http://fredchong.net/

or update your RSS feed to point to:

http://fredchong.net/feed/

• I’ve taken on a different role in the Microsoft consulting services organization and had the opportunities to do some really interesting work in emerging markets.
• you’ll notice that many links to diagrams and photos in my previous blog entries are no longer valid. This is the consequence of me ignoring my blog for a while and the fact that the old file server for those links has been retired. When time permits, I’ll try to fix up the links again for the more interesting posts.
• The world of SaaS (which I used to blog about) has moved on. Now, there are the new IaaS, PaaS and cloud computing monikers to reckon with.

I’m really excited to start blogging again, as there is so much to say about the new opportunities and architecture of cloud computing, and my ICT4D projects in emerging markets.

In this post, I want to share about the cloud computing pilot that I’ve been involved with at the Nghe An Tate & Lyle (NATL)sugar factory in Vietnam.

As many of you may already know, the number of mobile phone subscribers in emerging markets is growing at astounding rate, and Vietnam is one of such developing countries.

Another interesting data point is that mobile phone communication and applications play key roles in driving the economic development agenda in these nations. Here we are not talking about mobile applications that use high speed 2G/3G/4G mobile networks for sending and receiving data. Instead, we are referring to content services that use plain old SMS text messaging for transmitting information.

So why is SMS text messaging the preferred mechanism in those places? The reasons are quite simple:

• Mobile data communication infrastructure is just starting to get deployed in developing countries and the signal coverage is very spotty – often only available in the larger cities.

• Many developing countries have agriculture-based economy, which means that huge proportions of the population live in rural areas where there are no Internet or mobile data network. However, basic cellular network infrastructure is almost available everywhere, and signal coverage often reach remote farming sites as well.

• The cost of sending SMS text messages is very cheap in developing nations, often at a fraction of the cost in developed countries. Therefore, SMS is relatively affordable as a communication tool.

• Although Internet coverage may be more readily available than mobile data network, PCs are still not as affordable as basic phones. As power blackouts occur very frequently in rural areas, mobile phone batteries are able to yield longer service cycles. And finally, compared to desktop computers, portable phone are more convenient and better suited for roaming and rugged agriculture lifestyles.

At this point you may be wondering: “what does mobile phones and SMS text messaging has to do with improving quality of life”? Well, it turns out that SMS is a relatively simple protocol which can be used to enable many different applications that have query-response, update and notification-multicast messaging semantics.

For example:

To query for the current weather forecast for the Vietnam coastal city “Nha Trang”, the user could send a text message with this syntax: “weather Nha Trang”. The response could look something like “Stormy; 26 degree Celsius; strong wind and ocean waves; small boats advisory”. Fishermen living in that city may use such a SMS weather service to decide if it is safe to be out fishing in the sea on that day.

There are numerous examples of how SMS-text messaging enables just-in-time information services that could save lives and improve economic well-being:

• subsistence farmers may use SMS to look up average prices of fruits and vegetables sold in various markets. They can use the average price information to decide on the market to sell their produce so they can fetch the highest income.

• micro-finance customers may use text messaging to check their current loan balances, as there are usually no bank branches and ATM machines in rural areas.

• the ministry of health may broadcast avian flu alerts to chicken farmers so that preventive health measures can be quickly implemented in affected areas.

The next example shall take us to my project with NATL in Vietnam, and the video below is probably the best way for me to introduce the project:

As you can see from the video, text messaging satisfies a business need for the sugar factory and the farmers – it helps streamline the business communication between the factory and the sugar cane farming communities. The truck arrival schedule is critical information that enables farmers to time their harvesting activity properly, which results in better quality sugar for the factory and higher level of profits for the farmers.

So what does the project has to do with cloud computing, and especially Microsoft Software + Services (S+S) architecture model?

The following diagram shows the conceptual architecture of the NATL sugar factory solution. It should enlighten us on the question above:

When a sugar cane farmer sends SMS messages to query for truck arrival information, the messages are routed through local telco and SMS aggregator infrastructure deployed in Vietnam. In addition, the content of the SMS messages should all begin with a reserved keyword, for example: “NATL”.

