Google Base, the service where people upload all sorts of stuff to a Google database (mostly products, services and information for sale), is now asking uploaders some questions so it can better understand what is in the database. For example, Base will present a user with 30 questions, some looking at the product listing and asking if it is correctly identifying key words (like brand names).
Hopefully these questions don’t just help on the particular listing, but are part of some sort of attempt to make the Base algorithm smarter so it can do an even better job all by itself.
I wonder how popular Google Base is these days. Google’s created some really cool vertical search engines using Base, but without any idea if users are taking advantage of them, there’s no way to know if Base is fabulous or a flop.
Google Base has a bunch of search verticals, mini search engines built on data uploaded to Base, and they’ve changed the results format of those search engines. The old:
And the new:
It’s like this for all the verticals listed on the homepage, including recipes, housing, jobs, hotels, and anything else listed here. It’s a really nice improvement, besides looking better and having more white space, also giving you all those boxes to refine your search, and multiple views for the search results (list view, grid view and map view.
Thanks so much for Arie for pointing me at the new interface.
Google is now reporting how many of its users are subscribed to website’s feeds, by including the subscriber info in the header its Feedfetcher spider leaves when it grabs a feed. This means that if you look at the header, you’ll know how many users combined subscribe to that feed in Google Reader and the Google Personalized Homepage. With Bloglines, Yahoo and Google all reporting counts in an easy way, you can now get a really good idea at a glance how many feed subscribers you have.
Thanks to Google’s move, Feedburner users all across the web (who get the best automated and detailed feed statistics of anyone) are reporting huge jumps in subscriber counts. Yeah, turns out that Google’s Feedfetcher may very well be, in total, the biggest player in the RSS feed subscription game. Darren Rowse had 30% of his subscribers from old #1 Bloglines, now he has 39% from new number one Google and 17% from number two Bloglines.
Also, Randy Morin did a survey of blog search users, finding out that Google Blog Search is the favored blog search engine. Twice as many respondents listed Google as their favorite search over Technorati, showing how easily Google has taken over that market, just as it is muscling its way into RSS readers with a great product.
Hitwise has some info on how Google Base and Google Checkout are doing, with Checkout doing considerably better. While Google has just been burning cash to get users on Checkout, they haven’t really promoted Base at all, and a rumored replacement of Froogle with Base has still not happened. Base’s market share has actually declined over the last half-year.
Google Base now has a podcast section, along with a special promo page asking podcasters to upload their ‘casts to Base. I’m not sure how much free space Google allows per upload, but 7,060 podcasts have already been uploaded, although a lot of that is spam (how is a flashlight or headphones a podcast?). In a sense, this is now Google’s podcast search engine, letting Google have another vertical with minimal effort. For podcasters looking for free, reliable storage, Google Base is definitely worth checking out.
(via Steve Rubel)
One year ago today, Google unveiled Google Base, its vague-yet-powerful repository for “everything”. While Base is proving useful for certain types of queries, and is developing into some sort of replacement for Froogle, the typical user has never ever even heard of or used Base. It remains to be seen how critical Google sees Base as being to its long-term plans, as it has been a year, and many of the original promises of Base have not been fulfilled, or under-utilized.
The one thing we did see used from Base were extra search boxes and options above certain search results, like for “google careers”. Those boxes don’t show up anymore, meaning the promised integration into regular Google search results was actually rolled backwards. Meanwhile, Ask.com and others are rolling out lots of extra features in the same space we thought Base was going to integrate. It seems like Google’s desire to automate everything is holding it back from doing the same things many of its competitors are adding all the time.
The Guardian blog has a hilarious take:
Amazing to think that Google Base is one year old today. As predicted, it proved so powerful and so easy to use that it has already gobbled up most of the world’s information, made Craigslist redundant, and brought Amazon to its knees. It just proves how right Google was to ignore microformats etc. Just imagine how much more amazing it will be when it comes out of beta in another four or five years! Or not, as the case may be….
Garett Rogers discovered a major update to Google Base that was in testing, one designed to make Base able to replace Froogle for the holiday shopping season. It looks like the update is now live, at least for some results types. For example, check out Events and activities or People profiles and you’ll see a new view that shows all the search results on a Google map, so you can see what people or events are near you. It also works for Jobs, Vehicles, Services, and a bunch of other categories. It doesn’t work on Products searches, but if you filter by location, the map comes up, a really smart decision. All very interesting, and really worth checking out.
