SOA Guy Walks into a Bar (part 2)

Seems like SOA is continuing to take off. I'm interested in how these organizations are using EA to define the SOA investments. Also, what type of ROI is being captures from Application Modernization in these cases. Or is this simply web-inizing legacy applications.

Cloud Computing

The Silver Lining is Lighting Ready to Strike

Ok, here is my long awaited post on Cloud Computing. I recently got involved in a heated discussion over the “Cloud”. In an effort to more effectively organized and communicate my ideas, I thought I’d turn to my blog.

For background on Cloud Computing go here... (http://www.cio.com/article/439814/Demystifying_Cloud_Computing)

In this (rather long blog entry) I’m going to organize voice my concerns over the cloud and attempt to identify the potential hurdles that will hinder the adoption of cloud computing for running specific applications.

First I will outline the technical limitations of the currently available consumer cloud offerings and other cloud implementations. Next I will look at the consumer market place and discuss how corporate and consumer trends will impact the adoption of applications running in the “cloud”. Lastly, I will argue against three general statements that I recently heard about the cloud.

TECHINICAL

To begin with the technical evaluation of cloud computing that is available to the consumer, I looked towards the Google App Engine and Amazon’s EC2.

The article linked below outlines a number of key points about the Google App engine. The current limitations of the Google App Engine are as follows. First it doesn’t support SQL or Standard DB schema. Instead it uses its own proprietary query language (Yes, Google has something that is not Open) called GQL. Apparently this is to make their environment scale more effectively when performing calls from Web and Application Tiers into the Database Tier. While I’m sure it exists for a reason this is a large down-side that already will limit the effectiveness and adoption of the Google Cloud. Next, it currently does not support TLS or HTTPS which are common security protocols used on the web. In my opinion this will drastically limit the amount of e-commerce traffic that is deployed with the Google App Engine. And Lastly, while the article stresses many more points, it currently only supports development in the Python Development Language. Google does have plans to expand to other languages but has not done so to date. One additional concern has been raised about Intellectual Property of these applications and if publishing these sites is akin to “Socializing IP”. So who does own the code and data within the application…. Google or You?

1. Google Under Pressure As App Engine Requests Rise
http://www.cio.com/article/424864/Google_Under_Pressure_As_App_Engine_Requests_Rise/1

The article below details some problems identified with the Amazon EC2 platform related to the inappropriate use of customer data. Apparently spammers were using the EC2 platform to create malicious applications and spamming logic. The EC2 logic allows these solution to simply “hop” to another IP to avoid being shut down. As a result the EC2 space is now being actively blocked by Outblaze and Spamhaus, two market leaders in secure enterprise level email services. This represents a huge problem for the EC2 platform and will require Amazon to create a “trusted” partnership for programs running within its environment. This in turn would make it more difficult to create a new web solution and deploy it quickly. Amazon may now require code scans, system monitoring, and pilot time periods before your application is promoted to an “unblocked” range of its Domain. One current users expressed his concerns with Amazon’s lack of response to situation like this:

‘I got the impression the only thing Amazon considers abuse is use of their servers and not paying the bill. If you’re a paying customer, you can do whatever you like.’

Read the entire article and discussion below.

2. http://taint.org/2008/07/02/162007a.html
Amazon EC2’s Spam and Malware Problems

Below is a great article addressing the concerns regarding the cloud from the CIO's perspective.

3. The Dangers of Cloud Computing
http://www.cio.com/article/426214/The_Dangers_of_Cloud_Computing


CONSUMER AND ENTERPRISE VIEW

I believe that most large organization will begin to use ITO activities and allow their chosen vendor to implement their own version of the cloud. Nowhere, is this more apparent then in the Government. Silo'ed departments and offices may soon be mandated to utilize the cloud that has been built for their entire agency. This could have a significant impact on the amount of control that the CIO has. However, using Google or Amazon's publicly available cloud is not an option for most companies and government offices. There are too many concerns about security and vendor tie in.

4. 10 Reasons Enterprises Aren’t Ready to Trust the Cloud
http://gigaom.com/2008/07/01/10-reasons-enterprises-arent-ready-to-trust-the-cloud/

The end consumer will also be VERY SLOW to adopt the cloud. (See article below)

5. Are you ready to subscribe to Microsoft Office?
http://www.tgdaily.com/content/view/38221/140/

Microsoft is carefully targeting the introduction of an Online Microsoft Office subscription. It is tentatively being priced at $69.99 a year and bundles Email, Calendar, Virus and Computer Maintenance with Office 2007. In a consumer industry used to owning their software this would be a major leap for consumers. In addition, there are entire web companies whose revenue model is based on FREE RIA based office products. (See below)

6. What's Holding OpenOffice Back?
http://www.linuxinsider.com/rsstory/63684.html

So to sum up my points. Cloud computing presents an exciting way for IT Infrastructure to become increasingly more efficient and agile in order to support the business of the enterprise. In addition, System Administration procedures and management controls can become more widely adopted to provide more secure and available software. HOWEVER, contrary to other's opinions, all applications will not run in the cloud in the next 10-15 years. A) Organizations with concerns over information privacy and differentiation will build their own "grid computing" models. B) Some complex applications that utilize extensive data sets have no business being in the cloud because of bandwidth and scalability concerns. AND C) Consumer trends are very difficult to change. Subscription based models for software people are used to owning will require time to penetrate the market place.