During the pilot testing with farmers, Microsoft made commercial arrangements with a local SMS aggregator to reserve a mobile short code (that the farmers dial to send SMS messages) and keyword for the text messages to be routed to the Mobile Web Gateway (MWG) cloud service hosted in Microsoft data center.

In order to bridge the differences in communication protocols used by telco-centric SMS infrastructure and the Internet, my team has developed the MWG service to translate and route SMS messages between the two kinds of networks. For the NATL pilot, the MWG service is connected with the SMS gateway in Vietnam using the SMPP protocol. To retrieve content, the MWG service used SOAP protocol (it could also use REST) to communicate with the Supply Chain Management Web service hosted in the NATL private cloud. The SMS request is sent to the NATL Web service as a payload in the SOAP body.

When the NATL Web service receives a SOAP request, the service logic parses through the parameters of the SMS requests (keyword, farmer identification number etc…) and makes a SQL request to the content database that has the truck schedule information.

The application response gets sent in the reverse order, starting as a SOAP response with the requested information in the SOAP body. Next, the MWG service reformats the SOAP response into an SMS message and routes the reply back to the farmer’s mobile phone through the SMS aggregator.

The core S+S architecture model that integrates public cloud services with on-premises private cloud services, and  provides value-added content through ubiquitous SMS mobile channel, has enabled us to change lives in a developing country.

Since we completed the pilot last year, I have spoken to a number of other government agencies and NGO in emerging countries who are interested in adopting the same architecture model in their ICT4D initiatives. The scenarios are wide ranging from agriculture marketplace to education content.

The NATL sugar factory pilot is definitely one of the most fulfilling computing projects I’ve ever done, and sugar-coating it with an S+S architecture pitch has made it a double sweet success…

In addition, the paper also explains the different roles and benefits of application marketplaces to ISVs, hosters, VARs and enterprises.

Enjoy.

Here in this post, I'll share my thoughts about the architecture of such an application marketplace.

The high level application marketplace architecture consists of four loosely coupled modules: marketplace services, application designer and workspace, business operation services, and social networks. These four modules work together to provide an end-to-end solution implementation that would put the long tail forces into play.

Architecture Components of Application Marketplaces

Marketplace Services

Marketplace services are offered through a public facing website. Like any other commercial Websites, the user interface of the marketplace website influences and helps mold the first impressions of the marketplace customers.

The business model affects the types of marketplace services being offered to one or more groups of partners and customers: the independent software vendor (ISV) who sells applications (and optionally creates or hosts applications at the marketplace), end users who buy applications, and partners and value-added resellers who repackage and resell applications with their own set of value-added services. If the marketplace also provides application hosting services to its ISV partners, it may choose not have any hosting partners and therefore does not offer its marketplace services to other hosters (potential competitors).

The application catalog functions as a directory customers can use for discovering, viewing and buying published applications. In order to provide a better search and an improved buying experience, the application catalog may implement various types of filters that use additional criteria, such as application category and popularity, to help the users narrow down their choice of candidate applications.

In addition to the application catalog, the marketplace also exposes a set of administration applications that customers can use to edit their account profiles and manage business transactions at the marketplace. For example, an ISV can use the administration application to edit and post the company description, revise the pricing of their published applications, and view revenue reports of applications sold through the marketplace. Similarly, a small business can use the administration application to subscribe or unsubscribe to applications and check the outstanding payments for its marketplace account.

Business Operation Services

While the marketplace services provide the public interfaces for the marketplace customers to publish applications and perform commercial transactions, the business operation services are a collection of infrastructure and management services that operate behind-the-scenes to support the marketplace services.

The identity management module provides a core set of authentication, access control and entitlement services that other business operation services use to secure business data and marketplace transactions.

The monetization module makes it possible for application providers to sell applications using various pricing models, such as charging for user license subscription, application transactions or revenue from online ad-displays. To effectively support a monetization scheme, the marketplace should offer solutions to help ISV define and enforce their application pricing policy (e.g. $20 per user license per month), as well as implement revenue reporting engines that will provide visibility into the income stream for published applications.

