At the Friday Town Hall meeting for Microsoft employees, Gates shared a few parting sentiments — and, along with CEO Steve Ballmer — shed a few tears. (Microsoft blogger Steven Bink has actual video footage of the Town Hall meeting and the tears — see the clips starting with “Ballmer — Changing the World.”)

There has been so much coverage of Gates and his legacy over the past week-plus on TV and radio, in newspapers, magazines and blogs, that it’s impossible to provide a full list. Here are a few of the many clips. (And yes, I am favoring clips where yours truly and the Microsoft 2.0 book got a shout out.)

The Economist: After Bill

ABC News: Geek Goliath: Gates Says G’Bye to Microsoft

National Public Radio: Gates Retires from Daily Role at Microsoft

Investor’s Business Daily: Curtain Call: Gates Exits Main Microsoft Stage

Gizmodo: Bill Gates Retirement Party (review of Microsoft 2.0)

Wired: The Many (Geeky) Faces of Bill Gates

Reuters: Life After Gates

Reuters: Ballmer becomes lone voice at Microsoft’s helm

Meanwhile, here are the internal e-mail messages that Gates and Ballmer sent to the Microsoft troops on June 27:

From: Bill Gates
Sent: Friday, June 27, 2008 10:40 AM
To: Microsoft - All Employees (QBDG)
Subject: My last full-time day at Microsoft
I want to share some thoughts on my last day as a full-time Microsoft employee.

For the last 33 years, I’ve had the ideal job. It’s been incredibly exciting to come here every day to work with the smartest people in the world to develop breakthrough software. Together, we have built a great company that has profoundly changed the world for the better.

 After today, I will be shifting my full-time focus to the work of the Bill and Melinda Gates Foundation while keeping a strong connection to Microsoft.

The fact that I am making a career change does not mean that our work at Microsoft is done. In fact, the most exciting impact of our software is still ahead of us. Everything we have done up to now is just the foundation for the more dramatic breakthroughs to come. As you apply the magic of software to delivering a new generation of innovations, this company will continue to transform the way people communicate, create, and share experiences.

Microsoft is in an incredible position because we have momentum and a great pipeline of products and technologies. Even more important, we have great people at every level. In research and development we have great engineers focused on solving the most pressing challenges in computer science and turning new ideas into innovative products. In marketing, sales, and customer service, our world-class organizations keep getting better.

We also have strong leadership. As Microsoft has grown, one of the most exciting and fulfilling things for me has been to watch new leaders develop.

It starts at the very top. For the last 28 years, I have loved working side by side with Steve. Even now after all these years I am regularly impressed with his energy and insight. I think he and I have enjoyed one of the great business partnerships of all time. Steve has done a great job leading the company since he became CEO in 2000. Steve’s passion for democratizing the power of technology and inspiring customers, partners, and employees will keep us driving ahead.

I am thrilled to have Ray and Craig playing key roles in guiding the company’s strategy. For over a decade I had hoped that we could convince Ray to join Microsoft—and in the three years he has been here, he has made a huge difference in helping us focus on the challenge and opportunity of software plus services. I have worked with Craig for more than 15 years. His ability to anticipate the future direction of technology is a key asset, as is his deep interest in and understanding of emerging markets.

Of course, I’ll continue to be involved in the work of the company as part-time Chairman. As part of this I will help with a handful of projects that Steve, Ray, and Craig select.

As I reflect on the last three decades, the thing I’m proudest of is the role that this company has played in making the power of digital technology accessible and affordable. Software running on personal computers and other devices is the best tool for empowerment in human history. Microsoft founded the personal computer software business and we built the platforms that enabled the software industry to develop. Without your contributions, we would not have succeeded in making our dream of a computer on every desk and in every home a reality for more than 1 billion people worldwide—a dream we will extend to everyone in the future.

As I make the transition to focus more of my time and energy on the Gates Foundation, I am looking forward to applying the lessons I’ve learned—and, in some cases, the technologies that we have developed—to help address some of the critical issues that people around the world face in education, economic development, and health.

I want to thank all of you for your hard work and your dedication. It has been a privilege and an inspiration to come to Microsoft every day. I look forward to the amazing, world-changing innovations you will deliver in the years ahead as you continue the great work this company has always done.
Bill

From: Steve Ballmer
Sent: Friday, June 27, 2008 11:40 AM
To: Microsoft - All Employees (QBDG)
Subject: Bill’s Transition
I just wanted to add a few thoughts to Bill’s mail.

For the last 28 years, it has been my pleasure and privilege to work side-by-side every day with Bill to help build this great company. Of course it would be impossible to overstate Bill’s contribution as a technology visionary and business leader. Thanks to his vision and insight, Microsoft has delivered incredible innovations that enable people to achieve things every day that once seemed impossible.

For so many of us Bill has been a mentor, a colleague, and an inspiration. He has challenged us to do our very best work, and it has been an honor to try to live up to his expectations. For my part, I’ve had the incredible good fortune to have Bill as a great friend and a wonderful business partner. I’m grateful for the opportunity he gave me when he convinced me to drop out of business school to join the company. I can’t imagine a better way to have spent the last 28 years.

As much as I’ll miss Bill’s day to day presence here, I’m excited and deeply inspired by the step he is taking. The impact he will have on world health and education as he shifts his focus to the Bill & Melinda Gates Foundation will be amazing. Bill’s passion for empowering people and his commitment to making the world a better place have always been among his most important defining traits—this transition is a logical and important next step forward for him.

At the same time, we will continue to do amazing things here at Microsoft. Working together, we have created an incredible culture of innovation and accomplishment that will provide the foundation for future breakthroughs and even greater success.

During the last three decades, building on Bill’s insights about software and computing, we’ve transformed the way businesses operate and revolutionized the way people communicate, share information, and access entertainment. We helped create an industry that provides jobs for millions of people around the globe. We have delivered tools that have changed hundreds of millions of people’s lives for the better.

But for all we’ve achieved, I believe we’re just getting started. We’re in the midst of one of the most exciting periods in the history of this industry. Computing continues to become more powerful, more portable, and more affordable. Content, communications, and media are shifting entirely to digital formats. The combination of software plus services is transforming the way we create and deliver computing experiences. Online social networks are changing how people interact with each other. Gestures, voice, and other natural user interface capabilities are changing the way people interact with computers. New tools for developers are making it easier to take advantage of multicore processors and to deliver rich, connected experiences.

These trends are creating incredible new opportunities in our industry. And no company is better positioned to take advantage of these opportunities than Microsoft. No company has the tradition of innovation that we do, or the expertise across such a wide range of technologies. No company can match the breadth of our talent or the depth of our leadership.

Today, I believe that we’re poised to deliver a new generation of innovations that will have an even greater impact on people’s lives. That’s what this company has always really been about—finding new ways to use technology to make the world a better place. We’ve done that for a billion people during the last 30 years, and we’ll do it for billions more during the next 30 years.

This is not to say that we don’t face difficult challenges. But in the past, we have always done our best work when our job was to tackle the most pressing challenges. I’m absolutely confident this will be true as we move forward.

There’s no doubt that Bill’s last day as a fulltime Microsoft employee marks an important milestone in the company’s history. But the truth is that not much will change. Bill will continue to play a key role as Chairman of the company. And we’ll all continue to work together to deliver innovative products and services that improve people’s lives and create new opportunities for Microsoft, our customers, and our partners.

I want to close by extending a heartfelt thank you to Bill for his friendship, his partnership, his insight, and his inspiration.

Bill — It has been an amazing and wonderful 28 years for me, and I know that everyone at this company shares my respect and admiration for what you have achieved. We look forward to seeing what you do in the next phase of your career as we continue to build the great company that you launched more than three decades ago.

Steve

Microsoft futures of interest to institutional investors: The Citi Software folks graciously hosted me on a dial-in call for investor clients. The full Q&A is available by dialing in:

Replay: 888-203-1112 or 719-457-0820
Passcode: 4192523

Fox Business on the Microsoft vs. Apple tech rivalry: A video clip of my recent appearance on Fox Business, where I was charged with “defending” Microsoft against two Apple-friendly tech folks (Leander Kahney of Wired.com and Arik Hesseldahl, tech writer for Businessweek).

Windows 7 secrecy: My ZDNet blogging colleagues Larry Dignan, Ed Bott and I chat in this video clip about Microsoft’s decision to clamp down on all information pertaining to Windows 7 — and what that strategy means to customers and partners.

