Google I/O in Pictures: Google Culture at Work

by Tim O'Reilly | @timoreilly | comments: 9

I had a few miscellaneous notes on Google I/O that I wanted to share, including a few anthropological observations best made with pictures.

• I thought it was really interesting that there were more registration lines for Academia than there were for general admission. Google knows the same truth as Apple, that students are the future. They are making it really easy for students to get on board with their developer platforms. (I believe that there was a much lower student rate for Google I/O admission.) Good on them. (Note to self: we need a really good student rate for O'Reilly conferences too.)
  

• Google was having a bit of fun with their Streetview technology, with a bike riding around the conference creating a Streetview style experience of the event. Don't know if the view is up yet. If so, someone please post a link!
  

• I loved seeing the bike parking area, manned by a volunteer from the San Francisco Bicycle Coalition, who said that they were glad of the chance that Google was giving them to get additional visibility. (Again, note to self: we need to work with sfbike.org to do this for the Web 2.0 Expo, or just handle the bike parking ourselves. It would be great to do the same for Expo NY.)
  

Taken together, these images show a company whose unique culture shows through in small ways. This isn't marketing that you can fake; these small touches are a reflection of who Google is as a company.

As John Cowper Powys said, in one of my favorite books, The Meaning of Culture (1929):

Culture is what is left over after you have forgotten all you have definitely set out to learn...One always feels that a merely educated man holds his philosophical views as if they were so many pennies in his pocket. They are separate from his life. Whereas with a cultured man there is no gap or lacuna between his opinions and his life. Both are dominated by the same organic, inevitable fatality. They are what he is.

Great companies always have this sense of authenticity, while "me too" companies have a culture made up of the latest management fashions. I like to think that O'Reilly has an authentic culture. Our idiosyncracies are our strength, leading us in unexpected directions that is somehow true to something we might not recognize if we were following a map laid down by someone else.

Google Wave: What Might Email Look Like If It Were Invented Today?

by Tim O'Reilly | @timoreilly | comments: 80

Yesterday's Google I/O keynote highlighted the power of HTML 5 to match functionality long experienced in desktop applications. This morning, Google plans to announce an HTML 5-based application - still very much in the early stages of development - that represents a profound advance in the state of the art.

Lars and Jens Rasmussen, the original creators of Google Maps, will take the stage to unveil their latest project, Google Wave. As Lars describes it, "We set out to answer the question: What would email look like if we set out to invent it today?"

That is exactly the right question, and one that every developer should be asking him or herself. The world of computing has changed, profoundly, yet so many of our applications bear the burden of decades of old thinking. We need to challenge our assumptions and re-imagine the tools we take for granted. It's perhaps no accident that this project, carried out secretly at Google's Sydney office over the past two years, had the code name Walkabout. That's the Australian aboriginal tradition of going off for an extended period to retrace the songlines and learn the world anew.

In answering the question, Jens, Lars, and team re-imagined email and instant-messaging in a connected world, a world in which messages no longer need to be sent from one place to another, but could become a conversation in the cloud. Effectively, a message (a wave) is a shared communications space with elements drawn from email, instant messaging, social networking, and even wikis.

It turns out that Jens had the idea back in 2004, when Google first acquired the company that became Google Maps. As Lars tells the story:

We were excited to join Google and help create what would become Google Maps. But we also started thinking about what might come next for us after maps.

As always, Jens came up with the answer: communication. He pointed out that two of the most spectacular successes in digital communication, email and instant messaging, were originally designed in the '60s to imitate analog formats — email mimicked snail mail, and IM mimicked phone calls. Since then, so many different forms of communication had been invented — blogs, wikis, collaborative documents, etc. — and computers and networks had dramatically improved. So Jens proposed a new communications model that presumed all these advances as a starting point....

We started with a set of tough questions:

• Why do we have to live with divides between different types of communication — email versus chat, or conversations versus documents?
• 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?
• What if we tried designing a communications system that took advantage of computers' current abilities, rather than imitating non-electronic forms?

Responding in Context

Let's say I want to communicate with someone. I start a wave, just as I might start an email message. The recipient(s) see an incoming wave, just as they see an email today. Where the magic starts is with replies. In email, you have the choice of including no context, only a portion of the message you're replying to, or the whole thing. In the first case, you need to go back to the original message for context; in the second, you have wasted copies going back and forth. Come into the middle of a long thread and you may be replying to a discussion that has already moved on or covered the point you want to express. But what if there were only one message, shared in the cloud? Now, your comment on the second paragraph is attached directly to that point in the conversation. There are no redundant copies of portions of the message, as replies are seen in context.