The metering and billing module uses the defined policy and usage data logged through the monetization module to generate invoices and bill the customers. The billing engine can offer billing-on-behalf services to the ISV by collecting fees from the business users and crediting the revenue-split payment into the ISV bank accounts.

Application marketplaces frequently provide self-service registration to ISVs or business users for signing up as a new tenant. When the tenant management and subscription module receives a new registration request, a tenant provisioning workflow is started and interacts with the identity management module to create a new organization and administrator user for the new tenant account. The provisioning workflow also creates new entitlement records for the new organization so that subsequent application subscriptions can be properly tracked and metered. Finally, the workflow triggers off a sub-workflow within the hosted application runtime infrastructure to provision a new set of metadata that is used to configure and instantiate the application environment for the new tenant.

Social Networks

A social network at the marketplace not only enables the applications users and providers to interact with each other, it also relies heavily on user-generated content to help users find quality applications and to provide ISVs with market intelligence.

Through a ranking and reputation system, users can get a sense of what other users think about the usability and quality of the applications. These opinions also serve as information that the application providers can use to improve on future versions of the applications.

Many business professionals are now connected with their ex-classmates, colleagues and business partners through social networks. In this world where business success hinges on “who you know”, social network connections can be leveraged to mine valuable information stored in the subscribed applications. For example, a business partner from an external social network may recommend potential sales leads to a business tenant at the marketplace place. The recommendation becomes customer data that is stored in a CRM application purchased through the marketplace.

Application Designer and Workspace

The application workspace is the container that hosts the user interface for business users to interact with the applications. Besides providing the dashboard for using the subscribed applications, the workspace can also include features that enable the users to interact more directly with marketplace services, for instance, including links to purchase additional applications at the marketplace.

The kind of user experience that a marketplace can build into the workspace is virtually limitless. However, simplicity, ease of collaboration and the ability for third parties to monetize should be the key tenets for creating the application workspace fabric.

To mitigate performance and security concerns, there are usually constraints on the types of computations that the third party applications can execute in a hosted runtime environment. As such, the developer should only be allowed to create application features that are contained within the defined computation boundaries of the hosted runtime environment.

For example, the development environment could impose constraints over custom business workflows loops and implement branching logic to ensure that workflows will always terminate. In addition, the development tool can also set customization boundaries on application object models, optimizing system performance by ensuring that database schemas are not overly complex.

It is important to note that the hosted version of the application designer and workspace module is just one mode of user experience. This experience is most beneficial when the target user base has heterogeneous operating system platforms and the browser is the lowest common denominator that can be relied upon to interact with the applications.

There are other hybrid modes of design and runtime experiences where these workspace modules are not hosted by the marketplace. For example, the application may be designed with Microsoft Access running on the application developer’s computer. When the application is completed, its metadata is published to the marketplace. Subsequently, business users who have purchased the application have the option of running the application on their own desktop or within an application runtime environment hosted by the marketplace.

Creating a brand new application platform can be a huge undertaking. Instead of building one from scratch, third parties can create and customize their own application designer and workspace sandboxes by using existing products such as the Windows Sharepoint Services and Microsoft CRM 4.0. These platforms already offer tools for creating applications that can be imported and exported out of the runtime environment as metadata, which can facilitate the process of publishing and subscription applications to and from marketplaces.

+++

For those of you who have been following my SaaS architecture guidance content, you know that I'm a big fan of making the content hands-on and implementable with current shipping products and tools. As you may have expected, I've been working on an sample application marketplace solution using ASP.NET and CRM 4.0. This sample solution will be publicly available within the next few months.

Stay tune as I should have more to share in the near future.

As Chris mentioned in his book, the long tail market is really about niches and therefore long tail software are not the top selling applications.  Niche software could range from specialized business applications that are interesting to only a small number of audience to simple online social games for connecting with people. For example, the event management software your kids' soccer club uses to manage their league matches; and the custom project management application the construction company who remodeled your kitchen uses to schedule their suppliers and sub-contractors.