As I note in Microsoft 2.0, R&D spending doesn’t necessarily translate to more/better innovations. Plus, many of the “innovations” to which the Softies point aren’t seen by the rest of the industry as anything to write home about. And Microsoft Chairman Bill Gates has thrown his weight behind more than one concept (voice/vision-centric input; SPOT watches; Tablet PCs, Surface multi-touch tabletops) that haven’t panned out so well.

Privately, some Softies acknowledge that Microsoft needs to find new and innovative ways to innovate. In my book, I touched on some of the incubators, greenhouses and other new business ventures, products and initiatives Microsoft is testing as possible new innovation channels. A few of these had yet to go public by the time I submitted my book manuscript. But now there is more public info on some of them, specifically:

Microsoft’s Live Experimentation platform (ExP). The ExP team describes the platform as something that “enables product groups at Microsoft and later on will enable developers using Windows Live to innovate using controlled experiments with live users.”

Officelabs. A year ago, I was hearing talk about a new incubator in the Microsoft Business Division that was trying to become more agile and open. It took Microsoft until late April 2008 to acknowledge publicly the existence of “Office Labs.” One of the first Office Labs projects to see the official light of day is ”Search Commands,” a tool that was codenamed “Scout” — an add-in for navigating more easily Office 2007’s new Ribbon interface.

There are still other as-yet-unannounced innovation projects at Microsoft. Stay tuned for more….

Disbelief was followed by utter despair — and not just on Yahoo CEO Jerry Yang’s part. All I could think on February 1 was I was going to have to go back and revise every single one of my 300-plus pages. 

I did go back in and update my chapters to reflect the possibility Microsoft might end up buying Yahoo. Then I revised again to say Microsoft did buy Yahoo (given that much of the press in February made it sound like it was pretty much a done deal). Right before my drop-dead go-to-printer date, I revised one last time, saying that Microsoft might or might not buy Yahoo.

Well, as we now know, on May 3, Microsoft withdrew its takeover bid, after being unwilling to meet the higher per-share price that the Yahoo board was demanding.

As I noted in the book, if Microsoft had bought Yahoo, it would have taken the companies years to integrate. While Microsoft officials were predicting an almost immediate impact on their shared online services/online advertising strategies, few outside observers believed that the buy would result in any immediate changes — in Microsoft’s Online Services Business or any other parts of the Redmond software maker.

I think Silicon Alley Insider Henry Blodget had the best analysis of why a Microsoft-Yahoo combination would take forever (if ever) to begin to gel. Based on comments from many Softies, Yahoos and industry/market watchers, Microsoft’s ultimate failure to buy Yahoo may have been the best thing that could have happened to MIcrosoft, for a variety of reasons. The dissolution of the deal does beg the question, of course, of what Microsoft now plans to do to build its online ad inventory (and search market share) — the primary reason Ballmer & Co. said they wanted the deal in the first place.

Will Microsoft swoop back in later this year and try to buy Yahoo again? Will the Redmondians buy another online-advertising player instead? Will Microsoft do the seemingly unthinkable and completely withdraw from the online advertising business? Stay tuned….

In the summer of 2007, that gang of three issued a Windows Live Wave 3 planning document that
demonstrated just how much they planned to change the modus operandi of the Windows Live team. The thinking: Theme planning, milestones, vision checkpoints, and other Windows-like conventions, if successfully implemented, will make Windows Live services more predictable and reliable. (The addition of these more rigorous quality gates also risk slowing the Windows Live development pace, however.) Meanwhile, it will be interesting to see if and when Microsoft ends up acquiring Yahoo — or even if its acquisition bid ends up a distraction more than a reality — how these milestones and policies are impacted.

The internal Live Wave 3 Planning Memo Part 1 is here.  What follows is Part 2.

Delivering High Quality Connections

What are the new creative ways to drive connections to our services, other services, or merchants? Most people agree that the ideal is experiences that are relevant to the customer and drive theright traffic to the referee.  The specific things we’d consider include:

* What features will we design that drive more traffic across the network?

o To MSN
o To Search instant answers, etc.
o To advertisers or other merchants

* What features create new opportunities for merchants or delivery of value added services?

o Photo printing, events, affinity groups, lists

* For display advertising, how do we improve and target?  How do we manage the experience for customers (prevent irrelevant ads)? 

o How do we avoid “selling all the inventory at any price?”  This is the channel 35 late night TV example
o How do we create an experience that allows for the systematic collection of customer data about their location, preferences and interests that can be fed to the ad platform for better targeting which results in better quality of ads and increased yield.  This extends to having them add more information about their contacts and driving up the attributes per contact metric is currently low implying good breadth but not enough depth.

* How do we track and measure effectiveness?  How do we improve based on what we know?
* How does education and x-sell work in tandem? What is our SEO strategy for the suite, and itscomponents?
* How do our products (and specifically our user-generated content) effectively show up in our search results without violating user trust?
* Cross team – AdCenter

First Class Mobile Experience

* How do we create a compelling end to end mobile experience for users whose mobile device is their primary access point to the internet? (Particularly pertinent to certain markets such as China and Japan).
* How can we use mobile features to drive engagement with our services in the PC? And vice versa?
* How do we balance the need to reach every mobile user with the depth of experience with more capable phones?
* Phone number as a Live ID, phone number for SMS reset
* SMS, MMS, VOIP interop
* Contact roaming
* Taxonomy of mobile devices
* Onboarding and promotion
* Relationship with mobile client software team
* Messenger to mobile interop
* Relationship with network operators
* Can /should we do anything for location based services?
* What is our monetization strategy and role of our partners and how does that differ by market specially where online advertising is not as well developed but mobile use is increasing POP, IMAP, and other standards for connecting to non-Windows Mobile devices
* Cross team – Windows Mobile, Mobile Platform

Keeping in Touch with People

* What’s the list of contacts for any person?  How do contacts, buddies, friends, history of who I’ve communicated to all relate?
* What’s the taxonomy of ID to contact card to “extended contact card” to browse page to website?
* “Browse page” and “personal website” separation.  How do we separate the “browse page” (my view of your stuff) from your authored experience (personal website)?
* What is our model for groups?  What is a group, how does it relate to a distribution list (DL) and security group?  What is our taxonomy for personal groups (people in my contact list), shared groups (groups with membership visible to the members), and public groups?

o This includes roles (owner, contributor, etc), definition (does a group always have a space?  Does it have presence?), and platform (where are they stored, how are they shared)?

* What are the “1st party” experiences on our platform?  Who is the target audience and how do we measure success?

o Are we a blogging site, and how do we position relative to the competition?  Same with photo sharing or social networking.

* What is our approach to family scenarios?  How do I “administer” my family, share with family?  Is it a group?
* How do we enable basic sharing across home/work?
* How do we make it easy to solicit advice from your contacts, groups, whole WL network?

o How do we leverage all the ad hoc communication in our network to populate suggestions to users?
o Lists as first class things to share – should we invest?
o How can we best leverage content in ad hoc communication from IM and Mail to generate suggestions from your friends, etc?

* How can we expand/enhance the gleam concept to better surface user updates to others? How do we make sure users understand the gleam’s meaning and usefulness? How can we provide more granular detail via the gleam (or some evolution of it) with the least number of clicks/work as possible?
* What is the role of location based services in the context of social networking and keeping in touch with people?
* What is the plan to elevate contacts and presence across the network?

o How do we “aggregate” views of the people on other social networks into WL?
o What is our plan to extend beyond Live IDs to phone numbers and email addresses on a person’s contact list?
o What personalization do we provide to the communication recipient to decide what method they would like to receive (e.g. I prefer IM, Mail, SMS)?
o Do we provide a people directory on the network? How do we facilitate making new connections without becoming intrusive?

