Kevin Kelleher at GigaOm continues a thread started by Dave Winer in this strange little story about a talking pig, about how Google can take a successful idea started by Amazon with Amazon Web Services, and really do something with it. Kevin's quick description sums up how this is playing out in a way.

As valued Google workers pack up their desks and launch new startups, this is the single best strategy for Google to bring them back into the fold. And it’s a great way to pull the rug out from under Amazon, strategy-wise and profit-wise.

And this is where Google's brain drain might be perceived as not being a drain at all. Successful startup businesses operate without the control of the corporation, but have all the risk and therefore all the opportunity of innovating. If and when they are folded back into the company through acquisition, the reincorporation of the talent and the new techonologies they bring with them provide the bursts of innovation that keep Google's sails full.

It's an excellent analysis that I have to admit I've never thought about, and there doesn't seem to be any confirmation that this is what Google is thinking. But if they are, I would bet my dollars on that strategy.

Gmail updates labels to add coloring.

1) On the left side, you see the label panel. Clicking on the boxes to the left provides a drop down allowing you to provide a color for the label and now allows you to edit the label directly via the panel.

2) On the right side you see the label with coloring in the messages panel.

Google announced their Custom Search Engine Beta. Like the similar Rollyo service, Google CSE allows you to specify the sites you want searched. You can also tell the CSE to search the entire Google index, but give weight to those sites. It also gives you the option of excluding sites. You can also use patterns to specify parts of sites, e.g. to filter out urls including certain words.

There is an option to include the search form and results in your own site. To see that in action check out this job search page on my site.

One of the nicer features is refinement labels. This allows you to tag sites with a descriptive label that can be used to refine results. If, for example, you label sites in your set using facets, e.g. subject, type, etc., then those can be used to narrow the result set.

You can see the CSE I'm playing with for UX-related sites to see how this works (e.g. search for "sparklines"):

Or try this search of some of the newish niche web design/development job sites:

I've made these CSEs open so anyone can contribute to them by adding other UX-related sites to the list. I've also set the CSEs to limit to the sites specified. I tried with the option to search all of Google, but I noticed a little noise because many of the results were outside of the UX scope. It's simple enough, I suppose, to expand the search by clicking the "Web" radio.

I'm not sure how much I'll actually use this. I stopped using Rollyo after playing with it for a few days. But the added refinement functionalities make this service a bit more useful to me.

As you probably know, Google began it's Book Search project a while ago, scanning the collections of prominent academic libraries in the US and England. We're now beginning to see the most practical uses of Google Book Search. Google began offering free PDF downloads of books in the public domain -- books that were published before 1923 or whose copyright has expired. See for example this copy of John Stuart Mill's On Liberty scanned from the Library at Stanford.

I had to read this book in grad school. Students will be able to save quite a few dollars I suppose because of this, but think of the money they'll spend printing these things out. Someone's got to come up with a usable ebook reader. I'm certainly not scrolling through a book like this on my PPC PDA, even in landscape mode. When's the affordable, easy to use E Ink based product going to hit so I can read a novel like Jean Luc Picard in his "Ready Room"?

In any case, this is a great development. If we do see a usable method for toting these ebooks around, I can't see why Google couldn't begin selling ebooks like Amazon does. But maybe that's years away from being a reality. The good thing is the dead tree data is getting digitized in some standardized and accessible fashion.

Amazon's Online Reader is really nice, by the way. Here's a demo using a Beatles book.

I like the meeting format described in this Business Week article on Marissa Mayer, VP of Search Products & User Experience at Google.

New features are digitally projected onto the right side of a conference room wall, big as a movie screen. Everything Mayer and others say is transcribed and projected on the left. Underneath both looms a giant mega-timer. Everyone gets an average deadline of 10 minutes. Mayer and her team add and subtract to the feature as time runs down. It is iteration at lightning speed.

While the formal quick pitch format is unnecessary in small groups, what makes sense is that this format allows for new features to be proposed with frequency high up the chain of command in a large organization with some amount of structure in terms of time restraints. I assume engineers working on the project can pitch directly to Mayer. The critique format also allows a good deal of iteration while exploring ideas so they can be worked on some more and revisited. This excerpt describes the process:

What's most fascinating to me is the projection of the demo on the screen and the immediate capture of the discussion, which I assume goes directly into that internal project management system they talk about. That's excellent. Capturing these trasactions of verbal communications, although using brute force methods of manual transcription, is what knowledge management is about. The post-processing and information retrieval in their system is what glues it all together. That they're openly capturing everything in these meetings is what makes it effective KM work. It's not so hard to imagine all the pieces fit together into a product or at least a process that could be sold as an idea for realistic KM at work:

• WebEx type presentation software
• Transcription (manual now, voice recognition later in the retrieval system?)
• Search mechanism with some simple hooks for metadata (parsing for based on minimal formatting conventions, e.g. "[field name]:")

If enterprise meetings all went this way, mining of the types of tacit information usually floating around in conversations might begin to mean a bit more. Taken too far it could make a lot of information also mean less I suppose, but who cares as long as we have bigger hammers to tackle the signal to noise problem in retrieval. So I'm wondering if that type of process could be adopted as a model and be rolled into a solution. I want to see more under the hood at Google.

Baseline has a feature story exposing bits about How Google works and what we can learn from them. Most of the story focusses on the unique infrastructure Google has been building to support its expanding needs. But most interesting to me is the small bit that takes a lok Inside Google's Enterprise. The article refers to Page and Brin's pronouncement in their IPO that the company is not conventional and doesn't intend to become so. And this appears to be true judging by the way they run the company internally. They won't follow the pattern of what's been done to run businesses in the past if they can find a better way themselves.