Niche software applications have always existed since the early days of personal computers. Many of these existed as DOS and Windows freeware/shareware that are posted at online computer bulletin boards. Some of these applications are developed by PC enthusiasts who have regular day jobs but spend their evenings writing small applications for fun and extra cash. Others are written by "micro-ISVs" who hope to make it big when their applications pop into the top 100 chart some day.

Those were the days of niche software - it didn't have the three forces of Chris's long tail economics to ride on. Today, the creator, buyer and seller of long tail software are all better off because of the three liberating forces, and a 4th emerging one that I have added to the list. Let's examine them one by one.

Force #1: Democratize production

Whether it's building a business, consumer or social application, the tools for developing software applications have come a long way. RAD tools such as Ruby on Rails and Microsoft Access 2007 enable data-driven applications to be created without writing a lot of code. Furthermore, hosted application builders from startups like Coghead and Longjump provide hosted tooling and application runtime environment so that application developers can be productive immediately and do not have to worry about deploying their applications. Using mashups platform like Popfly, Web enthusiasts can easily create Web applications that overlay data and services from multiple sources. The current version of Sharepoint designer simplifies the tasks of building composite applications by producing application templates and Web part components that can be composed with other software modules to create custom business applications. In addition, Sharepoint designer's integration with Office Live enables custom business applications for small medium businesses to be designed, built and deployed quickly. No more fussing with the complexity of coding and deploying three-tier Web applications. (For those interested in diving deeper, here's more videos and demonstration on how to build custom applications for Office Live: http://www.innovateonmicrosoftofficelive.com/LearnBuild.aspx)

This point should need no further elaboration - the improvement in application development tools is key to the wide selection of software choices we have today, and the tools are getting even better as we read on.

Force #2: Democratize distribution

Compared to physical goods, software has some advantages when it comes to storage and distribution - thanks to it's digital form. In many ways, software has as "birth right" the capability to be distributed via the Internet. However, this ease of duplication is a sword double edged with the problems of software piracy. So for software, force #2 should really be "democratizing legal distribution". On the other hand, distributing niche software and physical commodities are similar in the sense that they can both use aggregators to lower the cost of selling. For software, the cost of selling is much more than transporting the bits from one computer to another - there are also the issues of consumer trust, brand and quality association and effective marketing to potential users. In the digital world of computer viruses, phishing attacks, mal- and spy ware, a lesser known micro-ISV has much to gain by selling their long tail applications through well established aggregators. A small business owner is much more likely to buy a project management application listed on the Windows marketplace than buying directly from an unknown micro-ISV's Website.

Although distributing software online does not always require trucking physical goods (except when selling shrink-wrapped software, which is still a common kind of commerce transaction) from stores to buyers, the delivery of hosted software services do require server infrastructure and data center operation support to host and run the application services. In many situations, it is not cost effective for an ISV to build and run their own operational infrastructure. Instead, it is frequently beneficial for SaaS ISV to have their applications hosted by professional SaaS hosting companies. Software service delivery is the subject of our recent paper. Look around at the companies providing application hosting services, and you will notice that many hosters are also aggregators of software services who provide the market channels for the ISVs they host.

However, the tasks of coordinating service delivery with SaaS hoster may still be too tedious for ISV producing long tail applications. This is why we are seeing new breeds of startups like Coghead and Longjump, where they provide a one-stop shop for ISV to develop, host, market and deliver their long tail applications.

Force #3: Connecting supply and demand

Technologies that connect consumers of long tail applications are what help drives increasing demand for niche software. Chris calls these types of technologies "filters". Search engine results, reviews, user recommendations, blog opinions, application usage data - all these information influence application adoption and can eventually introduce new user behaviors and habits. Take social apps on Facebook for instance, various filters at that site not only help reduce the cost of search, but also ultimately influence the applications' success and adoption:

As you may have guessed, a one-size-fit-all filter will not do the best job of connecting users with the applications they want to find. We see from the Facebook example above that depending on the user's goal, he may choose among the multiple filters to find applications categorized according to popularity, activities, active users or published date. Not to be neglected is the observation that the purpose of the Websites also act as filters: Facebook is the filter for social network applications; Office Live is the filter for line-of-business applications built for small-medium companies etc.