* Do we enable affinity networks (AARP, Boy Scouts, etc)?  If so, how?
* What’s the developer model for extending this experience?
* Cross team – ABCH, Messenger back end, XBox Live
* Deeping Our Relationship With Customers
* Lowering the barrier to entry
* One view of profile
* Encouraging use of more services
* Shifting from secondary to primary usage.  In each service, we have customers who have different levels of engagement.  How can we know, for each customer, whether they are primary or secondary on any service?  What features would we add to encourage deeper engagement? 
* Monitoring abuse and rewarding loyalty.  Another opportunity is preventing abuse on our system. Today people can sign up for free email accounts and then use these to send SPAM.  They can sign up for 30 SkyDrive’s and abuse our storage terms.  How do we know if a customer is abusing the system? Are there places we require deeper engagement or knowledge to use more of our services?  For example, should we offer 5GB of free storage but after 1GB is used require a phone number or other validation?  How do we balance this abuse monitoring with our desire to increase sign ups?
* Attracting customers to our services from our client experiences.  Our Windows-based suite of applications provides an opportunity for us to promote and encourage usage of our services.  For email, is there a “one click” way to migrate client settings to the service and roam them?  How does the experience within WL Mail get visibly better if a customer uses a WL Hotmail account?  In photos, do we have “one photo collection” which is stored with your LiveID that includes settings (like user name/password) from all sites you’ve published to?  In Live Writer,  do we make it easy to “keep track” of your blogs/comments on Live once you’ve used writer to publish to a blog?  We can attract customers with our best of breed support for standards and then encourage usage of our services.  In this theme we should consider ways to instrument and measure our success.
* What is our taxonomy and shared investment for “knowing more” about customers?  For each service, we should know if someone has not used, is a primary customer, or a secondary customer.  How do we improve this?  What’s the platform?

o Moving from secondary to primary
o Moving from visitor to author
o Pre-provisioning services
o Aggregating information

* How do we lower the barrier to entry?  What is the evolution of our experience for onboarding, sign in, sign up?
* How can we use Windows Messenger to encourage more adoption of our services – through notifications, recommendations, or other features?
* What’s the invitation model to join Windows Live?  (lowering friction)
* How do we help customers manage their personal information?
* How do I manage my “online persona”?  This should include home/work, home/gamer, and multiple IDs?  How does this relate to sharing, groups, and presence?

o Messenger – do I have one logon and presence for each ID?
o Spaces – do I have multiple spaces, one for each persona, with different permissions?  Or do I have one space with different views?
o Home/work – what if my work life is run outside of Windows Live?  How do we “link up” in the user experience?
o What are the implications for linked IDs, profiles?

* How do we know which devices and services are associated with an account?
* How do we enable migration from other services?  What is the flow and which scenarios are important?  How do we integrate compelling features that “pull” them into our service and entice them to migrate?
* What’s the relationship between your account and the rest of your Microsoft relationships (billing, newsletter communications, Xbox, etc)?
* Cross team — IDS, billing, Office Live small biz, advertising/adcenter
* Wiki Everywhere
* Community as editors
* Everything “editable” by everyone
* Defaults, implications, assume data
* Getting to Your Information Anywhere
* For each type of information (mail, calendar, files, photos, folders):

o What steps will we take to deliver for files what we have done with mail?  One view of all my information, on all my devices, from any device?
o How will we help customers aggregate their information across devices and access it on any device?  (e.g. POP aggregation, storing LiveWriter blogs, etc)

* For files/folders:

o How does our platform evolve to move beyond files to photos?
o How does the Windows Live experience improve when you add Office Live workspaces?  How is our combined offer superior to competitive offers (like Google docs/spreadsheets)?
o What is the namespace (both from the cloud and on a device) for files, folders?

* For social networks:

o Aggregating social networks – what’s our model?  How do we “aggregate” from other social networks and learn more about customers?  What do we enable customers to “project” into another social network and what do we allow them to “track” from ours?  Take Facebook, QQ, MySpace as examples

o Specific focus on Messenger as a “presence” platform – how do we make Messenger “the essential horizontal IM and presence platform for all social networks, with best experience on Windows Live”?

* Lists as first class things to share – should we invest?

o How do we enable community-generated content through rating and voting (to create aggregated lists like listible.com or 43Things.com)

* How do we bridge home and work environments?

o exchange interop
o running domains for small business, EDU, etc

* How do I manage my devices – which data where?
* What quotas do we have and how they are managed?
* What is the 3rd party opportunity:

o For connecting to information on Windows Live (publish photos onto a different social networking site)
o For putting information into Windows Live (adding photos to the aggregate customer view of “their stuff”)

* Cross team – WLC, Office Live, LPS Storage
* Securing the PC and Web Experience
* How do we take a leadership position in SPAM?

o How much of the SPAM problem is about giving customers easy management of gray mail? (inbox cleanup tool that helps unsubscribe to unwanted newsletters with a few clicks)
o How do we move to being competitive in terms email deliverability by offering tools, information and options to legitimate large and small senders (so they stop telling customers to not use Hotmail –some of them have done that in the past)

* What’s the next level of investment in phishing?
* How do we move beyond SPAM to reputation-based services?  Malware, IP reputation
* What is the experience when customers “update” to OneCare?
* How does our work improve (and naturally extend) the Windows and IE experiences?
* Should we increase investment in the administrative role?  How can we improve the experience for the “family administrator” or “custom domain administrator”? 
* What is our platform offering for Microsoft?  For the industry?  How do we enable solutions in the enterprise?
* Cross team – OneCare, ForeFront
Always Running Data Center
* Quality of service, benchmarked with top competitors
* Beta environment
* QOS reviews and focus
* Back end/front end collaboration
* Bridging Home and Work
* Calendar sharing
* Email
* Small biz and enterprise
* Enlisting the Army
* Partners
* Developers
* Community contributors

THEME OWNERS

Each theme will have a leader (generally a GPM and/or PUM), a product planner, and, in most cases, a design lead.  The leaders will pull together a virtual team responsible for refining and scoping the themes, and will produce a presentation and a high-fidelity click through prototype for each theme. 

The leader will coordinate the investigation of each theme, working with product planning, product management, design, development, testing, and other discipline leaders.  They will work across dependent teams as they are writing their drafts and make sure that scenarios or features that span teams are covered end to end.  They will outline the proposed scenarios and customer promise.  We expect to hold these theme checkpoint meetings in late October with the these leaders and members of their virtual teams…..

(Note: I removed the list of project owners’ names here.)

Schedule

Below is the detailed planning schedule, including owners, next steps, and deliverables

8/31/2007: Planning Memo/ Outlines themes and bets/ Chris Jones, David Treadwell

8/31/2007: Engineering Memo/ Outlines engineering framework for Windows Live Experience teams/    Steve Liffick, Arthur DeHaan

September/October: Theme Planning/ Teams take planning memo, plan their work/ PM leaders, productplanning, product management/ Ship Wave 2

September/ October: MQ Planning/ Prioritized list of MQ deliverables, MQ entry/exit criteria Dev and test managers with GPMs/ Begin MQ feature list and prioritization

October: Theme Checkpoint/ Includes prioritized list of scenarios, PPT, and click-through prototype. Scoped planning themes inform vision document, direction for feature teamsPM leaders, productplanning, product management

November: MQ  & feature team planning begins/ MQ work begins/ Dev and test managers with GPMs  Focus on MQ

November: Team Planning/ Teams take refined themes, continue to plan their work/ PM leaders/    Planning within feature teams, cross team dependencies included, early spec writing begins

Early December: Vision Checkpoint/ Teams review plans to support scoped themes, outline proposed feature team investments/ GPMs and/or PUMs

December: Vision Document/ Outlines pillars for release, describes team commitment/ Chris Jones, David Treadwell, Brian Arbogast

December/January: Team Planning/ Teams refine plans based on vision document, including M1 commitments/ GPMs and/or PUMs/ Planning within feature teams, spec writing for M1 continues

Q1 2008: M1 Entry/ MQ complete, enter M1/ All feature teams/ Begin M1 coding

In the summer of 2007, that gang of three issued a Windows Live Wave 3 planning document that
demonstrated just how much they planned to change the modus operandi of the Windows Live team. The thinking: Theme planning, milestones, vision checkpoints, and other Windows-like conventions, if successfully implemented, will make Windows Live services more predictable and reliable. (The addition of these more rigorous quality gates also risk slowing the Windows Live development pace, however.) Meanwhile, it will be interesting to see if and when Microsoft ends up acquiring Yahoo — or even if its acquisition bid ends up a distraction more than a reality — how these milestones and policies are impacted.

Here is Part 1 of Microsoft’s internal Windows Live Wave 3 Planning Memo.

TO: Windows Live Experience Team; Live Platform Services Team
FROM: Chris Jones, David Treadwell, Brian Arbogast
RE: Planning Windows Live Wave 3

INTRODUCTION

As we are nearing the completion of Windows Live Wave 2, we want to congratulate the teams on their work to date.  Windows Live Wave 2 delivers the first version of our integrated suite of software services.

Our mission is to deliver the essential suite of software and services for individuals around the world, designed to help them stay connected (browse, create, manage, and share with the people they choose, on any device) and protected (provide safety and security for their information, their families, and their devices), built on the leading platform for developers, merchants, and advertisers. We believe that users of these services will create a web of user defined content that will improve the traffic to Windows Live Search, create a valuable audience for advertisers, and enable the next generation of software and content publishers.  We believe our investments in safety, security and PC health will differentiate our experience and dramatically increase customer satisfaction and loyalty.  At the same time, we believe this combination of software and services will be an incredible benefit and differentiator for the Windows PC and Windows powered devices.  Finally, we believe that it is the work of the community of developers and content publishers that will enhance and uniquely differentiate the experience for customers.