As you can see in the screenshot below (click to enlarge), a Wave inbox looks much like an email inbox. But look to the right, and you can see how the replies are embedded right into the middle of the original message, so Stephanie's question about what camera Jens used for his photos appears right in context.

Now, you might ask how well this works for long, complex messages rather than the short one shown in the demo. I don't know the answer, but I suspect that Wave will be even stronger in that case. Our experience with collaborative editing of book manuscripts at O'Reilly suggests that the amount and quality of participation goes up radically when comments can be interleaved at a paragraph level.

Is it a particle or a wave? It's both.

First generation email/IM integration let you see when someone was online, and opt to instant message someone rather than send them an email. Wave simply erases the distinction.

If both people are online at the same time, a wave acts just like an instant message -- except that you see each character as it is typed, just like in subethaedit. "In our experience, a lot of time in IM is spent waiting for the other person to press 'Done'," says Lars. (However, it is possible to set Wave to hold your messages till you are done.)

A key point here is that Google's relentless focus on reducing the latency of online actions is bringing the online experience closer and closer to our real world experience of face-to-face communication. 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.

The real-time connectedness of Wave is truly impressive. Drop photos onto a wave and see the thumbnails appear on the other person's machine before the photos are even finished uploading.

Step by step playback draws a cheer

Let's say you are added to a conversation (a wave) that has been going on for a long time? You can be added at any relevant point, not just the end. But even cooler, you can do a playback of the entire evolution of the conversation.

But wait: there's more! Let's say you want to edit your message (or even a message that was written by another participant in the wave). Yes, you can. The original author is notified, but every participant can see that the message has been modified, and if they want, can replay the changes.

This leads to a change in behavior: conversations become shared documents. The screenshot below shows a simple example, as Gregory and Casey collaborate to produce a good answer to Dan's question. As Stephanie Hannon, the product manager for Googe Wave, said to me, "In Wave, you don't have to make the choice between discussing and collaborating."

As anyone who's used version control knows, a document with lots of discussion and edits can become pretty messy. No problem. You can export an edited wave as a new wave, and start over. "One of our design principles," says Lars, "is that the product of a wave can be as important as the original wave."

Nor do you need to include everyone in every part of a conversation. Essentially, Lars, says, "waves are tree-shaped sets of messages. You can shape a subtree, or a sub-conversation and limit the set of participants in any way you like."

(continue reading)

New Geo For Devs From Google I/O

by Brady Forrest | @brady | comments: 1

Today at Google I/O, Google has made several announcements for geo developers. To sum:

• Google is updating (not abandoning!) its Flash API, but it still prefers the Javascript one
• Google is pushing the Maps API into mobile (and performance is a big part of the push)
• Geolocation is going to be a part of every Google product eventually
• Android is being backed by deep pockets
• Google is preparing an army of Qualified Developers to bring more them more API customers.

More details are below. I chatted with Pamela Fox of the Google Geo team about the announcements.

Flash API with 3D - This upcoming release adds the ability to add a 3D perspective to Google's existing Flash Maps API. It's not a full blown implementation of the Google Earth API PLugin which provides Google Earth level 3D in the browser. Instead this release allows you to tilt the maps and play with perspective. The screenshot above shows how you can tilt the map on its canvas. As Pamela said during the demo "we hope it's useful when you need 3D, but the Google Earth's 3D is overkill." (paraphrased - she said it much better in the session)

Qualified Developer - Google has expanded their Qualified Developer program to include Maps. Very soon the team will be releasing a semi self-serve process for getting the Google seal of approval on your Google Geo API skills. The qualification process will be a test (in the process of being written), referrals, sample code and your participation in the developer forums. Your qualification status will help you stand out in the Google Solutions Marketplace.

Map API v3 - The Maps team also chose Google I/O as the place to launch the new version of their Javascript Maps API. The key improvement is in the performance area and a reduction in initial download size. The other big push was in the mobile arena. Both Chrome and iPhone Safari mobile are supported (Android mobile still has some bugs though). They've also done an overhaul of the Geocoding API (good!) and enabled the default update of the mapping controls on third-party sites.

Client Location on Google Maps - At both Google I/O and Where 2.0 Google showed the ability to locate yourself when on Google Maps. By default it geolocates you by IP, but if you have Gears it will take advantage of Gear's Geolocation API. The feature will show up in the mapping controls as a blue dot. No word on when this feature will be released. I hope this feature eventually becomes available in the Maps API.