Force #4: Democratize Capitalism

This particular force is my contribution to long tail software economics. I believe that enhancing the opportunities to make money can further increase the supply of long tail software. The recent proliferation of Facebook applications illustrate this point very well - the authors of Facebook applications get to keep 100% of the ad-revenue attributed to their applications. Furthermore, incorporating an ad-funded software revenue model shifts the economics of when and how software is paid for, which can be explained through a fishing service metaphor. Using ad-funding is as if the long tail aggregator is running a "fishing service" business and software applications are hooks which baits (advertisements) are attached to entice the users to transact for some goods and services. The merchandise and service sellers are willing to pay for the "fishing service" when there is fish action. Effectively, hooks are essential components of the fishing service and the aggregator splits the ad-revenue received from the merchandise sellers with the long tail application providers. The ad-funded model not only provides opportunities for micro-ISVs to get paid, it also reduces the up front cost of trying and using software, thus removing some barriers to drive demand further and encourage software adoption.  

Ad-funding is not the only model to monetize long tail applications. Traditional licensing, subscription and transaction models are all still valid means for the ISV to get paid. The key point here is that the supply curve of long tail software is likely to grow when there are effective mechanisms for the developers to be compensated for their work.

Most long tail applications will never command the kinds of price tags that top selling software are sold at. However, with production tools and distribution democratized, it may only take a very short time and almost no capital investment to build long tail software. In fact, when the long tail software economy takes off, it is very conceivable to see many more micro-ISVs creating applications as fast as them can (in matter of days or weeks) and live off the the aggregate income from multiple long tail applications.

+++

Amazon and Netflix have already proven the profitable business model of selling long tail physical goods and entertainment content. The same cannot be said about the business of selling long tail software yet. However, there are existing and emerging businesses that are standing behind the forces we described above, as noted below in the table:

To prove that the market for long tail software is in fact a profitable one, we'll need to orchestrate the integration of long tail development tools, software marketplaces, filters and monetization schemes deliberately.

May the force be with us...

Happy reading!

Well, here's my theory:- The majority of SMBs in existence are fighting for survival in the market they are competing in, hence they are not (yet) going to spend their hard earned money on software that they are yet to relate with their profits. Unlike the large enterprises and established companies who may be looking to optimize, streamline or differentiate through IT automation, the number one priority for a large chunk of the long tail is finding customers to pay their bills.

When I was working in China earlier this year, Alibaba was one of the Web companies that caught my attention. For those not familiar with Alibaba, it is China's (if not the world's largest) B2B online marketplace. You can find many small and medium sized companies selling all kinds of merchandise through this online exchange. Essentially, Alibaba is offering (market) channel as a service. It also turns out that Alibaba's CaaS strategy attracted the world's largest gathering of SMBs to its B2B service. Serious viral effect is happening at their marketplace.

What does all these mean to SaaS ISVs and catching the long tail?

Well, to begin, if you are serious about selling to the long tail, you should first seek to become their market channel, or minimally, the short link to their channel. Essentially, this is the value proposition you need to create. The SaaS value propositions that convinced the large enterprises do not always work well for the elusive critter looking for the channel to their cheese.

Compared to the aspirations of B2B exchanges like Commerce One and Ariba in the late 90s, Alibaba does not attempt to provide a marketplace that is overly complicated - there is no automation of RFQ, goods requisition etc, just good old basic "listing" and "search". Although the features are very limited, SMB saw the business value immediately. Alibaba's Web site claimed 21 million and 3.6 million registered users in China and outside of China respectively.

Interestingly, Alibaba launched its software company, Alisoft at the beginning of this year with the mission to sell line-of-business applications to SMBs. This press release wrote that Alisoft already has 500,000 active users and 3 million registered users for its free LOB software.

My takeaway from Alibaba's business direction is that bootstrapping for the long tail needs to focus on solving business priorities for the SMB, which to me, is clearly helping those businesses find their customers. As software companies, selling software is of course the eventual goal, but it needs to be bundled with a customers' business channel story.