This document is the planning memo and it outlines the shared assumptions in our planning work across Windows Live Experience and related teams in Live Platform Services and Mobile Services:

* It covers the state of the market – including customers, partners, competition, and challenges. 
* It outlines our strategy for Windows Live. It sets the shared themes and bets that span all the features in Windows Live Wave 3 and investments for Wave 4. 
* It outlines the feature teams and areas for investment in Windows Live, as well as the questions to be asked and answered as part of team planning. 
* It sets the timeline and schedule for both our plan and Wave 2. 

Together we will build a plan that spans the work in Windows Live Experience and Live Platform Services and supports the business priorities established by the Online Services Group.  This plan will cover the work and commitments for the teams over the next year.  A wave describes the deliverables that happen throughout the year, including smaller improvements that are made to our services on an ongoing basis.  As we plan, it is important to recognize that the platform group needs more notice and time to build out both back end infrastructure and operations.  So when we plan Wave 3 we need to also look out enough to plan the platform work for Wave 4. We are making a few changes from Wave 2 based on our experience as a team.

* We are going to focus our planning work on themes that span feature teams.  We hope in doing so to improve the experience for customers when using our suite together.
* We are going to enter planning as a team with a clear definition of planning themes, including click-through prototypes.  As we do this we will front load design of major cross Wave dependencies.
* We will have clear entry and exit criteria for MQ, and use MQ to invest in design and implementation of major cross team dependencies.

The following figure illustrates our planning process.

This planning memo starts our planning process and is intended to be used by teams to develop the vision and the feature set for their areas.  It describes bets and themes for Wave 3.  These themes in many cases span experience and platform, client and service, and are used during planning to structure our investigation of what to build.  In some cases, this means we will decide not to pursue a theme.  In other cases, we will merge themes together.  In others, we’ll narrow our definition.  And in some cases, a new set of scenarios will emerge.  As we investigate the themes, and scope what is possible, we will write the vision document, which defines the pillars for our release.  As with Wave 2, the vision document describes our team commitment to deliver.

Some people have asked “what’s the difference between a theme and a pillar?”  Themes are used in planning, and outline the areas of exploration for a Wave.  By design, we have more themes than we can fit in a Wave.  In the course of planning, we will refine the themes into work that can be achieved in our release timeframe.  This exploration, based on detailed plans from the feature teams, results in the pillars for the release.  It is possible for a theme to become a pillar.  It is also possible that in planning we merge themes or come up with a new set of scenarios that become a pillar.

In Wave 2, themes were provides as guidance to feature teams.  In Wave 3, themes will be the foundation of our planning process.  Each theme will have an owner (generally a GPM and/or PUM), a product planner, and, in most cases, a design lead.  Each theme owner will produce a presentation and a high-fidelity click through prototype for each theme.  The role of the owner will be to coordinate the investigation of each theme, working with product planning, product management, design, development, testing, and other discipline leaders.  They will work across dependent teams as they are writing their drafts and make sure that scenarios or features that span teams are covered end to end.  They will outline the proposed scenarios and customer promise.  We expect to hold these theme checkpoint meetings in late October with the theme owners. 

These theme checkpoints provide scoped and refined themes to the feature teams, who will then work on planning their work for Wave 3.  Once we have scoped the themes, we will have a set of feature team checkpoints, where the teams will describe what they believe can be delivered for Wave 3 based on our themes.  These feature teams are the experts on the scenarios and specifics for their area and are responsible for building best of breed solutions to meet customer demand.  Any conflicts or disagreements between teams should be resolved as part of the checkpoint meetings. We will hold checkpoint meetings with each feature team in December, where the feature teams will outline their plans for Wave 3. 

Following these checkpoint meetings, we will decide on the pillars for Wave 3 and write the vision document.  The GPMs and/or PUMs will work with Chris Jones and David Treadwell to create a single draft vision document that spans the work in Windows Live Wave 2.  This will include the value proposition, tenets, top level schedule, shared bets, and feature commitments across our teams.  We will load balance as required across teams to make sure that the themes and scenarios are delivered for the Wave.  We expect to publish the vision document in December.

The feature teams will use the vision document and resource plan to build the final feature list and schedule for their area.  Following their detailed schedules, we will have a vision week with team members and partners where we walk through the vision, demonstrate the prototypes, and commit to the shared schedule.  For Windows Live Experience teams, we will then move into M1 and coding for Windows Live Wave 3….

BIG BETS

There are a few bets we will make as a team that we know now will require work from all of us to achieve the vision.  These bets represent initiatives across the team.  In some cases, they will be realized by work that spans feature teams, and in others they will be largely owned by a single feature team.  Each feature team must understand the work required to support each bet as part of their Wave 3 planning.  As with Wave 2, we have picked a targeted number of bets for Wave 3.Windows and Internet Explorer.

While Windows Live is a service that is available across devices, we know most customers connect to their services on a Windows PC using Internet Explorer.  We have a unique opportunity to provide a seamless experience for customers who choose to use our services with Windows and Internet Explorer.

While we will target a seamless experience on Windows Vista, we will make a bet on the Windows 7 platform and experience, and create the best experience when connected with Windows 7.  We will work with the Windows 7 team and be a first and best developer of solutions on the Windows 7 platform. 

Our experiences will be designed so when they are connected to Windows 7 they seamlessly extend the Windows experience, and we will work to follow the Windows 7 style guidelines for applications.  We will work with the Internet Explorer 8 team to make sure we deliver an experience that seamlessly extends the browser with our toolbar and other offerings.Search and MSN

Our network of services combined completes the experience for our customers, advertisers, and partners.  We will be on MSN as the unified portal and customized home page for Microsoft’s services.  We will bet on Live Search to connect customers to information when in our experiences. 

We will optimize our experience for customers who use MSN and Live Search, and create unique experiences that work together across Microsoft’s network of services.
 
Beta and Service Deployment

We will invest across our suite in improvements to beta and service deployment, with a particular focus on web-based services.  We will make it possible to “self-host” and “dogfood” all services ontop of “live” data, so it easy to test and use the products before they are deployed to production. 

We will invest to increase the stability and maintenance of our INT environment.  We will make it possible for customers to “opt in” to beta versions of our services so we can introduce betas and get customer feedback without updating the entire customer base.

PLANNING THEMES

Everyone should think about planning themes as the rough draft for the vision document.  They form the areas for exploration and set of questions we will ask and answer as a team as part of our planning process.  By design, we have more themes than we can fit in a Wave.  These themes in many cases span experience and platform, client and service, and are used during planning to structure our investigation of what to build.  In the course of planning, we will refine the themes into work that can be achieved in our release timeframe.  As we investigate the themes, and scope what is possible, we will write the vision document, which defines the pillars for our release.  As with Wave 2, the vision document describes our team commitment to deliver.  In Wave 2, themes were provide(d) as guidance to feature teams.  In Wave 3, themes will be the foundation of our planning process. 
 
Refining the Web Experience

As part of our vision to help customers get to their information from anywhere, it is essential for us to have an integrated, browser-based way to deliver our services. With Wave 2, we started our integration with a shared home page and header for navigation.  This in turn uncovered a new set of seams in the experience.  We’ll add more services in Wave 3 – for example device management, RSS feeds – that will need to fit in to our experience.  We need to rethink the overall navigation model between the current header and secondary navigation in spaces. 

What is the difference between the “what’s new” view on spaces and home.live.com?  Should private messaging and the inbox come together as a single concept?  Are events part of the calendar or separate?  How do we bring together the concept of home/start and dashboard, or should we?  Do we have a navigation model pivoted on people (my stuff, friends stuff) or data (photos, files, etc)? Or some hybrid? 

Our customers will use both MSN and Live Search, in addition to Windows Live, and we want to support navigation across the network.  Today our header is optimized for “my view of my stuff” – showing customers their services – and is scoped to a few elements.  We have an opportunity to connect our networks together and optimize the depth scenario.  What should my view of my stuff be?  How does the experience change when viewing someone else’s space?  How do customers move from MSN to Live Search to Windows Live?  As we move to MSN as the “home page,” what is the evolution of home.live.com, how does cross navigation work? How does Office Live extend the experience? 

Some customers will run on a Windows PC and have our client software and use it together with browser-based services.  In Wave 2 we delivered limited integration between our client and browser experience.  How should we change our web-based experience when we know customers are using our client software?  How does our experience change if the toolbar and /or Messenger are installed? What if the customer is on their primary PC?  How do we use our web-based suite to encourage use of our client software?

