to the left side navigation.
Sounds cool? Let’s see what Google will release. “Google Wave will be available later this year,” Google writes on their Wave homepage, and the following pieces are available elsewhere:
- The Google Wave API. The API is supposed to allow “developers to embed waves in other web services and to build extensions that work inside waves”.
- The Google Wave Protocol letting you build on top of the (open-sourced) platform.
- A Google Wave press site with photos, team info and more
- A Google Wave Sandbox, but you need the right login.
- A Google Wave Developer blog. Co-editor Tony stumbled upon the Google Wave product name and blogs earlier this month and was wondering what it was... now we know!
- Google’s introductory blog post, offering a bit of historical origin of the service (the idea was thought up by Sydney office based brothers Lars and Jens Rasmussen of the Google-acquired Where 2 Technologies). Lars writes: “Could a single communications model span all or most of the systems in use on the web today, in one smooth continuum? How simple could we make it?”
- Brian Ussery, who reports back for us from the Google I/O conference, has more pics from the presentation.
- Explanatory videos at YouTube.
The following screenshots further show off the functionality of the program, courtesy of Google, and if you want to know more, Techcrunch has a good write-up too. Without a real live app to test we need to keep in mind though that Google may talk about all the glory of the app but hardly jump to point out potential quirks, bugs and shortcomings. I’m curious how well the app will live up to the very interesting bits we were offered, which browsers exactly it supports (and at what speed), and how its features compare to those of blogging platforms, chat programs, mail clients, wikis, real time collaboration tools like Etherpad, and social news sites like Friendfeed.
Collaborative (rich) text editing
An extension to Wave handling event planning
A map is included
Another wave extension: a game
Adding a participant for the wave
Wave’s context-aware spellchecker
[Thanks Franta H., Brian, Jérôme, James & Tony!]
Android 1.5 Being Rolled Out
Ron of Totlol.com sent in below three images and writes:
Early this morning Google installed over the network the new version of the Android operating system on T-Mobile phones in the US.
Notable changed spotted early on:
- Amazon MP3 app pre-installed.
- a unified"search/url” bar for the browser called “Go” that defaults to Google search.
- a nice way to display embeded YouTube videos that makes them clicakble.
- a lot of eye-candy
Dpic in the forum also has the new release – codenamed Cupcake – and comments that it doesn’t “exceed” his expectations but “definitely meets them”. DPic says, “This basically brings android up to the standards i had originally expected. Everything seems more polished and solid.”
[Thanks Ron and DPic!]
The Ideal Amazon Site Would Not Show a Search Box...
Greg Linden writes:
Long ago, when I was at Amazon working on personalization, we used to joke that the ideal Amazon site would not show a search box, navigation links, or lists of things you could buy. Instead, it would just display a giant picture of one book, the next book you want to buy.
[Via Ionut at Friendfeed.]
Google Apps Script: Expanding the Google Office With Your Own JavaScript
A public document footnoted with “Google Confidential” mentions Google Apps Script, a framework that looks like a kind of macro language (and more) for Google’s web office. The terms of service mentions a “Trusted Tester Agreement”, but Google did blog about this so it’s no secret; Google says “we’re inviting a limited number of Google Apps domains – about a thousand organizations – to start playing with Google Apps Script and giving us feedback so we can quickly understand which tricks would be the most beneficial to learn next.” (A preview sign-up form is provided, too.)
According to the intro page, with scripts you can:
- “Create your own custom spreadsheet functions”
- “Automate repetitive tasks (e.g. process responses to Google Docs forms)”
- “Link multiple Google products together (e.g. send emails or schedule Calendar events from a list of addresses in a Spreadsheet)”
- “Customize existing Google products (e.g. add custom buttons or menus to run your own scripts)”
To edit a script, you’ll be using JavaScript within Google’s web editor (or another desktop tool of choice). Some basic sharing is supposedly supported as well. However, the JavaScript is not executed client-side, but server-side instead. “As a result, direct operations on the client-side DOM are not supported, although some restricted functionality is provided.”
The introduction links to a documentation which mention “Google Web Scripts" for the services Google Base, Calendar, Mail and Spreadsheets. The following sample images are provided in a Google tutorial:
Google explains, “A dialog box should pop up in the Spreadsheet and request you to enter a number. Enter one and click ’Ok’ ... Then, the spreadsheet cells A1 and B1 are updated. Finally, a second dialog box shows the value you entered plus 1.”