The combination of offering CaaS and SaaS makes the ISV business model more interesting but also more complicated, such as offering a few different channel + software services bundles with hybrid monetization schemes. For example, the entry level service offering may need to rely on an ad-funded or transactional model, so that SMB that are new to LOB apps can begin to see the value of the market channel and the software. For the more established SMB, the ISV may consider up-selling them to a full featured application that can take advantage of all aspects of the channel offering, and charge a subscription fee for using the application.

Another industry example of players who are well positioned to sell software to SMB are the telcos. Many of the ones I've spoken to plan to become SaaS aggregators. Telcos are not really SMB's channel, but by offering communication services, they are vital links to the SMB's channels. By tightly integrating line-of-business functions (especially those application features that enable interaction with their customers or leads to sales) with communication services, telcos can entice SMB to upgrade their services to include LOB apps.

These last few years, Internet search, consumer-based ad business models and social networks have drawn a lot of high tech investment energy. I suspect that we will soon see a revived interest in online marketplaces since the last of the B2B exchanges dwindled away in the early 2000s. Perhaps this time around, online marketplaces will focus on being the business channels for the SMBs - and they need to in order to fulfill the SaaSy Long Tail prophecy.

More often than not, my attempt to reassure them that my architecture team in Redmond is now feverishly working on SaaS hosting architecture guidance does not seem to appease their anxieties and concerns. (My previous post here described the high level architecture capabilities of a SaaS hosting platform.) I can understand their phlegmatic reactions - hosting is not a core competency for most software vendors and it is no wonder that they are looking for alternatives other than the laborious option of building their own operational capability.

Generally speaking, SaaS providers may consider the following hosting solutions (which can also be thought of as a continuum of options. If you've been following my posts you should now conclude that I really like the continuum way of representing architecture decisions) :

• Software vendor self host and manage their own operational environment
• Software vendor partially or entirely outsource hosting to 3rd party hosters
• Software vendor OEM and license software to 3rd party hosting partners

The right end spectrum of the continuum represents a software vendor's decision to self-host and manage its entire operational infrastructure. The other opposite end of the same continuum represents the decision to OEM and license the SaaS solution to business partners who not only take ownership of the operational responsibilities, but also own the customer relationships with the application tenants. The points between the two ends of the continuum represent decisions to outsource one or more components of the hosting infrastructure to third party hosters. For instance, a software vendor could decide to self host and operate everything except for the billing system which it outsource to a billing solution provider.

For the rest of this post, I want to share some thoughts on the technical and business implications brought about by the above hosting decisions.

When software vendors self-host, they (willingly or unwillingly) become the trusted custodian of their customers' data. The customer data include actual business data as well as workflows and business rules that are configured to run the customers' business processes. The software vendor is (legally and contractually) responsible for the "well being" of the hosted customer data.

Frequently, the self-hosting software vendor is actually sitting on a gold mine of information - lots of valuable business knowledge can be mined from analyzing the stored data. For instance, if the application is an inventory management SaaS application, business intelligence about best selling products of the month and supply forecasting data can be derived by doing trend analysis on the stored data. Therefore, having access to customer's data allow the software vendor to offer valua-add services and information back to the tenants, potentially for additional fee.

Technically, it may also be possible for the software vendor to perform cross section analysis of all the tenants' data in order to generalize business practices used by the clients. For instance, a CRM software vendor may be able to analyse and derive from all the configured workflows, patterns of sales processes with the shortest sales closing period. Such knowledge may then be used in business practices classes offered by the software vendor. Of course, such usage of customer data remains a controversial topic as many tenants will be nervous about their trade secrets being published as best practices.

Publication and resale of identifiable private data obviously violates legal regulations and privacy laws. However, it is debateable if derivation of business knowledge by SaaS software vendor should be permitted, especially if the mining process is non-trivial and involves analyzing large samples of data.  Perhaps we will see more intellectual property protection discussions around this topic when SaaS adopters see their business practices being "popularized" through their service providers. For those who are skeptical about such practices, let me remind you that the financial industry has been capitalizing on customer data and behavior for a long time. (Next time you take a class on stock market trading from an ex-broker, guess where he learned all his best dog tricks.)