Many customers will use Internet Explorer to connect to our services.  We have a unique opportunity to extend the browsing experience for customers using Internet Explorer with our toolbar and additional services that enhance the browsing experience.  What scenarios and features will we enable that are unique for customers using Internet Explorer?  How should our web-based experience change if the toolbar is installed?  How do our services make the browser experience better?

Certain web-sites today have a flair or particular shared set of experiences that make them a family of sites.  There are “signature elements” of their design.  With Wave 3 we have the opportunity to redefine the look and feel of our site.  What is the next generation of our visual style and interaction model?  How do we take a quantum leap in our web-design?  How will Wave 3 “pop” for customers?  What should stay consistent in our web design, product taxonomy, information architecture in order to create a sense of familiarity for our customers? What should our next generation of standard controls be?  Will we pursue a ribbon or other common element?  What is our platform target for the browser?  What is our down-level target?  Mobile?  AJAX?  What is the role of SilverLight in our design?

This theme involves close collaboration with the Internet Explorer, MSN, and Search teams.  Examples of features we could build to support this theme include:

* Make it seamless to navigate between content on MSN, information on Live Search, and Windows Live services
* Dramatically update our web-based experience with professional themes and controls that have a new level of performance, quality, and interaction
* Provide a richer browsing experience when running on Internet Explorer, including roaming of favorites and browser settings, and a toolbar and suite-header that naturally extends the browsing experience.

Seamless Windows Experience

As mentioned in our bets, we will invest to deliver a seamless experience for customers who own a Windows PC. We have a unique opportunity to remove the seams between Windows, our applications, and our services. Windows Live Wave 3 will be designed so it feels like a natural extension of the Windows experience.

We have an opportunity to make it much easier for customers to “get started” with Windows Live. Our goal should be to have customers log in, type their Live ID, and then they are automatically “set up” with Live. For new machines, we want Windows Live to come with the experience and will consider investments to make this experience easy. For customers who are upgrading from Windows Vista to Windows 7, we will explore ways to make it easy for them to get Windows Live – particularly for photos, calendar, and movies where our applications complete the experience.

We will “light up” the Windows experience with Windows Live. One way to think about this goal is that from 10’ away a customer can tell that a Windows PC has Windows Live – whether through a new theme or other feature. What does it mean to “light up” the start menu, taskbar, sidebar, and folders? What happens when a customer types their Live ID in their Windows account? As an example, we could “light up” the user tile on the start menu with their picture, add presence information, and automatically replicate and roam their documents, photos, and other media. We could roam a set of Windows settings, including background bitmap and other preferences, making it easy to make one PC look like another PC. Our family safety solution could naturally extend the Windows experience for parental controls, providing reporting and content filtering as well as account management.

What’s the relationship between a Windows account and a Windows Live ID?  Should we have a LiveID connected to account settings?

The Windows 7 platform provides new enhancements that allow us to deliver even richer experiences for customers. We will invest in differentiated features that “light up” on Windows 7, and in this theme we will identify these “signature elements” – gestures, ribbon, or other – that make our suite best on Windows 7. We will explore innovations in graphics and presentation, including window management and high-DPI support, that make our applications feel distinct and “pop” on the new platform. What experience will we provide when we “light up” Windows with Windows Live? What is better with Windows 7? What experiences or scenarios are Win7 only? How do we take advantage of or lay the foundation to take advantage of some of the hardware innovations already available or planned for Windows 7?

Windows Live will have value for every Windows customer. If you have an email account and use the Internet, Windows Live will make your experience better. (add more here…) For customers who have Windows Live Messenger, we will explore using Messenger to recommend and “upgrade” their experience. For example, if a customer is using Messenger on their primary Windows PC, Messenger can recommend “getting all of Windows Live,” download the software, and enhance their Windows experience. How can we use Messenger to increase the depth of engagement of our customers in our software and service suite?

Our client software experiences today have different experiences for the user tile, toolbars, menus, spelling, and navigation. While there is a cost to sharing code, there is a benefit to customers who will have a consistent experience with our site. What are the common elements that define our client suite? What is our approach to common controls?  Should we have a shared sign in for Live ID or keep it separate? What is the evolution of setup and update?  Should we invest in other shared infrastructure – spelling, editing, parts/extensibility? Should we have “parts” that are shared between Live Writer, the Photo Gallery, and Mail, that enable connection to 3rd party services?

Beyond shared components, what are the shared scenarios for suite? Today it is hard to share photos and add a blog or start a blog and add a photo album. How can we bring our experiences together for publishing, sharing, and communication?

Many customers will use Office and Office 14, and we will work to connect these customers to our experience. What happens when a customer sets up Windows Live and uses Office? It should be easy to use Windows Live Messenger and our communication services with the Outlook client. It should be easy to publish from Office applications to Live Folders.

This theme will involve close collaboration with the Windows 7 and Office 14 teams. Examples of features we could build to support this theme include:

* Make it easy to get set up with Windows Live by typing your Windows Live ID, and automatically download the information and applications required
* Enhance the Windows desktop with Windows Live services and a new theme so customers feel their Windows PC has “come alive” after Windows Live is installed
* Support Windows 7 platform enhancements so Windows Live feels like a natural extension of the Windows system, including gestures, ribbon, and other elements
* Enable a “one-click” way to take my settings, get a Live ID, and “move them” to the service so that POP/IMAP import happens in the cloud.
* What’s our next level of investment in family safety?  What is the experience of parental controls and account management (with Windows 7)?

Part 2 of the Windows Live 3 Planning Memo.

Following a couple of service pack code and information leaks in July 2007, Sinofsky wrote a post on his internal blog explaining why he believed “translucency,” instead of “transparency,” is the best approach for his team and for customers. Because Sinofsky’s philosophy is so integral to how Microsoft 2.0 is attempting to operate, I decided to share it. This blog post was provided to me by a source, who asked not to be named.

7/9/2007
Transparency and disclosure
 
Transparent. Easily seen through or detected; obvious.
Translucent. easily understandable; lucid.
 
Today was a pretty exciting day for the folks working on servicing Windows Vista as there were a number of breathless stories about SP1 including dates and features. These stories caught us (management) by surprise since not only have we not announced any of the things in these stories, but much of what was reported was not or will not be the case. This is not a good situation to be in and I thought I’d offer some words on how we think of “transparency” relative to disclosure.
 
One topic I have been having an interesting time following has been the blogs and reports that speculate about how Windows will go from being an open or transparent product development team to being one that is “silent” or “locked down”. Much of this commentary seems to center around me personally, which is fine, but talks about how there is a Sinofsky-moratorium on disclosure. I think that means I owe it to the team to provide a view on what I do mean personally (and what I personally mean to do), of course I do so knowing that just by writing this down I run this risk of this leaking and then we’ll have a round of phone calls and PR management to do just with regards to “Sinofsky’s internal memo on disclosure”. But I thought it would be worth a try.
 
The most important thing I believe we owe our shareholders and customers relative to how and what we communicate is that whatever communicate to people be accurate and truthful relative to the work we have going on. This does not mean free from ability to change down the road. It does not mean silence until the very last minute. What it does mean is that we should recognize the potential impact our communications can have on customers, partners, and our industry and we should treat folks with great respect because when we do disclose what we’re working on people pay attention—and they do more than listen as they make plans, spend money, or otherwise want to count on what we have to say. When we have to change our plans, modify what has been said, or retract/restate things we not only look like we don’t have our act together, but we cause real (tangible) pain to customers and partners. One need look no further than the Longhorn/Vista product cycle and the cost to the PC ecosystem of us being out there talking broadly before we really were able to speak with the accuracy our customers and partners assumed. Plans were made. Plans were remade. And then finally people just decided to wait until we really delivered, with some folks not really believing us until the DVD was in their hands, which meant they were no on board with drivers, compatible applications, or the support their customers expected. That example is close at hand, but we can look at examples for Server 2008, ship dates that came and went for any number of products, or even recent examples with Windows Live. This is a challenge that spans all of Microsoft, not just Windows.
 
All of these challenges come about because there is a mismatch between expectations and reality—that mismatch or gap is the heart of customer dissatisfaction. What we can do is be thoughtful about planning and then just as thoughtful about how we communicate those plans. That is what we are doing.
 