Android Developer Challenge V2 - Google has announced V2 of its Android Developer Contest. They'll be awarding prizes across multiple categories including geo-friendly ones like Travel and Productivity. The winners will be chosen mid-November.

Of these announcements I find the Qualified Developer program to be the most significant. By certifying developers Google will be enabling developer to get more Google more API customers. The program started last year with Gadgets Ads. Once they work the kinks out I am sure that this program will extend to every API that Google has and will be quite the moneymaker for the participating devs.

Google Web Elements and Google's Iceberg Strategy (Google I/O)

by Tim O'Reilly | @timoreilly | comments: 5

At Google I/O this morning, DeWitt Clinton announed Google Web Elements, a new simple interface layer to Google Ajax APIs. The goal is to make bringing Google features to other sites as easy as cut and paste. And indeed, the cut and paste functionality is impressive: Add news, custom search, conversations, maps and more to your site with only a few clicks. If the earlier HTML 5 announcements were for developers, these announcements are for everyone else. Any blogger can easily incorporate Google services.

No need to show this via screen shots. I can easily embed live widgets.

Here's a News widget, searching for news on Google I/O: There are two standard sizes, the one shown to the right (350x250), and one in "leaderboard" size (728x90). It would be nice to see user-configurable sizes. Though you can edit the provided HTML, Google does not do automatic size detection.

You can also embed Google docs, spreadsheets, and presentations. (Of course, you can do the same with slideshare and scribd - embedding is the new black.)

The search widget is shown below. It automatically knows to search the site it's placed on. No configuration needed. If you're an Adsense for search customer, you can include advertising. The widget below is live. Type in Google I/O to search radar.oreilly.com for posts relating to Google I/O.

The search widget has been available before, but it's got some new features, as shown in a before-and-after example in the image below. On the left is the previous incarnation. On the right, the new version uses AJAX tabs for improved performance, as well as featuring new kinds of ad elements. And of course, search results can be styled to your site with CSS.

Here's a "conversation element" - you can comment right here on the page rather than in the normal comments. And as a special bonus, I'm told that if you comment in another language, Google's automatic machine translation comes into play.

Elements like these embedded on other pages around the web are the underwater portion of what you might call Google's iceberg strategy: a great deal of their usage is not on their own site, and so not measured by Comscore and others who measure search market share. In his keynote, Vic Gundotra mentioned that Google is now supporting more than 4 BILLION API calls daily across more than 60 different APIs.

A backchannel conversation with one attendee suggests that search API traffic alone might well be larger than the next biggest search destination on the web. Another tidbit from the backchannel: the aforementioned Google language translation API is a sleeper hit, used worldwide for translation of user generated content.

Google Bets Big on HTML 5: News from Google I/O

by Tim O'Reilly | @timoreilly | comments: 65

"Never underestimate the web," says Google VP of Engineering Vic Gundotra in his keynote at Google I/O this morning. He goes on to tell the story of a meeting he remembers when he was VP of Platform Evangelism at Microsoft five years ago. "We believed that web apps would never rival desktop apps. There was this small company called Keyhole, which made this most fantastic geo-visualization software for Windows. This was the kind of software we always used to prove to ourselves that there were things that could never be done on the web." A few months later, Google acquired Keyhole, and shortly thereafter released Google Maps with satellite view.

"We knew then that the web had won," he said. "What was once thought impossible is now commonplace."

Google doesn't want to repeat that mistake, and as a result, he said, "we're betting big on HTML 5."

Vic pointed out that the rate of browser innovation is accelerating, with new browser releases nearly every other month. The slide below, from early in Vic's talk, shows the progress towards the level of UI functionality found in desktop apps through adoption of HTML 5 features in browsers. This looks like one of Clayton Christensen's classic "disruptive innovation vs sustaining innovation" graphs. It's also fascinating to see how mobile browsers are in the forefront of the innovation.