The second hosting model frees the software vendor from having to build and operate all or part of the SaaS hosting infrastructure themselves. In most third party hosting agreements, the software vendor would still maintain access rights to their tenants' data, so there should be little contractual restriction limiting the software vendor from pursueing the business ideas mentioned above. Practically speaking, since the hosting is outsourced to third party, there may be technical barriers that needs to be taken care of before the software vendor can have timely access to their customers' data. For instance, the hosting provider may limit when hosted data can be bulk replicated to a data warehouse. This is because the hoster needs to consider the network bandwidth impact on other hosted software vendors when massive amount of (replication) data is being transmitted over the networks.

The third hosting model actually involves a major shift in the software vendor's business model. In this model, the software vendor provides the solution to their hosting partners who host and resell the solution through licensing and OEM agreement with the software vendor. Profit made is often split in varying proportions between the software vendor and the hosting partner. A key implication here concerns the software vendor's access (more accurately, potential lack of access) to the tenant data. In many OEM arrangements, the custody of the customer data belongs to the partner hosters. Obviously this shift of responsibility comes with benefits and downside for the software vendor. While the software vendor is no longer responsible for the operational availability, integrity and security of the tenants' business data, the vendor is also giving up potential opportunities to monetize on those data assets.

In addition to value added services that could otherwise be provided, the software vendor may also be foregoing a channel of knowledge about the application usage and users' behavior, both of which can be accurately obtained through the operational environment. Furthermore, successful Websites often trial test new features by deploying and making pre-released features available publicly on the Internet. Without an operational environment, the software vendor will have to find partner hosters who will stage and trial run its alpha and beta release features.

Although this post may read like I'm trying to convince you that you should be self-hosting, be assured that this is not my intent. There is really no absolute right or wrong hosting model. What's important is that you as a software vendor understands the price and reward that comes with your hosting decisions.

It seems like a lot of good things in this world are bitter-sweet, and now that you are aware of the implications, you'll have to accept that the after taste of your SaaS hosting decision is no exception.

2 weeks ago, I had the unanticipated opportunity to be involved in the organization of Imagination Festival - an event held at Beijing University celebrating the past, present and future of innovation in China. University students and high tech entrepreneurs attending the event seemed to be inspired by Bill Gate's keynote speech titled "Innovation through the Digital Decade". 

Normally, I would not expect to see an honorary tie to SaaS in a technology celebration event like this. However, I was pleasantly surprised that day.

During the event, Bill Gates presented awards to 5 "innovation heroes" as tokens of appreciation for being engineering and entrepreneural forefrunners of the software industry in China. 2 of the 5 heroes honored by Bill have been involved in SaaS-related endeavors:

o    Wang Xiao Cun is a program manager at the Suzhou SaaS innovation center and has been instrumental in the development of the Suzhou software park's SaaS incubation program.

o    Mr. Guo is the founder of MaKaDao - a small SaaS ISV startup that is applying the academic theory of marketing science to real world business problems through the company's SaaS application.

Left to right below (Bill Gates delivering the keynote; China Imagine Cup winners; me with Mr. Guo, the honored innovation hero):

Another highlight at this event is the final competition and announcement of the China Imagine Cup winners. The Imagine Cup is the world's premier student technology competition challenging university students all over the world to innovate with their programming skills. The winner of the China Imagine Cup gets to represent their country in the grand finals that will be held in Korea later this year.

Among the China Imagine Cup entries is one from XinHua University in ChengDu, where the students have implemented a multi-tenant SaaS application for creating and managing university portals. The team also sumitted a commercialization plan intending to capitalize on the long tail of Chinese universities where custom designed college websites for managing student and faculty activities are still considered unaffordable operational expenditure. Although this entry is not the final competition winner (perhaps the demo factor is not as cool), I think the SaaS business plan looks very promising if executed properly.

So there I was, witnessing and enjoying the presence of SaaS at a non-SaaS event.