Customers and partners want to know about SP1 for Vista. Actually they need to know. We want to tell them. But we want to do so when our plans and execution allow that communication to be relatively definitive. We are not there yet. So telling folks and then changing the plans causes many more challenges than readily apparent. While it might sound good on paper to be “transparent” and to give a wide open date range and a wide open list of release contents, we all know that these conversations with customers don’t end with the “we’ll ship by <x> date and we’ve prioritized <quality>”. Folks do want to know “did you fix this <example> bug?” That is reasonable, but we don’t have all those answers and thus we cannot have a reasonably consistent and reliable communication…yet. We are working towards that. While there is clearly a challenge in the near term in not offering details, this challenge is much less than if we get the wrong information out there and we have to reset and unset expectations. Even among our enterprise customers, for whom this type of information is routine, we have a long history of really scrambling these most valuable customers with “information” that turned out to be “misinformation”. The difference we are trying to highlight is the difference between transparency in what we’re “thinking” and transparency in what we’re “doing”. Everyone wants to know what we’re thinking, but making it clear that those are thoughts versus “doing” is a subtlety lost on a mass audience.
 
So our goal as an organization is to be much more thoughtful and considerate with what we disclose. Premature disclosure might make us feel like we were helping. Heck it might even make some customers and partners feel good, and some partners might even understand of the challenges we face in managing our projects. But on the whole it did not make Microsoft a good citizen of the ecosystem and it certainly did not make us good enterprise partners. Being thoughtful and considerate means we will be just as open and just as transparent about roadmaps and plans as we ever were (meaning the contents we disclose) but we are going to work to eliminate the premature disclosure that has low reliability and high error rates—we will have the right materials for enterprise customers, brief industry analysts, and work with partners all with valuable and timely information. Notice that these audiences are our customers and partners and that a non-goal is allowing the news cycle or needs of the press to drive disclosure timing and contents.

 
Just as we plan the software we will plan to disclose our work. It means that we will develop the messages (so expectations are correctly set), the supporting information (so all the details are there), and the overall communication plan (so we don’t leave anyone out). Product Management owns and drives this. In many ways this is their product deliverable. Just like we don’t want people running to demo a feature hot off a build machine, we don’t want to rush to disclose until we have these plans in place. Our PMG team is dedicated 100% of the time to communicating in a planful way this information to the Microsoft field, customers, partners, and the press. They are not perfect, but like all of us their strive to do their best, learn, and improve each turn of the crank. This is a key point which is that we are trying to be new and improved with respect to disclosure, and one thing we need to do is go out and make sure we set expectations on what new and improved means and how we will be working.
 
But our PMG team cannot do their job effectively if they end up in reactive mode. Stories like the ones about SP1 (or similar leaks about Live Services) make getting the word out pretty impossible. It puts us on the defensive. It confuses customers. It makes it so the message we want to get out there—the features we delivered, the quality of the work, the scenarios we enable, etc.—just doesn’t make it through the cacophony of chatter about the rumors, partial information, and other guesses. Of course we can’t be proactive about how we wish to be new and improved if we are always responding to these situations.
 
We also have ongoing disclosure with customers and partners of all types as we get their feedback and input about how we should evolve Windows. These discussions are about what we’re “thinking” and when done in a manner in which expectations are clear are super valuable and critical. We do this in a deliberate and constructive manner. These are dialogs. They are not press releases. We work with customers. We provide tools to the field that talk about what we do know about the next releases of our products. We train the field to deliver those messages. Is there enough detail relative to expectations? Never. That is a natural outcome of making sure what we do say meets our over-arching goal of being truthful and accurate in what we say.
 
Some folks think that it is a good idea to tip off the press or give a customer (even under “NDA”) early details of what they are certain we will be communicating in the future. Please don’t. This doesn’t help. It only feeds the frenzy and diverts attention from doing a good job. This is especially true when we burn “news cycles” responding. The ripple effect of the SP1 stories is immense—our PR team, OEM teams, enterprise sales, IHV relations, and on and on all spent the past 24 hours (yes these folks all are on call) scrambling to address the rumors. Ultimately, this means we spend less time planning how we will talk about and disclose the work we are doing. And ultimately it just causes problems for everyone. Even if one person, somewhere and for some reason, felt like it was the right thing to do by disclosing what they believed to be the case.
 
All I’m asking folks to do is think before they disclose—in person, in blogs, over the phone. Our product management team owns disclosure and owns communicating with the world the work we are doing. They take this work seriously. They have a very strong desire to tell people about what we do—it is their job. They want to do this well and that takes discipline from everyone involved. Please help.
 
I know many folks think that this type of corporate “clamp down” on disclosure is “old school” and that in the age of corporate transparency we should be open all the time. Corporations are not really transparent. Corporations are translucent. All organizations have things that are visible and things that are not. Saying we want to be transparent overstates what we should or can do practically—we will share our plans in a thoughtful and constructive manner.
 
The upside of being deliberate is that we hope to exceed expectations with what we do. That is not to say that if we are silent people will expect nothing so anything we deliver is great. Rather since we will be talking all along about what we do (in a planned manner) when we show off the software it comes to intrigue and excite people because it does what we said it would, but it does so in an elegant and thoughtful manner, and that it really does what we said it would do, and it does so spectacularly well. We are different than some companies in our industry because our success is dependent and intertwined with that of thousands of other companies. We take that extraordinarily seriously and thus our communication is designed to take that into account by sharing actionable, accurate, truthful, and complete information in a timely manner—timely means that there is time to act and if acted upon the results are what we collectively hope to achieve.
 
–Steve

In the early Microsoft days, these papers were secret. But in the middle of this decade, Microsoft began sharing these Think Week papers publicly inside the company, allowing employees to comment on them and to see Gates’ and other key Microsoft executives’ comments on them.

One of these papers, shared with me by a source who requested anonymity, provided a good sense of some of the “cloud-computing ” infrastructural issues with which Microsoft has been — and needs to be — grappling.

Because Microsoft is spending so much on building out its datacenters and people in the online business space in order to gird for the Web 2.0 and 3.0 battles, the issues described in this paper are especially interesting. And there are some hints about the still-under-wraps Microsoft CloudDB and Blue technologies that are also rather intriguing.

An Edge Computing Network for MSN and Windows Live Services
Author: Jason Zions
Date:  12/15/2006

Abstract

Microsoft’s online properties create and monetize rich content and innovative end-user experiences. To meet their business objectives while providing the quality of service needed to attract and delight an audience, they must overcome a variety of technical and operational challenges. Many of these challenges arise from the current architecture of the properties themselves and from limitations of the infrastructure which supports them.

The Edge Computing Network (ECN) extends Microsoft’s existing core network and data center infrastructure with intelligent computing nodes at the “edge” of the network cloud. This distributed computing network provides a set of net¬work, com¬puting, storage and management resources and services closer end users.  The ECN goes beyond traditional Content Distribution Networks to enable a wider range of application architect¬ures that offer improvements to performance and robust¬ness and reduction or elimination of some operational challenges.

This document identifies the challenges faced by Microsoft’s on-line properties (as well as some problems created by current implementations), lays out the vision for the Edge Com¬puting Network, describes the progress already made towards achieving the vision, and works through two scenarios showing how the ECN could be exploited by a property.
 
Challenges Faced by On-Line Properties

Microsoft has roughly 150 on-line properties covering a tremendous range of scale, customer base, and functionality. Despite that huge variation, they each face a fairly consis¬tent set of challenges that fall into a few broad categories.

Network Challenges

Any internet-facing service has to deal with network latency, which degrades the usability of the service. The larger the latency, the worse the problem; the usability impact is mul¬tiplied by the number of round trips. Common web site development can results in tens or even hundreds of round-trips to display a fairly complex page; each separate graphical element gets retrieved independently. Various technologies have been created to deal with this. Content Distribution Networks (CDNs) like Akamai and Savvis sell distributed small-object caching services, relying on a global network of points-of-presence (POPs) housing web servers which serve static graphical elements. These POPs are located so as to have much lower latency (with respect to the end user’s browser) than that of the owner of the actual on-line service.

Similar problems arise for streaming video, although the problem is due more to packet loss and variation in packet delivery times than in pure latency. Streaming as well as straight downloading of large files raise issues of bandwidth provisioning; serving many downloads and streams from a single origin would require very large (and very expensive) egress from that origin to the Internet, and the Internet’s backbone itself isn’t growing in capacity as fast as Internet traffic overall is growing. CDNs sell services which solve those issues as well, serving large files and streams from many distributed POPs around the world.

The past few years have seen the rise of Distributed Denial of Service (DDoS) attacks, in which a large army of zombie attacking nodes attempt to overwhelm a single service. While usage of CDNs can protect static content from DDoS attacks, a single origin hosting dynamic content is still vulnerable.