I won’t be back online till Sunday night, so everybody enjoy this one last giant link post.
Looks like Jason is running his own Google community at Crispynews, which uses a familiar voting-on-stories concept. Check it out.
Google Desktop ships with a warning only really visible when stacking the files in Windows Explorer. Cute.
Click fraud made the cover of BusinessWeek. Lots of exciting stuff in there. Kinda makes me wish I had a subscription.
Meanwhile, Newsweek had Marissa Mayer on the cover and nobody noticed for days. Ouch. Looks like one of those interesting Women in Power pieces. I’d like to see a cover story about old white guys who are changing the world.
Today was “OneWebDay”, when we were all supposed to get all flowery and write about how the internet has enriched our lives. Jeez. When did it become okay to invent days? I like the Talk Like A Pirate Day better.
I love this sign. Google is going to be a religion one day. Or the devil. Could go either way…
Google plans to put a Google Base search box above the results every time you do what appears to be a product search. Say ta-ta, Froogle.
Some online publishers are working together to create a universal way to tell Google and other search engine when to cache and not cache their content. Or, as Andy Beal says, they are looking to spend half a million dollars (more like $583,700) to build a robots.txt file. Their SEO is a genius, to have fooled all these idiots.
Bloglines introduced a nice mobile search option that also reformats linked pages on the fly, like Google Mobile does. The question is, can Bloglines’ RSS reader do this too?
Finally, from Niniane Wang:
What do Stanford students call themselves after they graduate?
Me: They call themselves Googlers.
Google keeps signing up companies to peddle their wares. This time, they’ve gotten Intuit to slap a Google logo on the cover of Quickbooks, and integrate Google products into a number of Quickbooks’ features. Andy Beal reports that users will be able to advertise on Google AdWords or in Google Maps, post products for sale on Google Base (no Checkout?) and get a $50 AdWords credit. Also, Quickbooks will bundle Google Desktop Search.
I hate bundling, and have made that abundantly clear in the past, so I won’t get into another rant. However, Andy does wonder:
One puzzling question, why didn’t Microsoft think of this tactic with their own Money software?
I’m guessing Microsoft is too terrified. I’ve heard that every time someone at Microsoft uses the word “bundle”, a Justice Department lawyer gets his wings.
Google Base now lets you attach documents to your items. You can attach TXT, DOC (Microsoft Word), PDF (Adobe Acrobat), PPT (PowerPoint), XLS (Excel), RTF, ASCII, Unicode, XML and WPD (WordPerfect) files, up to 15 per item for a total of 20 megabytes. In addition, they shipped some usability fixes (finally someone gets that a “Browse” buttom to upload a file is just stupid), a virus checker, Safari support and an upload queue. Hopefully they keep shipping updates to Base, and maybe it’ll be enough to take the service to the next level.
Mark Otuteye writes about ten products that Google should be developing, all based on existing technology that they could leverage to expand their reach (without copying their competitors). There are some good ideas on the list, and others that we know might already be in development, like Google TV Search and free Google Storage. Google Genetics might be a little far-fetched (and likely to start a congressional inquiry), but most of them are solid and realistic.
I was particularly interested to see Scanbuy in the list of technologies Google should take advantage of. Scanbuy is a company that develops software that lets you take a picture of a barcode with your mobile phone, and then look up things like prices, share personal information (via a 2D barcode on the back of a business card) and develop barcode/camera applications using an SDK for all the major mobile platforms (Symbian, Windows Mobile, Palm, Java, others).
I am an investor in Scanbuy, and have, on occasion, an advisor. The reason: I see so much untapped potential in this space, providing a cheaper, more secure and easier system than RFID, and I hope to see someone do some great things with it. Certainly, Google’s the type of company to develop some powerful mobile applications that involve camera phone pattern recognition, and could at the very least use it to improve their current mobile offerings, like Froogle.
It should be noted that Google is folding Froogle into Google Base, a great idea and way to put all sorts of data on equal, usable footings. The more Google services that run on Base, the more Google can then concentrate on improving just Base, which would bring new features to a multitude of services at the same time.