The following screenshot from a different tutorial illustrates adding your own menu entries to Spreadsheets:
Another tutorial, labeled “Advanced”, aims to show how to “collect information from different users in a Spreadsheet using Google Forms, then leverage it to generate and distribute personalized emails.”
[Thanks Jérôme!]
Wednesday, May 27, 2009
Google Web Elements and More Announced at Google I/O
By Brian Ussery & Philipp Lenssen
Google at today’s I/O developer conference announced Google Web Elements. Web Elements lets you simply paste a URL from a public Google spreadsheet, calendar, presentations and more, and then offers you some gadget code to paste into your site. Also, you can add modules like custom search, maps, or custom Google News. Mostly, the stuff behind these Web Elements was already available through Google’s apps. (For instance, in a Google Spreadsheets you can pick Share -> Publish as a web page -> HTML to embed in a page.)
Also at the Google I/O event, Google says that Java for the App Engine is now rolled out for everyone (it was announced as “early look” already last month). It’s the second supported language for Google’s framework (after Python).
If you’re curious about the I/O event, Brian Ussery is there and sends us some pictures. Brian tells us, “This year’s conference focuses on web standardization via HTML 5”. For more on that, see Google’s “Choose Your Own HTML5 Adventure” page.
Google IO 2009 areas of focus
Breakfast at the event
An interactive map of Google I/O
Google CEO Eric Schmidt
The “holodeck“
“Canvas is Intrinsic to the Web”
Brian continues, “In addition to the demonstrations, Google announced that each of more than 4,000 conference attendees will receive a free Google phone.”
[Thanks Brian for the photos and report, thanks Tony & Jérôme! This post may be updated as new information comes in. Tags: io2009]
Google Holodeck for Street View
By Tony Ruscoe
Some of you may remember seeing screenshots of a humorous email sent by a Googler requesting a holodeck to be installed in one of the Googleplex buildings.
Perhaps that wasn’t such a joke after all. Anyone attending the Google I/O developer conference in San Francisco can try out the Google Street View Holodeck for viewing Google’s street-level photographic imagery in almost 360 degrees.
As the board outside the holodeck explains:
Holodeck
The holodeck began as a tool for Street View engineers to evaluate picture quality. But it’s also a great way to experience other parts of the world.
Over on Search Engine Land, Danny Sullivan has posted more pictures from when he first saw the Holodeck at Google in October last year.
Monday, May 25, 2009
Google Homepage With Yellow Ribbon
Why does Google have a yellow ribbon on their homepage today? It’s because it’s Memorial Day in the US. Wikipedia writes, “Memorial Day is a United States federal holiday observed on the last Monday of May (May 25 in 2009). Formerly known as Decoration Day, it commemorates U.S. men and women who died while in the military service.” In the past, certain groups and people criticized Google for not changing their logo on this day, while others have congratulated Google for ignoring the holiday.
The Google homepage also showed ribbons of other colors before. A red ribbon was put up on World AIDS day, and a black ribbon was put up during bomb attacks in Madrid in 2004, and London in 2005.
Also see the Google Homepage FAQ.
[Thanks Ionut!]
Sunday, May 24, 2009
A DoubleClick Ad For a Link Buying Service
When you enter http header checker into Google, one of the first or the first result is a page by Webconfs.com. On this page, I noticed an ad for a “link building service.” The ad is by Google’s DoubleClick, and it points to a page at ForumLinkBuilding.com which promises that “Your links will come from pages with only 5-15 outbound links!”, that “Some of your backlinks will come from high Google PageRank pages: up to PR6!”, and that they “use a unique link building technique: our links will come from high quality forums with high Google PR - up to PR9!”. Buying links with them will improve “your website rankings for your targeted keywords on Google”, they say. The also write that they “follow all Google Guidelines”, but doesn’t this look a whole lot like text link buying, which is against Google’s webmaster guidelines?
YouTube’s Copyright Notice on Muted Videos
Not new, but I just noticed that on the bottom of this YouTube video showing how a 80x80x80 Rubik’s cube is solved, there’s a copyright notice. It reads: “Notice ... This video contains an audio track that has not been authorized by WMG [Warner Music Group]. The audio has been disabled. ... More about copyright”. The link points to YouTube’s copyright tips.
How many videos are muted like this? Hard to tell exactly, but the Google search query ["This video contains an audio track that has not been authorized by” site:youtube.com] returns around 342,000 hits (including some help pages, and perhaps some regional YouTube content too).
