On the flight back from Burbank last Friday, I sat next to Nitin Duggal. Nitin is a Product Manager with Yahoo! search.

We quickly got to talking about Web Office / Enterprise 2.0 and the strategies pursued by the big players: Microsoft, Yahoo!, Google and EBay.

Nitin said something that struck me as brilliantly insightful. Nitin said that Microsoft is losing their Network effect.

People used to have to buy Windows machines because that was what you needed to run MS Office and most other software. But people aren’t building applications on the Windows any more; they are building them on the web.

Nitin is right. All the coolest new applications are Web based. WordPress, Movable Type, Flickr, SalesForce.com, Gmail, Sphere, Basecamp, Joyent, Zimbra, and LinkedIn are all web based.

Google’s new spreadsheet tool is yet another example. I tried it out today. It’s a capable spreadsheet engine. Vlookup works. You can fill down with Ctrl-D. You can fill right with Ctrl-R. Some things are a little different. When you set up a formula such as sum(c2:c20) you can not select the cells with your arrow keys. Instead, you have to select the range with the mouse.

Will this kind of spreadsheet engine cause people to jump off MS Excel? I am not sure. It’s nice to be able to work on a spreadsheet together at the same time, and this might solve the problem of “who has the latest version”. That the spreadsheet works with google’s chat tool is really interesting. Chat, combined with more robust access control that is integrated into the email directory and an audit trail tool would begin to represent a radical improvement over today’s tools.

The Implications - Web 2.0 means new opportunities to create Network Effects

The implications of this are obvious:

1. Web 2.0 means the network effect is now on the network.

   MSN and Office Live now count more than the number of existing installed users. From blogging tools like WordPress and MovableType, to Wiki tools like Wikipedia, and SocialText, web based information will replace information stored in MS Word. And because it is web based, it can be searched, it can be leveraged through cross-links, and behind the scenes, it can be captured and stored in a structured format, which in turn can be used as the basis for all sorts of Mashup based application.

2. Web 2.0 means knowledge workers don’t necessarily need Windows.
   

   Web 2.0 tools, including basic tools like blogs and wikis mean you can access and work with information from any operating system. If you are writting a blog post, you can be running Apple’s OS/X, Ubuntu Linux, Solaris or Windows.

3. Network Effects in the Web 2.0 world will come from Social Networks and, perhaps more importantly, from Standards.

   
   

   The new network effect will be built upon two things.

   The first comes from social networks built up around these new Web 2.0 applications. Once you start to use Flickr to share your photos, you are more likely to stay than to switch. This isn’t always true and it doesn’t mean that users will never move away.

   The second network effect comes from the spread of information and communication standards. The most popular include plain old HTML, RSS and ATOM. This presents an interesting dilemma for vendors: do they play nice with others and try to compete on an open level playing field of common standards, or do they try to go it alone. The down side of competing on an open level playing field is that you truly have to have the best solutions to win. Considering, however that the network effect is now on the network, you do not really have an option?

There is two Achilles heel that could stop the movement of applications from the Windows desktop into the Web Office powered world of Enterprise 2.0. The first Achilles heel is interoperability standards. The second has to do with deploying the servers.

Interoperability: Feeds, Cut & Paste, Widgets, Mashets and Microformats

There is a simple reason why Windows and MS Office became so successful. It was easy to cut & paste between applications. Sure it didn’t always work, but generally you could copy a table from Excel into Word or PowerPoint. This made companies adopt the technology. Once corporate America has standardized on MS Office, consumers followed suit. What was a “home office” if you could use the same tools you used at the “real office”.

In the Web Office / Enterprise 2.0 world, creating a standardized way of moving information between applications is a must have requirement before wide spread adoption.

Some of those issues have already been worked out. Feeds powered by RSS / ATOM have created a smooth read only way of moving information from one application to another.

Interestingly enough, Microsoft’s Ray Ozzie solved the pure humane powered cut & paste problem in a very generous and open way. Ray’s LiveClip idea is simple: add a button to a web site that puts structured XML based information onto the browser’s clipboard. For example, you can copy an event in a format that conforms with the hCalendar microformat from one page, go over to your online calendar, hit the paste button and boom, the event is plugged into the right section in your calendar.

As an aside, Technorati has just recently added Microformats to their search engine.You can learn more about their plans in the Technorati Kitchen.

Widgets, which fold information from 3rd party sites into your web page are starting to become prevalent. Wordpress has a widget tool, as does TypePad. These are similar to Microsoft Gadgets, Google Modules and Yahoo! Widgets. Check out LabPixies. The method of integrating these mini applications into a web page is not yet standardized. It needs to be.

In the Web Office / Enterprise 2.0 world, this concept will be carried one step further. For example, I have already talked about how companies will use People Pages powered by blogging software to help employees communicate about the projects they are working on, and the skills they bring to the table.

I have described the next step as an AJAX Badge, but I think that a Mashet might be a better name.

The different between a Mashet and a Widget is that a Widget goes on a home page or news reader page, while a Mashet is something you post inside you existing blog posts.

The first real example I have seen comes from MajikWidget. Guy Kawasaki pointed me to MajikWidget. I think what MajikWidget are doing will revolutionize the way people think about blogging. No longer are blog posts just words, now blog posts can become containers for a full-blown application. Here’s an example of what they look like.

There are other Widget tools out there, but MajikWidget is different because it stores state and provides an interactive interface. MajikWidget could be used to replicate everything you can do on all of the 37 signal offerings including track milestones, collaborate on a document or share document. However, with MajikWidget you can do all those things within the context of your blog posts.