The Google Base Blog has a post clarifying how Google Base entries are being integrated into Google. Products and vehicles will appear in Froogle search results as long as all required attributes are filled in. If a location is specified products can be listed in Froogle Local. And local business listings submitted to Google Base will appear in Google Maps. Listings can also be included in some experimental tests in Google search results, such as displaying recipe information.
Looks like a welcome addition as long as content is filtered and spam-free.
Via the Google Base Blog
The Google Base blog announces that you can now advertise your Google Base items in Google AdWords, finally releasing the “Automat” system we saw in a patent filing last November. From the Base control panel, you can create an ad, and watch as Base automatically finds queries and areas to target your ad against.
While this may not seem like a very useful system for those selling a single item, Google’s low prices make it a good deal. Since the minimum cost per click is just five cents, if you are selling to non-competitive keywords, you can sell your item pretty easily. Think about it: If you are trying to unload a $200 couch, would you pay $2.50 for fifty people to look at it? Of course! This is the smallest section of long-tail advertising, but the one that has least been tapped, advertising by individuals, not companies, for stuff they are trying to unload.
Of course, Google could have even more luck if they gave those using Google Base a Snap.com-like pay-per-action model, where you only paid for the ad when someone bought something. This way, sellers could decide that the sale of their $200 couch was worth as much as, say, seven dollars, and as long as they were bidding more than the other couch sellers for that keyword, that ad would appear in the top spot until the couch was sold. The seller would pay Google seven dollars when the couch sold, but would pay nothing until it was. Considering that clicks for car sales could go for hundreds of dollars in a cost-per-action auction, Google could rake in even more than the fifty bucks it gets for mesothelioma.
Steve Rubel, in two seperate posts, links to new types of searches Google is serving, from auto searches to real estate searches. But is this really news? Last week, job searches were showing up in the regular search results, served via Google Base, and the auto and real estate results are built off the same model.
In fact, the Google Base homepage lists twenty seperate types of specialized search Base offers, including blog search, podcasts, recipes, tickets and content for mobile phones. Some are more featured than others: the recipe search has criteria for main ingredients and ethnic cooking, while clinical trials search has types of diseases, trial stages, and testing centers.
Now, we all made fun of Google Base when it was introduced as being too vague. “Upload anything”, they said, and many were just confused. Now, Google is looking brilliant, as usual. Google is builting specialty search engines, simply by having users and companies do all the work for them, loading the info into the Google dataBase, and building a simple search engine front-end around it. As soon as they have a critical mass of data, they add the search as a OneBox result in regular Google search, and thus, they’ve added a brand new targeted search engine, with a minimum of effort.
John Battelle writes that if you search in Google for “google careers“, you get a OneBox result that lets you see jobs available at Google. Click the link to view it, but it follows this structure (and the example below actually works):
As John points out, there is no result for Yahoo careers, but I did find one for Microsoft careers. Basically, if you have job listings in Google Base, and especially if those jobs are through Career Builder, they will show up in Google search results.
Now, who thinks it isn’t worth putting stuff in Google Base anymore? Not me.
Yesterday, three important things arrived. The first was my first free issue of Wired, with Will Wright as guest editor. The second was a much-needed 300 gigabyte hard drive. You’d think nothing could top that…
So, who’s got a great memory? Anybody remember that I made an order from Google Base a few weeks ago, partly in order to see how the process goes, and partly to see if it netted me a free t-shirt? Do you remember what I ordered?
Well, lets takes a look at this new screenshot/picture gallery I’ve uploaded to Flickr. For the screenshots, click on the image to see a larger version.
Here’s Google Purchases order history page:
Here’s the order summary page:
Once your item has been shipped, you can leave feedback for the seller. Look at how happy I am!
Afterwards, you can review any feedback you previously left.
Now, onto the photos:
Here’s the item, resting on its UPS envelope and order slip:
Notice that they didn’t even bother to send a proper receipt in the package! Just a strip of paper that reads:
Purchased from Google Base - pet pebble. Order #**********
How am I supposed to declare my wonderful $1.15 pet pebble on my taxes (as the business expense it clearly is) without a proper printed receipt? Where’s my certificate of authenticity? How do I know this pebble came from the Googleplex itself, and not some regular rock garden?