On a related note, Neowin.net on May 21st reports that “YouTube is to run a trial of ’pre-roll’ video ads on selected content from several broadcasters starting from today. The pre-roll ads are short video ads shown before the video you requested and could be up to 30 seconds long. Advertisements will start appearing around videos from BBC Worldwide, ITN, Discovery, National Geographic and – from today – Channel 4.”
[Thanks WebSonic.nl!]
Thursday, May 21, 2009
YouTube vs Porn Day
According to CNet, Google-owned YouTube “was busy deleting porn videos on Wednesday after users of forums at a rival site and an imageboard site declared a ’Porn Day’ campaign against the popular video service.” The sites in question are said to be Ebaum’s World [perhaps not: see update] and 4Chan, and Ars Technica says some of the tags used to upload the videos are “marblecake” and “jonas brothers”. CNet writes:
As one might expect, the pornographic clips are being uploaded without any indication that they’re for adult eyes only, making them easy to happen upon by casual searchers. As the upload-fest has progressed, users are also uploading what seems to be legitimate content, but is in fact a porn video that simply has 20-30 seconds of non-porn content (a newscast, an interview) at the beginning.
Searching YouTube for the tags in question, you’ll notice a couple of NSFW thumbs, and a lot of messages reading “This video has been removed due to terms of use violation” and “This account is suspended.”
[Via Reddit.]
Update: Ebaum’s World might not have been part of this at all... see the comments for more, there’s indicators here that they’re merely being wrongly blamed by 4chan.
Book on Using the Google App Engine
Charles Severance’s “Using Google App Engine” is a new book by O’Reilly. According to the O’Reilly* description, it’ll help you build “scalable web applications” using Google’s framework, touching on the subjects “Python, HTML, Cascading Style Sheets (CSS), and HTTP”. A sample chapter [PDF] is available.
[Via DeWitt.]
*disclaimer: I wrote a book for O’Reilly before.
Wednesday, May 20, 2009
Translate Your Gmail Inbox Messages With a Click
This has got to be my favorite Gmail feature rolled out in the Labs: message translation. Enable it in the Labs settings, and next time you see a mail written in a language other than your own, you can hit the link reading e.g. “Translate message to: English” above, and the message will magically appear in English. (Well, it would be magic to someone from a decade or two ago, even though we start to get used to these things!) To change your default language for this tool, go to the general settings tab, and you’ll find a selection box labeled “Default translation language.”
I wonder if Google will roll this feature out for everyone after initial public testing and feedback-gathering. Some of the Gmail labs “experiments” seem to be too useful to have them be lost in opt-in settings not everyone will see.
These days, translations of texts we don’t understand are often triggered with a click. In the future, as translation quality improves, perhaps instead of having to click to translate, the software could simply add a disclaimer footnote that “this message was auto-translated for you” with a link to the original. It would also not strictly be necessary to always implement this on a per-app level; it could be built into a tool like Google’s Chrome browser, automatically translating every web page you visit, as you visit it. There are two sides to the coin here: on one side, mistranslations can cause misunderstandings, which in turn could cause all kind of misguided actions (admittedly, nothing new in the history of communication). On the other side, there’s potential for a deeper understanding of other cultures, once you’re able to browse the foreign web as if it were your native language one.
[Via Hebbet and Niranjan, who saw it at the Gmail blog.]
Game Based on Google Earth 3D Plugin
Ships is a simulation game letting you steer a ship through different locations – and it’s all based on the Google Earth 3D plugin. This app by Paul van Dinther comes complete with sound effects, keyboard controls, and a proper good looking interface. However, note you can drive your ship everywhere as there’s no collision detection. [Thanks Juha-Matti, who saw it at the Google Earth Blog!]
Tuesday, May 19, 2009
Mistakenly Logged In to Google As Someone Else?
Several users from Singapore are telling me that their Google accounts get mixed up. Derick N. at his blog posts about how Google thinks he’s another user, and how furthermore, messages are posted under his name which he did not post.
Priyadarsan Venugopalan, who first contacted me about this back on April 3rd, tells me that while he was browsing the Google Apps help center, he suddenly realized “that instead of my email ( priyadarsan...@gmail.com ), the interface is showing Maximilian...@gmail.com. I thought it’s some kinda minor bug and it will be prompting for my account once i refresh. It didn’t happen. I tried posting a message in one of the thread and to my surprise, the post came in as Maximilian...’s. I dropped an email to the account holder as well.”