While the entire HTML 5 standard is years or more from adoption, there are many powerful features available in browsers today. In fact, five key next-generation features are already available in the latest (sometimes experimental) browser builds from Firefox, Opera, Safari, and Google Chrome. (Microsoft has announced that it will support HTML 5, and as Vic noted, "We eagerly await evidence of that.") Here's Vic's HTML 5 scorecard:

1. The canvas element provides a straightforward and powerful way to draw arbitrary graphics on a web page using Javascript. Sample applications demoed at the show include a simple drawing area and a simple game. But to see the real power of the Canvas element, take a look at Mozilla's BeSpin. Bespin is an extensible code editor with an interface so rich that it's hard to believe it was written entirely in Javascript and HTML.
2. The video element aims to make it as easy to embed video on a web page as it is to embed images today. No plugins, no mismatched codecs. See for example, this simple video editor running in Safari. And check out the page source for this YouTube demo. (As a special bonus, the video is demonstrating the power of O3D, an open source 3D rendering API for the browser.)

3. The geolocation APIs make location, whether generated via GPS, cell-tower triangulation or wi-fi databases (what Skyhook calls hybrid positioning) available to any HTML 5-compatible browser-based app. At the conference, Google shows off your current location to any Google map, and announces the availability of Google Latitude for the iPhone. (It will be available shortly after Apple releases OS 3.) What's really impressive about Latitude on the phone is that it's a web app, with all the platform independence that implies, not a platform-dependent phone application.

4. AppCache and Database make it easy to build offline apps. The killer demo is one that Vic first showed at Web 2.0 Expo San Francisco a few months ago: offline gmail on an Android phone. But Vic also shows off a simple "stickies" app running in Safari.

   (I love the language that Vic uses: "You can even store the application itself offline and rehydrate it on demand.")

5. Web workers is a mechanism for spinning off background threads to do processing that would otherwise slow the browser to a crawl. For a convincing demo, take a look at a web page calculating primes without web workers. As the demo says, "Click 'Go!' to hose your browser." Then check out the version with web workers. Primes start appearing, with no hit to browser performance. Even more impressive is a demo of video motion tracking, using Javascript in the browser.

During his keynote, Vic was joined on stage by Jay Sullivan, VP of Mobile at Mozilla and Michael Abbot, the SVP in charge of application software and services at Palm. Both showed their own commitment to working with HTML 5. Jay expressed Mozilla's commitment to keeping the web open: "Anything should be hackable; anything should be scriptable. We need to get out of plugin prison." Javascript rendering in Firefox 3.5 is 10x faster than in Firefox 2, with support for video, offline storage, web workers, and geolocation.

Michael showed how Palm's WebOS relies on HTML 5. "You as a developer don't need to leave your prior knowledge at the door to develop for the phone." He demonstrates the power of CSS transformations to provide UI effects; he shows how the calendar app is drawn with Canvas, how bookmarks and history are kept in an HTML 5 database. Michael emphasized the importance of standardization, but also suggested that we need new extensions to HTML 5, for example, to support events from the accelerometer in the phone. Palm has had to run out ahead of the standards in this area.

If you're like me, you had no idea there was so much HTML 5 already in play. When I checked in with my editors at O'Reilly, the general consensus was that HTML 5 isn't going to be ready till 2010. Sitepoint, another leading publisher on web technology, recently sent out a poll to their experts and came to the same conclusion. Yet Google, Mozilla, and Palm gave us all a big whack upside the head this morning. As Shakespeare said, "The hot blood leaps over the cold decree." The technology is here even if the standards committees haven't caught up. Developers are taking notice of these new features, and aren't waiting for formal approval. That's as it should be. As Dave Clark described the philosophy of the IETF with regard to internet standardization, "We reject: kings, presidents, and voting. We believe in: rough consensus and running code."

Support by four major browsers adds up to "rough consensus" in my book. We're seeing running code at Google I/O, and I'd imagine the 4000 developers in attendance will soon be producing a lot more. So I think we're off to the races. As Vic said to me in an interview yesterday morning, "The web has not seen this level of transformation, this level of acceleration, in the past ten years."

Vic ends the HTML 5 portion of his keynote with hints of an announcement tomorrow: "Don't be late for the keynote tomorrow morning."

Additional Resources

Here is a convenient list of the HTML 5 demo apps shown in the keynote this morning. Be sure to look at the page source for each of the applications.

New developer features in Firefox 3.5

To learn more about these HTML 5 features, check out these tutorials from the Opera, Mozilla, Palm, and Google teams (plus a few others):

Canvas: HTML 5 Canvas: The Basics
Painting with HTML 5 Canvas

Video: A Call for Video on the Web
HTML 5 Video Examples

Geolocation: Track User Geolocation with Javascript

Web cache and database: Palm WebOS HTML 5 DataBase Storage
HTML 5 Features in Latest iPhone Applications
Gmail for Mobile: Using AppCache to Launch Offline

Web workers: Using DOM Workers