Here’s the li’l rock by itself:
And here you can see that it is slightly larger than the Google Base logo:
But hopefully you get an idea of its size by seeing it in my relatively normal palm (as opposed to my other palm, which is anything but normal, or my Palm, which could get scratched):
And finally, my Google Base pebble, in all its glory:
The first podcast is online! I don’t like the levels so much, so will try to work on it for next week. Keep your volume a little low while listening.
Now, show notes:
- 00:00 - Show starts. Song: Stain by Megaphone
- 01:04 - First caller: Devin Reams. Topic: Origami
- 07:19 - Cut to music. Song: Wired For Loud by David Henderson
- 08:30 - Second caller: Jason Schramm. Topic: Calacanis taking over Netscape.com
- 11:20 - Making fun of AOL users
- 11:50 - Reminiscing about Netzero and Juno
- 13:37 - Apple blogs, Apple’s paranoia, Apple Store and employees
- 16:24 - Apple’s R&D budget isn’t increasing like it should
- 17:29 - Google not focusing well, Google Pages, Google Base, Google Desktop, Picasa
- 20:21 - Google buying companies for people, not products
- 20:57 - Google’s history with Blogger
- 23:10 - Google buys Writely and Sketchup
- 24:37 - Why Stu couldn’t call
- 25:18 - What exactly an Origami is, who uses Tablet PCs
- 28:03 - What Writely is like, who should use it
- 31:52 - SketchUp is cool, as are killer robots
- 34:39 - Cut to music. Song:Down With Everything by The Transfer
- 35:41 - Third caller: Miel. Topic: His birthday, nudity on his blog
- 38:10 - Blog readers in strange countries
- 41:03 - Developer Days
- 41:24 - Microsoft’s Atlas (AJAX) toolkit
- 42:45 - Windows Live Local Streetside
- 43:59 - Microsoft’s great interfaces
- 45:39 - Chocolate Xbox 360
- 46:56 - Office 2007 and XML file formats/containers
- 49:54 - Office 2007 UI and the Ribbon
- 51:06 - The Floaty and Clippy (funny stuff)
- 53:42 - The changes from Word 1.0 to Word 2007
- 58:05 - Read Jensenn Harris’ blog! Also, writing XML for the Office 2007 UI, and Office add-in compatibility
- 61:20 - Old DOS games, arcade emulators, emulators on your iPod
- 63:11 - Difference between Origami and UMPC
- 66:07 - Emulating Origami screen sizes on your PC for testing
- 66:52 - Pretending you have an Origami
- 68:10 - Do road warriors need Origami? Plus, Origami vs. SideShow
- 69:30 - Miel needs an MP3 player, got a new phone, my phone stinks
- 73:38 - Wrapping up.
- 74:44 - Ricky Gervais’ podcast (download free shows)
- 76:33 - Closing song: Two Minutes Of Mayhem by CJACKS
Size: 71.9 megabytes
Recorded: 5-6:30 pm on 03.16.2006 in New York
All music is from Podsafeaudio.com
Subscribe to the RSS feed.
Barry has been accepted into Google Base and given the ability to sell items through Google Purchases. He’s posted a lot of screenshots of the process involved, including the part where Google asks for your bank account information so it can make a tiny deposit and verify it, exactly like PayPal does. Google Base can pay you directly into your bank account, which is just peachy, in my opinion.
Scot Wingo has screenshots of a new feature in Google Base: a five-star reputation rating system. After doing business with someone, you can give them a star rating and leave comments for them. The stars are defined as:
- One star - poor. I’d discourage others from this seller.
- Two stars - below average
- Three stars - average
- Four stars - above average
- Five stars - excellent, I’d definitely buy from this seller again.
Also, if you rate at one or two stars, you get a “Remedy” link to try to fix the situation (maybe getting your money back). Reviews can be as long as your want. You can’t edit a review, but the seller can respond, and you can respond to that. For now, only buyers and sellers can see the reviews.
I noticed in my Google Purchases control panel a link where you can see all the comments you’ve made in the past. There is a 4-question FAQ about ratings Google has published.
(via Andy Beal)