Priyadarsan says he was asked to login when he tried to access Gmail or the My Account page, but that other pages were browsable (he says he was able to access data from Google Notebook, Google Profile, iGoogle, Google Video, Analytics and more). He also says this happened in both Chrome and Firefox, and that he’s not using a proxy. “I posted the issues at security@google.com, which went unanswered,” he tells me today (I’m not sure how long the wait was so far).
Peter B., another user whose account was mixed up (and who also uses Google Apps), says “At the time, I though perhaps google was somehow defaulting to a ’pretend’ person until my own profile was created... obviously this mustn’t be the case.”
[Thanks to everyone involved!]
Google Calculating Who’s About to Quit
The Wall Street Journal today writes:
[Google] recently began crunching data from employee reviews and promotion and pay histories in a mathematical formula Google says can identify which of its 20,000 employees are most likely to quit.
Google officials are reluctant to share details of the formula, which is still being tested. The inputs include information from surveys and peer reviews, and Google says the algorithm already has identified employees who felt underused, a key complaint among those who contemplate leaving.
Laszlo Bock, responsible for human resources at Google, is quoted saying “We haven’t seen the most critical people leave”. He also says that the algorithm helps Google “get inside people’s heads even before they know they might leave”.
Tags: The Algorithm, Philip K. Dick
[Hat tip to /pd and Kevin Scott.]
Monday, May 18, 2009
Designs Google Tested for Their Search Options Feature
Last week, Google at their blog published below image of alternative designs of their recently rolled out Search Options feature. “The team began by generating a broad range of directions, which we refined through design sessions and reviews. Confident about a handful of concepts, we began to seek out user feedback using eye-tracking and usability studies.”
The design that’s currently live:
Commencement Addresses of Google’s Eric Schmidt, Larry Page and Marissa Mayer
Google made several videos of university commencement speeches available:
- Larry Page at the University of Michigan (the sound issues with this one have been corrected; also see the transcript)
- Eric Schmidt at Carnegie Mellon University
- Marrisa Mayer at the Illinois Institute of Technology (Marissa also appears in another recent video)
[Thanks Jérôme and Roger!]
Google Street View Tricycles
A couple of photos of a Google Street View Trike can be viewed in a Picasa web album (there’s another one at Flickr and via the Google Japan blog). I believe this is official press material from Google, as is the video presented at PressAtGoogle, which I’ve uploaded to YouTube.
Sunday, May 17, 2009
Google Squared Live... Kinda
By Tony Ruscoe
When Google recently announced they’d soon be launching their Google Squared service, this wasn’t what we had in mind...
Google Squared will be coming soon. For now, why not have fun squaring some numbers?
If you want to have fun squaring some numbers, you’d better head over to www.google.com/squared now before the real version is made live!
And if you’re after even more fun, try entering these Easter eggs:
- times
- be there
- alpha
- 42
You can see some screenshots of the real Google Squared here.
[Thanks beussery!]
Thursday, May 14, 2009
Google Shows Survey Questions on Some Health Searches
Google announced that they’re starting to show survey questions at the bottom of a small percentage of health-related search queries. For instance, a search for headache may trigger the question: “Did you search because you or someone you know has a headache?”
Why is Google doing this? They say getting better stats can help with flu trends detections, as well as improve the relevancy of Google’s search results in the future. If you participate in the survey, no personal Google Account data will be associated with the results, Google says, though they will in some way use it together with their search server logs, as the FAQ explains (“Google’s servers automatically record information when you respond to this poll, including a cookie, IP address, browser type and language, and the date and time of your answer”).
Search Engine Land has another screenshot, and a couple of more survey questions:
- “Did you search because you or someone you know has ankle pain?”
- “Did you search because you or someone you know may be experiencing a reaction to poison ivy?”
- “Did you search because you or someone you know may have streptococcal pneumonia?”
- “Did you search because you or someone you know is taking ibuprofen?”
I’m curious how skewed results to such questions could be; for instance, if I search for “reaction to poison ivy”, I might be more inclined to take the time to answer survey questions if I’m doing this as research for homework, versus when I’m actually experiencing the reaction myself.
If you come across such a survey box, please post a screenshot! Google in their FAQ on the subject disclaims though that this experiment will only be displayed for “users from United States IP addresses”.
[Thanks Juha-Matti! Image by Google.]