Operational Challenges

Properties have to acquire and provision server hardware to host their services, and they must size their resources to match their expected peak load. Any time the load on a service is below the servable peak, money is being wasted, both in the form of under¬utilized capital equipment and in the form of power to run and cool unneeded servers. Worse yet, pro¬per¬ties must continue to accurately predict their peak usage as they grow; insufficient capacity for peak load results in slow or interrupted service, reducing customer satisfaction and leading to defection of users to competing services.

Since properties acquire their own servers, they typically attempt to optimize their choice of equipment for their specific application. This means that, across all of Microsoft’s on-line properties, there is only limited commonality of hardware. Excess servers cannot be easily repurposed to other properties. Also, the cost of maintaining large numbers of servers scales linearly with the number of SKUs; we can’t fully leverage the sublinear scale of cost versus the total number of servers. Finally, OS deployment onto such a large variety of ser¬vers is quite complex, introducing still more cost as well as greater risk of misconfiguring some systems or missing a patch. Rigorous standardization of servers is of only limited help; specific hardware models eventually leave production lose support from vendors, and new models must therefore be introduced.

Each property is responsible for its own monitoring, reporting, and logging systems. Most early-stage or small properties can’t afford to invest significant time or money in auto¬ma¬tion; these tasks are handled in a more expensive ad hoc manner. There are some projects under way to build common tools in this space (e.g. MAGIK) , but these tools are designed for the needs of the largest properties (particularly Live Search and Hotmail) and are too inflexible to meet the varied needs perceived by the great majority of properties.

Many existing properties have centralized architectures that cannot be easily geo-distri¬buted. As a result, there are limits to their growth based on the pace at which Microsoft can build sufficiently large datacenters to hold their infrastructure. These limited architec¬tures made sense early in the life of the property when servers were hosted in only one datacen¬ter; the simplifying assumptions made possible by that design are pervasive through¬out the code, though, making the overall implementation inflexible. Some assump¬tions render the implementation highly fragile, e.g. those related to multicast architectures or round-trip time to access various elements of the pro¬perty. For some properties, capacity growth can be achieved only by adding new racks of servers physically adjacent to already deployed servers. These kinds of constraints make consolidated manage¬ment of datacenter space extremely difficult; smaller properties are often “bumped” from datacenter to datacenter, sometimes repeatedly, to allow larger, monolithic/fragile pro¬perties to expand.

CDN as Partial Solution

CDNs can be used to solve some, but not all, of the network challenges described above. Some properties use one or more services from various CDN providers to address the specific challenges they face. Very few properties leverage the full suite of available CDN services, and many properties use none at all. Various factors play into these choices.

CDN services are not inexpensive; Microsoft spent about $40 million on CDN services in FY06. Projections of future growth (based on expected growth in the number of properties, amount of traffic, and usage of CDN services) show this growing to more than $130 million in FY11.

CDNs can only be used to handle static content. While there is some limited capability for hosting application code remotely via a CDN (e.g. Hotmail’s “AATE” component on Savvis), there are significant drawbacks: cost is significant, capacity is limited, management and deployment tools are primitive, and Personally Identifiable Information (PII) cannot be used there.

All CDNs provide a “traffic manager” or “global load balancer” (GLB) service which directs user requests to the location most appropriate for serving the request. The GLB services provided by the various CDNs are limited in sophistication; because of their general-purpose nature, they cannot take into account application-specific quality-of-service needs, and they cannot route traffic based on Microsoft-specific business logic (e.g. from which locations can Microsoft serve this traffic without paying for bandwidth, based on our current peering relationships, time of day, link utilization, etc.).

The logs created by CDNs are difficult to access for specific purposes. Raw billing data is expensive to retrieve and process. While Microsoft can meet its regulatory and forensic needs in cooperation with CDN vendors, the process is sometimes quite complex.

Edge Computing Network Vision

Partial solutions can only get us so far; Microsoft’s properties need an integrated and complete solution to their technical and business challenges. The Edge Computing Network is one such complete solution.
The ECN vision is based on the following four assertions:

1. Quality of Service (QoS) and Scalability are critical to the success of Microsoft’s online properties.

2. When delivering global online services/applications, centralized architectures are in¬adequate because they lead to poor QoS and do not scale well.

3. For the same purpose, distributed architectures can achieve superior QoS and scalability.

4. Centralized technologies and the people who are familiar with them are plentiful whereas distributed technologies and the people who know them are few.

The Edge Computing Network will contribute to the success of Microsoft’s various online properties by enabling properties to make use of a comprehensive set of optimized and easy-to-use distributed computing services and resources. Properties can focus on de¬li¬vering compelling end-user experiences and addressing their business priorities without having to make the daunting choice between spending unsustainable amounts of money to compensate for architectural limitations of the centralized approach or developing and operating a proprietary distributed computing network.

The Edge Computing Network comprises roughly 24 nodes distributed worldwide. Most Internet users will be no more than 20 msec roundtrip time away from at least one node. Each node provides traditional CDN services and also provides distributed computation and storage services capable of hosting elements of Microsoft’s own on-line properties. Each node would have egress capacity to the Internet end-users in the region. The nodes would be connected to each other and to existing Microsoft data¬centers by a network overlaid on the Internet and on private links leased or owned by Microsoft.

CDN-like Services

Based on the service needs of properties desiring access to the customers within a region, the ECN node serving that region would be provisioned so as to provide an appropriate subset (and capacity) of these services:

• Small object caching
• Large file downloads
• Media streaming
• Peer-to-peer file transfer
• Smart (business-rule and load aware) traffic management and load balancing
• Traffic and user analytic data
• Logging to support billing, regulatory compliance and forensic demands
• Monitoring and management of services

By providing these services in-house, Microsoft can extend and enhance these services beyond what is possible through external CDNs. Properties can more effectively use these ser¬vices when when developers have visibility into the details of their actual functioning. For example, we can do a much better job of distributing downloadable content around the world in res¬ponse to sudden changes in demand patterns. Traffic management can take into account information which Microsoft is unwilling to divulge to third parties. Our control of the soft¬ware we ship can permit us to build P2P distribution mechanisms so they improve per¬formance for end users while reducing costs to Internet Service Providers and to Microsoft.

More importantly, Microsoft can control the exact “footprint” of our network, citing nodes in locations which are most important to our business and using our bargaining power and relationships to do so in the most cost-effective manner.

Many of these goals could be achieved in partnership with a third-party CDN; however, that CDN would be free to sell those same enhanced services to others, including our com¬peti¬tors. Our intellectual property would be used to the benefit of the very companies over which we seek to build competitive advantage.

Microsoft already has considerable intellectual property in this space; we hold a substantial portfolio of patents across these technologies. While we could implement all of these ser¬vices from scratch, the most cost-effective way to get these capabilities deployed and working to our advantage is to acquire existing, functioning technology from one or more CDN providers.

Distributed Computation and Storage

Based on the service needs of properties desiring access to the customers within a region, the ECN node serving that region would be provisioned so as to provide an appropriate capacity of these services:

• Application containers
• BLOB storage (replicated and local)
• Transaction-oriented (database) storage (replicated and local)
• Distributed file system
• Logging to support billing and diagnostics
• Monitoring and management of services

An application container is an isolated environment for running application code. Con¬tain¬ers come in a single “size”; that is, each container exposes to the code it hosts a specific amount of physical and virtual memory, a single processor of defined speed, and a maxi¬mum amount of internet egress. Property developers construct elements of their applica¬tions to fit those containers. Scale of service for a property is achieve through running code in the desired number of containers, rather than through increasing the capacity of a single con¬tainer; that is, capacity is provided in a scale-out, rather than scale-up, fashion.

Each container is isolated from other containers running on a given physical system and from the host OS which runs directly on the physical system. Applications are unaware of the precise mechanism used to provide this abstraction or virtualization. The provisioning infrastructure is capable of starting or stopping instances of a particular computational element based on a variety of criteria:

• Scheduled by time-of-day or time-of-year
• On demand of operations staff
• Automatically in response to measured load on already-running instances
• Automatically in response to load on other nodes

Scheduled instance provisioning is the easiest to provide but still provides property own¬ers with solid tools to control quality of service versus the cost to provide that service. Financial properties (e.g. MS Money) would schedule increased capacity during end-of-month, end-of-quarter, and end-of-year periods as well as tax-preparation periods; this could easily be adjusted to match regional fiscal reporting requirements (e.g. April 1-15 in the US but April 16-30 in Canada). Operations staff could easily override these schedules to meet unusual demand.