In the new Web Office / Enterprise 2.0 world, the network effect that came from cut & paste within MS Office, will be replaced by a network effect built upon interoperability. Specifically, standards around Feeds, Cut & Paste, Microformats, Widgets and Mashets will define how this interoperability works.

It is clear that while Microsoft can and is contributing to the definition of these standards, Microsoft can not establish any Web 2.0 standards in a way that excludes their competition. Success for everyone requires cooperation.

Deployment Part 1: AAA= Authentication, Access Control, Audit

Big companies need to be able to authenticate their users, limit access control to certain tools and pieces of information and keep an audit trail of some things.

Most Web 2.0 tools were not designed with these objectives in mind. Bloggers, open source advocates and the folks who contribute to things like Wikipedia are generally not interested in creating barriers to information exchange. However, when you are dealing with things like account information at a bank, or medical records, you have to be able to restrict access and create audit trails. Web 2.0 tools can be used within banks and hospitals to facilitate internal communication, but Authentication, Access Control, and Audit trails have to be added into the mix.

If Microsoft moves their current Network Effect from Windows into a position where they sell the tools that provide these critical AAA services, then they have an opportunity to maintain their dominance in the enterprise space.

Microsoft has done some interesting thinking in this area. Kim Cameron, has a great blog that deals with many of these issues. I recommend checking out here white paper, The Laws of Identity, if you want a good place to start.

This is not to say that all three AAA services do not exist in many Web 2.0 tools. SocialText has built some of this functionality into their enterprise wiki tool. As has Blogtronix and iUpload. Movable Type Enterprise and WordPress both have LDAP plug-ins. However, there are no consistent AAA standards across these tools, and working with them from an Enterprise IT perspective requires a unique solution and a unique integration effort for each vendor.

Deployment Part 2: Software as a Service / ASP Vs. Conforming to Life Behind the Firewall

Compounding the issues of associated with authentication, access control and audit trails are problems of physical deployment.

Here’s an example of what I mean:

• Movable Type Enterprise works with IIS, but requires MySQL or Oracle as a database.


• WordPress recommends “Apache or Litespeed as the most robust and featureful server for running WordPress, but any server that supports PHP and MySQL will do.”
• SocialText is deployed on an appliance. The box is run on a LAMP stack, but IT shouldn’t care. If they do care, they have to have Linux, Apache, MySQL and Perl experts in house.
• Joyent runs on Ruby on Rails. I personally am a huge Ruby on Rails fan, but many IT departments do not know what it is.
• iUpload runs on Microsoft .Net, and can be used as a Software as a Service, or in a large installation, can be brought behind your firewall.
• Blogtronix runs on Microsoft .Net, and can be used as a Software as a Service, or can be brought behind your firewall.

Most companies do not have the internal resources to deploy solutions on a wide range of O/S, database and web server stacks.

In most Banks, for example, they try to rigidly stick to just one stack. EJB or .Net paired with Oracle, SQL Server or Sybase are the most common.

Also, most large companies have not worked out a standardized why of working with Software as a Service vendors, although many of their employees are clearly using tools like SaleForce or LinkedIn.

The folks at SXIP have a really interesting solution to this problem.

Their SXIP server provides a basis for maintaining control over services that must be behind the firewall, like authentication, while providing a consistent standards based gateway to Software as a Service vendors.

With SXIP, you bring one gateway server behind your firewall. Then, ask your employees to work with only SXIP enabled Web 2.0 vendors. SalesForce.com is one example.

The SXIP server lets the company provide the authentication, while letting SalesForce provide the app.

If an employee gets fired, you turn off their permissions in the SXIP server, and the fired employee no longer has access to the SalesForce.com account.

The advantage to SXIP is that it is a standard. If you decided to host enterprise blogs at iUpload, you could use the same SXIP authentication server. The same could work for a company group on LinkedIn.

I think that SXIP is the right approach to these kind of problems. When combined with standards for access control, encryption and audit trails, this could give corporate IT the basis for maintaining control while leveraging the best there is to offer in Web 2.0 / Web Office / Enterprise 2.0 service offerings on the open Internet.

Microsoft Could Win Anyway With Web 2.0 Gateway Servers

Imagine a platform of Web 2.0 gateway servers, beginning with SXIP authentication, that worked with your Active Directory or LDAP server and your VPN encryption engine, and provided corporate IT with a secure and standardized way to let their users any Software as a Service offering?

The company that builds this platform and sells it into major enterprises will own the new Web 2.0 network effect. That means they will own the Web 2.0 network effect that is inevitably going to replace Microsoft’s current Windows network effect.

Right now, ownership of this platform is up for grabs. Although SXIP has a nice technical lead. Microsoft, on the other hand, has opportunity to leverage their existing relationships with corporate IT to sell these Web 2.0 gateways as an extention to their existing LAN management tools.

Microsoft’s Robert Scoble and Don Campbell have both stopped by Innovation Creators, so there is every chance that the Microsoft team will be reading this.

Given Ray Ozzie’s foresight with Live Clip, there is also every chance that Microsoft is already thinking about this.

So, is Microsoft losing it’s Network Effect? Web 2.0 means that the value of the Windows network effect is certainly going down. But Microsoft has plenty of time to use their current install base to springboard into ownership of something just an powerful.

Is Google paying attention? I’m not sure. They have a fancy web based spreadsheet, gmail and a calendar tool, but so far, it seems that Google’s only enterprise offering is search.