Hypothetical Question In Regards to AdWords
You are given $1 million to spend on Google advertising (to be shown in Google search results or on the AdSense network). What do you advertise?
Google Slow (or Down) for Some
Some of us are having problems accessing google.com, YouTube, Gmail and others. [This post may update if there’s further info.]
Update: And it seems to be back up now (18:15 CET).
Update 2: Google explains what happens: “An error in one of our systems caused us to direct some of our web traffic through Asia, which created a traffic jam. As a result, about 14% of our users experienced slow services or even interruptions.”
[Thanks JEShack and Tony!]
The Problems With the (Great) Wayback Machine
In the future we might have web archaeologists digging for old sites just like we have them in the real world. It won’t be an easy job either, as a single tool won’t suffice. I just wanted to check what Amazon looked like in around 1999, and the following were the first things I clicked on at the Wayback Machine:
Dec 12, 1998:
Jan 25, 1999:
Apr 30, 1999:
Oct 13, 1999:
The first three screenshots above are broken, apparently not showing what Amazon looked like back then (we get a rather new design, a blank page, and an error page). The fourth sample shot is better and may be what Amazon appeared back in 1999, but how can I trust it to give me the real deal considering the other errors of the Wayback Machine? Should a historian accept this without further verification? And what if I’m unsure if my browser of 2009 shows the same as users saw with older browsers back in 1999? And what if I want to see Amazon.com in its launch yearh when the Wayback Machine has 0 pages from 1995?
What will be the tools with which we’ll recover old websites in the future?
Google Timeline Misses the Mark
There were mentions of the term “Android” as far back as 1270, according to Wikipedia*, but when Google’s Timeline feature finds 1788 in a query for android developer they still miss the mark:
Not the first time Google’s otherwise smart Timeline gets it wrong. In this case, the string in question which confused Google is “The Top 50 in 1788 entries”, which their algos seem to think is a year.
Can you find other Timeline hit & misses?
[Thanks Alistair!]
*Wikipedia says “The term [android] was first mentioned by St. Albertus Magnus in 1270 and was popularized by the French writer Villiers in his 1886 novel L’Ève future, although the term ’android’ appears in US patents as early as 1863 in reference to miniature humanlike toy automations.”
Nostalgic News: Indexing 200 Million Unique Web Addresses
A news bit of 10 years ago, from BBC:
A search engine that promises to index all of the World Wide Web has been officially launched.
Alltheweb.com says it already holds 200m unique Web addresses in its database, or 25% of the 800m URLs (Uniform Resource Locators) estimated to be out there.
Google to Reshoot Japan Street View Imagery After Privacy Complaints
Google Street View Japan
Reuters reports that Google will reshoot all Japanese pictures for its Google Maps Street View feature... using lower camera angles, because there were complaints “they were capturing images over fences in private homes.” Already, Google blurred faces and license plates in the past after complaints.
[Thanks Juha-Matti!]
Update: TomHTML in the comments points to Google Japan’s blog post on this (see translation, which covers more privacy changes), which has before/ after photos of the cam-equipped Street View cars:
[Thanks TomHTML!]
Which Page of Yours Has the Highest PageRank?
To find out which page of your site has the highest Google PageRank, log-in to the Google Webmaster Tools. Click on your verified site (or click “Add a site” and follow the instructions), and then Diagnostics -> Crawl stats (in the new interface). Scroll down on the page and you’ll see which of your page had the highest PageRank, broken down into the previous three months. Usually, the highest-PR page will be the homepage of your site, i.e. the root of the domain. Any of you out there for which that is not the case?
Google Toolbelt Tells Which Of Your Pages Were Recently Indexed or Linked
Google’s new “show options” feature*, rolled out for everyone this week, lets you restrict results to just pages from the past 24 hours. You can use this feature in combination with Google’s “site” and “link” operators, too. Here’s how:
- Finding newly added pages: Say your site is example.com. Enter site:example.com into Google, and you’ll see all pages Google indexed from you. Now click the “Show options...” link on top of the search result. In the left hand bar that appears, click “Past 24 hours”. (You can additionally “Sort by date”, but I noticed that this would sometimes not show all pages.) Doing this for blogoscoped.com, I’m seeing the frontpage (because it was changed), new forum threads, and new posts. (I’ve also spotted a somewhat malformed link that was indexed.)