This is exactly the attitude that has slowly been building up a revolution inside the ranks at enterprises large and small. Knowledge workers, fed up with the way things are have turned away from conventional software to manage they way the work in favor of better, simpler applications that get out of their way and let them get on with with it.

The article talks about how Google uses a simple system that manages project information using relatively unstructured email as the interface. The system mails employees every week asking what they worked on the week prior and what they plan to work on during the current week. The response is parsed, fed and indexed into a searchable system that is open to the enterprise so that anyone else can track other employees projects that they are interested in. They call it "living out loud".

What they're doing is creating an open system that matches an open knowledge sharing ecology. That openness allows for the "cross pollenation" of ideas. Even better, it provides opportunity for the one thing that is driving nearly every aspect of the innovative web today -- open conversation. They're creating a system that better ensures sustainability because it works with an existing, accepted process -- communicating through email. This removes barriers to use because email is easy. The unstructured nature of the format also means that it can evolve with the needs of the system on the back end. The computer works harder so that the knowledge worker can just dash off a note and get on with their work.

Wow, right? That's revolutionary thinking, and it's so simple on the user-facing end that you hardly have any excuse for not participating. And opportunists that exploit the system by mining and tracking with it will benefit from it immensely. This is the evolving face of knowledge management. The idea of telling the technology to get out of the way so we can do work is what's driving the enterprise blog and wiki revolution. We all need to publish, share and collaborate, but we want to do it as simply and effortlessly as possible. Google embodies that idea completely inside and out.

A few new interesting labs projects were announced at a Google Press Day. What caught my eye:

Google Notebook (to be released at http://google.com/notebook) uses the familiar Gmail UI to give you an easy to use text editor application. Seems to take cues from the wiki-hybrid text app http://www.writeboard.com/. It seems to integrate with Search somehow so you can take notes while sifting through results. Screenshots of the app are available via EricaJoy's photos at flickr.

Google Trends provides simple graphing of search and news reference volume around search terms. The UI is similar to the Finance timeline, without the slider controls. See for example this search for ajax or even more interesting is this search on several different content management systems.

Google Gadgets are little apps that can be floated on your desktop similar to Konfabulator's widgets or the Mac's Dashboard. Gadgets will be most interesting when used for showing stuff like changing data in stock quotes or weather, but I presume will be created for nearly any type of information Google retrieves.

Google's new Finance site is really quite elegant. The site offers information on North American stocks, mutual funds and public and private companies along with charts, news and fundamental financial data. Different things to watch for here are interactive charts, and the blog and discussion group retrieval. Most of the other tear sheet type information, e.g. news, company profile (description), and finances you'll find on all of the other finance sites as well.

The line/spark line chart scrolling is cool. it automatically scrolls to the news for the period you are browsing in the chart. You can also change the range of dates in the chart by resizing the year widget -- mouse over the years at the top of the chart and a little resizing widget appears. When you drag and resize the date range, the main line graph shrinks or expands to show better detail on that range and the news box on the right refreshes to show only the items in that date range. Very nice, clean and simple use of AJAX.

Don Norman recently attempted a simplicity backlash after a few articles touted Google's simple UI as one of the reasons for it's success. Most of these simplicity articles talk about the spareness of its search interface as opposed to Yahoo's, for instance. Finance people are also saying that Google is not presenting a clear enough strategy and that their tools are all over the place. I might agree with that. They have a lot of applications that never seem to make it out of Beta.

Norman says that the simplicty factor breaks down when you try to do anything outside of searching web corpus. His argument is valid. If you view Google as a suite of tools for retrieving information, there is often a disconnection between the bodies of indexed data. The problem is rooted partly in poor information architecture problem and partly in poor interaction design. Norman is saying, I think, that the site doesn't yet allow the integration of the pieces into one UI, and rather segments it by application (and dare I say, by working group within Google?).

But when they do rich applications like Google Maps and this new Finance site, they DO do it rather simply and elegantly. (The Google News Reader on the other hand, ugh! That thing needs to take a lesson from these Beta apps.) With Maps and Finance their focus and execution on the functionality of simple little interaction widgets, e.g. moving a Google map around with a cursor, changing a data set range with a scrolling widget, is what sets them apart. In the end, our discerete interaction with specific tools is what is simple, and it's why I continue to use them over other sites. I don't care if their products are siloed and perhaps require poking around in the labs or clicking tabs to find them. When I get there, there is very little menu cruft in the way and it lets me get the job done quickly and efficiently.

The Google Talk beta is now live. The IM client uses your Gmail account and has built in voice functionalities similar to those offered by Skype and Yahoo! Messenger. I'm looking forward to using this, but I can't see any contacts yet. If you're installing it and want to test it out, you can add me as a contact. My handle is jibbajabba (of course).

Mac users can sign onto Google Talk using a client such as iChat or Adium by setting up a new account. Google Talk uses Jabber. Instructions for configuring a Jabber account on non Google Talk clients are available. This thread TUAW.com goes over specific settings for getting it to work in iChat. Note that the server is talk.google.com, not talk.gmail.com.

Pretty cool, and should certainly send a warning out to Skype. If they refine this product until it's a viable replacement for regular voice calling (e.g. include in and out calling to regular phone numbers), I can see using this more frequently. But, I hate the idea of having to do this all on a PC. The voice functionalities only work on the Google Talk client, and that client like a lot of other Google software including Desktop Search doesn't work on a Mac. I wish the guys at Google would commit a little bit to making software that works on the Mac OS as well.