More interestingly, the provisioning infrastructure could monitor some property-defined load measure and dynamically adjust the number of instances of distributed elements of that property to keep the load within a specified range. Code would compute that load fac¬tor so that it accurately reflected the quality of service provided to the end user. De¬pending on the nature of the property, live code could be run on the end-user’s system which meas¬ured actual elapsed time for various operations rather than simply relying on brute-force measures of round-trip times; these real time values would be combined with server queue lengths, memory pressure measures, etc. to help determine whether ad¬ditional instances would improve subpar response times or to decide that the property is over-provisioned and can release one or more instances.

Automating the balancing of service between ECN nodes is a key element in providing pro¬tection against DDoS attacks. By integrating this automated balancing with the routing ca¬pability provided by the Traffic Manager, the ECN will be capable of providing service at “nearby” nodes in the event a single node is under attack. The provision of additional serv¬ers and addresses for a service will dilute the impact of the attack; the fact that the addi¬tional addresses are more distant (network-wise) from the attackers will decrease the rate of the attack as well.

Storage in the ECN node comes in a variety of types reflecting the varying needs of app¬lications. All storage usage would be limited by per-property quotas and tracked for billing. Application code (exe and dll files, etc.) need to reside someplace visible to various app¬lica¬tion containers within the node; a distributed, replicated file store would address code distribution needs, and Microsoft has several such file systems either in release, in development, or in research.
Applications need to store chunks of data for a variety of uses; some are purely local to an instance, others are intended to be shared amongst all instances of the application running anywhere in the Edge Computing Network. An application would tell the blob storage infra¬structure about the replication needs for each category of blob data: local to this instance, local to the node or a set of nodes, and/or rep¬li¬cated to a backing store in one of Microsoft’s large-scale datacenters. The blob store would be built on top of the Cheap File Store (used by Windows Live Spaces), Blue (used by Windows Live Mail), or some variation on those which accommodates the extended rep¬lication needs of ECN.

Some applications need transactional storage; again, replication needs would vary by ap¬plication. For properties needing distributed database semantics, a service like CloudDB would be provided. Properties needing a database purely local to a node would use SQL Server on top of local non-replicated storage; a property which needed off-node replication of data for business-continuity and backup could build such a mech¬an¬ism using existing and forthcoming services provided by the major datacenters (e.g. Blue).

Overlap of CDN and Distributed Computation Infrastructure

While the features and implementation of the CDN and Distributed Computation infra¬struc¬tures have been described separately, there’s no reason they need to remain separate in practice. Over time, the various systems which provide CDN functionality would become applications running within Application Containers in the ECN node. Stor¬age previously dedicated to small-object caching, large file download, and streaming would be integrated into the blob storage infrastructure of the distributed computation environment. Logging and monitoring systems would converge.

No CDN provider today does things this way; none has a general-purpose distributed com¬pu¬tation environment. The ability to dynamically reallocate resources to match needs across the spectrum of CDN and computation services is a unique advantage of the Edge Computing Network we envision.

Status of the ECN Program

An attempt was made to acquire a CDN vendor; changes in the economics and valuation of the various players in that market space drove the price of the acquisition outside the ac¬cep¬table range. As a result, negotiations have taken place to license tech¬nology from one or more CDN vendors; this technology would serve to jumpstart the ECN implementation. The agreement will include consulting services and operational assistance for a period of time, enabling Microsoft to acquire a set of best practices from a successful CDN.

A list of 24 broad locations for ECN nodes has been roughed out; selection of the first three and second three specific sites is under way. Site selection is strongly influenced by the ex¬perience of the Windows Live Infrastructure team responsible for siting and managing Mi¬crosoft’s major data centers. The current plan has the first three sites built out by the end of FY07, the second three sites built by mid-FY08, and the remaining sites from the full list of 24 coming on line by the end of FY09.

Selection of technologies for the various components of the distributed computing environ¬ment is under way. Of particular interest is the technology used to support the application container itself. An appropriate balance needs to be struck between isolation and perfor¬mance; the selected technology needs to provide enough management tools and hooks that beta-quality services can be provided in three sites at the end of FY07. Microsoft has a va¬riety of virtualization and isolation technologies entering the product stream over the next two years (e.g. Silos); ideally, deployed properties would be abstracted from the app con¬tainer technology so as to allow the ECN implementation to change over that time without requiring major rework.

Scenarios and Examples

A wide variety of scenarios were used to drive the development of requirements for the Edge Computing Network. A small set of scenarios are presented here to give a flavor of the ways properties can leverage the ECN to change the way they do things.

Cricket Match Play-by-Play

The population of nations which play international-level cricket exceeds one billion. During the course of a five-day test match involving his national team, a cricket fan is likely to be viewing a small web service which displays end-by-end descriptions of play and the cur¬rent score; users typically check the window every 4-8 minutes for at least six hours.

This type of property is ideal for deployment within ECN. Minimal state information needs to be saved, and none of it is per-user. The dynamic content (play-by-play data and current scores) is generated near the edge and can easily be injected there. Service instances can be started to meet current demand and can be shut down at the end of each day of play and at the end of a match. No servers need to be provisioned in any central datacenter; compu¬ta¬tion resources are only consumed (and thus only paid for) during a match. Archival infor¬mation (play-by-play of the previous day’s play and previous matches) can be stored using replication services; access to archival information wouldn’t require caching the data at an edge node, but the application could detect broad interest in particular archival data and elect to cache that data.

More interestingly, a peer-to-peer network could be built dynamically which would reduce the amount of network egress bandwidth required from the ECN node while increasing the speed at which updates could be pushed to the entire network of users. By taking advan¬tage of knowledge about ISP network structure (i.e. which IP address ranges belong to which ISPs) and structuring the P2P network accordingly, ISPs would also see reduced in¬ter¬net downhaul costs; information would flow from the ECN-deployed application into a small subset of an ISP’s total user base, and those users would pass the information amongst themselves within the P2P network.

Knowledge about a user’s location and interest in cricket and in their nation would be used to drive the delivery of appropriate ad content. Since update frequencies are predic¬table, the AdCenter property could spend longer than the normal 1-2 ms time budget allocated for selecting the ad which generates the highest income to Microsoft. Ad content would be dynamically cached at the ECN node; some ads (e.g. those tied very specifically to a par¬tic¬ular series of matches) would automatically expire and be flushed from the cache, while others would remain to be reused by AdCenter through other properties.

Hotmail

Today’s Hotmail service is monolithic; it isn’t easily geo-distributed, it has huge storage requirements, and it relies upon other services (e.g. address book, passport) with which it expects to be collocated. Given the capabilities of the ECN, some parts of this can change.

The Passport service itself could be deployed into the ECN. Each user’s passport profile would be stored in the ECN node closest to that users normal location with backing store in a primary datacenter. User sign-on would be much faster, since long round-trips to a single datacenter in the US would be eliminated. The secure nature of an ECN node allows us to store passwords and PII information right on the edge nodes.

Hotmail itself could cache mailbox header information in the ECN node closest to the user. By caching just headers, the page-load time for the initial mailbox view would be drama¬tically reduced, addressing one of Hotmail’s biggest competitive disadvantages against Yahoo Mail and Gmail. The cache could be pruned by code written by the Hotmail team and deployed to the edge independently from the Hotmail service code.

Intelligent pre-fetching of individual email messages could be performed; the Hotmail application code can predict which message is likely to be the next one the user wants to view. With the cooperation of an AJAX-style client, the ECN-resident component of Hotmail could use various forms of HTML and TCP compression to more efficiently send content to the client browser.

Acknowledgements

Several people have contributed to the vision and ideas described in this paper: Jeff Cohen, Vidyanand Rajpathak, Hang Tan, Ben Black, and Robyn Jerard. Thank you for your know¬ledge, review, and spirited argument.

You’ve come to the right place if you are looking for the book site/blog that is meant to complement the book Microsoft 2.0: How Microsoft Plans to Stay Relevant in the Post-Gates Era. The book is due out in early May 2008 from John Wiley & Sons, just a couple of months before Microsoft Chairman Bill Gates relinquishes his day-to-day duties at the company he founded more than 30 years ago. As of July 1, Microsoft officially begins its next, post-Gatesian chapter.

This site will include information about the book; supporting material that didn’t make it into the printed pages and regularly updated information on the products, people and strategies detailed in Microsoft 2.0. Because I had to submit the final manuscript earlier this spring — shortly after Microsoft made its $44 billion acquisition bid for Yahoo — the book had to be “frozen” in that moment of time. But as Microsoft’s acquisiton moves forward (or doesn’t); as more Softies quit (and new ones join); and as Live Mesh, Windows 7 and other new products begin to take shape, I’ll be covering all that and more on this site.