- Finding new links to your pages: To find backlinks, simply repeat the procedure as above but use link:example.com instead. After you clicked “Show options” and “Past 24 hours”, you’ll find other websites which linked to your site. This can help you uncover the discussion happening around your site. (Note another way to find backlinks is to enter your page’s full URL into Google Blog Search.)
*It’s codenamed “toolbelt” at Google, I’m told.
Update: Michael VanDeMar comments:
The link: command is treated as a text based search, not a special operator, when combined with the date based parameters. It has always been like this.
In many cases mentions of a domain name, especially those that are not keyword based, will also happen to have links to that domain, but that doesn’t make it a link: search, just a mention one.
[Thanks Michael!]
The Google Way (Book)
James Pyles over at the Million Chimpanzees blog posted a review of the book “The Google Way: How One Company is Revolutionizing Management As We Know It”, by Bernard Girard:
Whenever I hear a term such as “the Google way” or “the HP way” or “the giant-Fortune-500-megahuge-corporation way”, I think of the mechanics a large company uses to operate on a day-to-day basis. My understanding of these “ways” is that they result in an evolutionary slowness in getting even the most simple and mundane tasks done (...)
The general theme of this book, is to describe for the reader the origins of Google and how the development of the “Google way” resulted in it’s tremendous success, including possible future directions for the search engine giant. To accomplish this, Girard pulls not just from all of the publicly available information on Google and its founders, but interviews with former Google staffers, general history, philosophy, and classic corporate strategies.
For instance, lessons learned by Ford’s revolutionary creation of the assembly line, creativity and Edison, and the invention of the steam engine, are all brought into play to illuminate the various concepts and methods of developing something that had never existed before.
Another recent book on Google with a focus on business is Jeff Jarvis’ “What Would Google Do?”. “Blog impresario Jarvis uses the company’s success to trace aspects of the new customer-driven, user-generated, niche-market-oriented, customized and collaborative world,” Publishers Weekly writes.
[Thanks Phil!]
Wednesday, May 13, 2009
Google Calendar Tasks Live
Google Calendar went live with the Tasks feature that was spotted a while ago. Just click on the link “Tasks” in the left hand, and to the right side, a widget appears. This widget shares its data and behavior with the Tasks feature you can opt-in to at the Gmail Labs.
Adding a new tasks consists of simply typing at a new position within the gadget; more item features can be added by clicking the “>” icon to the right. To delete a task, you can hit backspace to clear the task text.
[Thanks Hebbet!]
Google Webmaster Tools Redesigned
Google gave their Webmaster Tools site an overhaul. The left hand navigation now has an expand and collapse menu. On their motivation behind the changes, Google explains, “Over the years we’ve constantly been adding and updating new features. The result was a set of tools we’re pretty proud of – but also a site that had become pretty unwieldy and often difficult to navigate ... Features are now grouped together in three categories: Site configuration (where you provide Google with information about your site), Your site on the web (where you can view Google data about your site), and Diagnostics (where you can get reports on any problems we encountered crawling your site). Our goal was to make stats and data more discoverable, while removing redundant pages and features (such as Index stats) that cluttered the interface and reproduced information available elsewhere.”
For instance, here’s the old settings page:
And the same in the new app:
[Thanks Hebbet!]
Tuesday, May 12, 2009
Screenshots of Google Squared
Update: Google Squared is live now!
One of the most interesting announcements at today’s Google Searchology event was Google Squared, which Google says will be appearing in their Labs later this month (apparently at www.google.com/squared). Google’s Matt Cutts at his blog wraps up the Squared demo: “[I]f you typed in ’small dogs’ then Google would try to return types of small dogs, along with facts like how much they weigh. It’s easy to add a row to the Square, so you could add a row for Lhasa Apso and Google will try to infer the relevant facts from the web. You can also add new columns, e.g. if you type ’energy level’ then Google will look for corroborating facts across the web and try to guess the energy level of each type of dog.”
Here are screenshots, courtesy of Google (Google says as this is not public yet, the final release may change):
The Google Squared homepage, complete with a Labs icon, offers a search box and a “Square it” button. Suggested searches include “roller coasters”, “strollers”, and “hotels in Paris”.
The search result page for the query “small dogs”. According to Google, Chihuahuas weigh 6 lb and belong to the group “toy”.
A detail of above screenshot showing different types of small dogs.
Please add your comments to the existing thread.
>> More posts
Advertisement
This site unofficially covers Google™ and more with some rights reserved. You can subscribe to the feed, email your tips and join our forum!