Myth #1. “Eventually Everything Will Run in the Cloud” – This statement is absolutely untrue. While enhancements and uses of to Internet Development Technology (such as AJAX, JSON, and Ruby) have expanded to allow for the creation of Client Style Applications or Rich Internet Applications (RIAs), there are a number of applications that due to security and bandwidth efficiency make more sense being run on the client machine. In addition, the Google or Amazon Cloud creates a single point of failure between the client and providing solutions. There are still a number of emergency response and disaster situations that will require the client pc to be capable of running independently of the server.

Myth #2: “The Cloud is More Secure” – As mentioned above the cloud creates significant efficiencies for operations and management of computing resources but can also create a single point of failure. As we’ve seen with new roll-outs of Windows, IPhone OS, and new Firmware, security flaws are inherently found after the fact. If the Google Cloud was running 90% of all E-Commerce sites and a vulnerability was inadvertently deployed they a hacker could gain access to a larger amount of data than ever before. While I agree that this risk exists today with e-commerce sites it exists in “silos”. The separate servers and applications create an environment where it is nearly impossible to access a wide collection of information in one hacking effort.

Myth #3: “All Computers in the Future will Run Applications Directly Through the Internet Browser” – While I agree that the Internet browser will continue to dominate the user interface development for a vast majority of client applications, I still think that Client Applications will exists. Java, .NET, and other languages that allow you to create a client application can offer three things that the cloud can not. First, is the ability to function without a connection. The ability to type a word document, edit a picture, or listen to music while my Comcast connection goes down for the 4th time that day is invaluable. Second, client applications can also offer the developer the ability to utilize system resources and memory more efficiently than most browser platforms. As javascript and RIAs become more prominent it may become difficult to ensure the machine is capable of withstanding the needs of the application, specifically when working with complex and sensitive data sets. For example, when analyzing market research data and comparing it with the latest census data, it would seem more appropriate to me that these actions be performed by the client than using up unnecessary bandwidth do an ad-hoc data mining function. Lastly, when attempting to use proprietary plug-ins and visualization techniques (think Adobe Shockwave) a client installation may be required to allow the application to run. The use of a proprietary or new software processing logic remains a key factor in the ability to differentiate your product offering, for video, picture, and media editing especially. Allowing these codecs and development frameworks to become open is contrary to the needs of a sound

CEO Guide to Technology

Found a great website from BusinessWeek.com that has published a CEO Guide to Technology.

http://www.businessweek.com/technology/ceo_guide/index.html

I won't comment too much on this site other than to say I know a lot of people who would find this useful. Everything from SecondLife, Wikis, Web 2.0, Mashups, Mapping, etc. All boiled down into somewhat easy to understand concepts.

Now taking these ideas and impacting your customer and marketplace. That is hard.

New Proposed Rules Governing Intellectual Property

The Senate has proposed new set of rules towards increasing damages to IP infringement. While we have done some extensive research into IP as it relates to software in technology, it certainly seems like something that has to be experienced to understand the intricate details and nuances of the legal processes. It is important for a company to have a strong in-house team or partnership with a legal provider in order to protect its interests. Companies like IBM have mastered this art over time and Microsoft, Apple, and Google are right behind them. These new laws will only increase the importance of IP protection.

http://blog.wired.com/27bstroke6/2008/07/senate-introduc.html

User Experience is Paramount

Ok, its been a while since I've written a post. Its been a busy month at work and school. However, I've just gone through my Google Reader account and came up with a ton a great articles that I wanted to share with you.

I've always been a big fan of "usability". While the term sounds simple, creating a powerful application for technology requires the use of a number of different techniques. In my opinion this all starts with strong executive sponsorship for an Agile Development Methodology.

I've found two articles that highlight the business value of having usability as a key design component of any type of produce. We're starting to see the promise of connecting the "problem space" and "solution space" using technology.

First the article "Design Up Front for HP" describes their approach to "The Computer is Personal Again". One could argue that they are the closest "apple-like" hardware manufacturer in the market today.

http://www.cio.com/article/439232/Design_Up_Front_for_HP

The second article references a recent memo sent by Steve Ballmer the CEO of Microsoft. He outlines Microsoft's approach towards conquering Google and Apple by providing a more "apple-like" user experience. He recognizes that these vendors are thriving "Because they are good at providing an experience that is narrow but complete, while our commitment to choice often comes with some compromises to the end-to-end experience." He brings up some great points. While Microsoft has been capable of creating a variety of complex solutions into its products (Word, Visio, Excel, etc) the user experience has often suffered from these variations. I believe that a free Online Office version would give Microsoft the ability to demonstrate/compete with both of these vendors. While SaaS (Software as a Service) has a number of concerns, I do not believe that Microsoft is agile enough to make the changes necessary to compete. I would expect for the next Microsoft Mobile device to be more like the current Iphone.

http://www.appleinsider.com/articles/08/07/24/microsoft_sets_sights_on_providing_an_apple_like_experience.html