SOA Trends – Minus the Hype

Why has SOA risen like a phoenix from the ashes after it was declared as dead by Burton Group (now Gartner) analyst Anne Thomas Manes? Is it because the end of slowdown is in sight or is there some other reason?
Recent statistics from Forrester says that the adoption of SOA is strong even during the slowdown. That means that SOA adoption never stopped. Probably there is a difference in perception on what SOA is between the Analysts and the people down in the trenches.

SOA started its life as an architectural technique for achieving loose coupling but within a couple of years became a business initiative to create reusable services mirroring the real world business. However, in reality, it primarily gets used to build multi-channel applications where common business logic is encapsulated inside services.

Now coming back to question 3 & 4 – [You need to read this post in conjunction with my earlier post where I had raised 4 questions and answered 2 of them].

3. If the current trend continues then where will it be in one year time?

  • SOA will become the standard principle for multi-channel application design.
  • However there will be a realization that services designed for multi-channel application will not be suitable for hybrid cloud.
  • A new terminology may get coined for SOA.

4. What happens if you take no action on the specific technology for next one year?

If you don’t have to service multiple channels then don’t bother with redoing applications the SOA way. However, if you are building new applications makes sense to attempt loose coupling through use of services.

Though it is not mandatory to purchase SOA products, if you are interested in updating yourself on who the SOA product providers are have a look at this report from Gartner.

SOA – the Phoenix act

In 5th January 2009, Anne Thomas Manes declared that “SOA is Dead; Long Live Services”. The original post is no longer accessible but the Google cached copy can be accessed. Just before this Gartner had announced that the Number of Organizations Planning to Adopt SOA for the First Time Is Falling Dramatically.

Prior to this SOA has been moving along nicely on the Gartner hype cycle for emerging technologies (though it had mysteriously gone missing on 2006). However, it has again gone missing from the 2010 hype cycle.

In the recent survey Forrester claimed that Adoption Of SOA: Still Strong, Even In Hard Times.

Even Gartner recently has gone on to declare that …SOA becomes like electricity. It’s just there

What is SOA?

Is SOA about …

  1. …design of the services – which mirror real-world business activities, or
  2. …achieving loose coupling among interacting software agents

The Open Group says that it is the former – The SOA Work Group : Definition of SOA. It is the reflection of the thinking among the experts around 2007.

But, if you roll back to around 2003 you would have selected the later – What Is Service-Oriented Architecture. Even, Gartner had published a white paper which leans towards the second definition – Predicts 2003: SOA Is Changing Software.

What do you think?

Advertisements

Is Facebook at the Peak of its Hype

The record of Facebook so far is impressive … in fact it is unprecedented. Facebook with $40 billion valuation & 500 million users – Mark Elliot Zuckerberg, the youngest billionaire & Time “person of the year” … the list goes on. So, where does Facebook go from here? Are we looking at the tip of the iceberg or are we standing at the top of Mt Everest?

Here is a 2007 May Fast Company article – Facebook’s Mark Zuckerberg: Hacker. Dropout. CEO. It makes interesting reading. The article goes on to says “…TechCrunch posted documents said to be a part of an internal valuation of Facebook by Yahoo. The documents projected that Facebook would generate $969 million in revenue, with 48 million users, by 2010…”. They have been wrong by an order of magnitude on the number of users!!!

Going forward, here are five possible scenarios for Facebook.

  1. Next big thing happens: A new and innovative entertainment channel opens up and becomes very popular. Though Facebook continues to enjoy a huge user base, total time spend by users on Facebook saturates and starts declining.
  2. Remains a social networking site: Their user base grows to about 1 billion and also gets a fair share of advertisement revenue. It remains, by far, the largest social networking sites but fails to expand beyond that. People continue to spend time on Facebook to socialize and play games but not for anything else.
  3. Encroaches into others territory: It builds up a loyal user base that uses Facebook of almost everything. It encroaches into the territory of Craigslist, eBay, Amazon, YouTube … and creates its niche. Significant portion of SMB completely depend of Facebook of all customer interactions.
  4. Seriously threatens Google and others: Proves that knowing people is vastly superior to understanding web sites. Business of all sizes starts to flock around Facebook for advertisement, sales and customer relation management. All internet segments are affected. Google sees a significant decline in ad revenue.
  5. Achieves complete dominance of the web: Facebook impacts every facet of our life and becomes synonymous to web. Majority of our web interaction happens through or with the aid of Facebook.

Can Facebook really reach level 3 and move beyond?

Here are some of the challenges that need to be dealt with.

  • Currently, Facebook does not recognizes the nuances in relationship (see this great slide show by Paul Adams who is slated to join Facebook from Google) – question is can this nuance be introduced without sacrificing the simplicity of interface?
  • Facebook is blocked in most large enterprises – and wining over enterprises is always a big challenge – Google is learning it the hard way. Even in SMB survey indicates that half of them don’t allow access to Facebook.
  • So far Mark Zuckerberg has unilaterally defined the rules of the game. Not the best way to take the rest of the world on – it requires alliance – it requires collective thinking. Tim Berners-Lee, the creator of the web standard has called Facebook a Walled Garden in Scientific America (here is the original article)
  • Knowing people and behavior is important but not everything. The importance of data – information – knowledge cannot be ignored and today in the area Facebook depends totally on Microsoft. Here is an interesting analysis – What Social Signals Do Google & Bing Really Count?
  • Privacy concerns – monopoly concerns – government concerns … these are real concerns and more voices will be raised as and when Facebook spreads its tentacle. This post (Facebook’s Anti-Privacy Monopoly) summarizes the current concerns of people.

Finally, here are some thoughts on where social web is going:

Web 2.0 Trends – Minus the Hype

Web 2.0? What is Web 2.0 … Oh you mean Social Media!

Yes, like a Chameleon, Web 2.0 has changed its color to become Social Media. But, what does this change of terminology imply? Web 2.0 has forked into 2 distinct streams.

  1. SOCIAL Media = Leveraging Collective Intelligence: The first fork, the original promise of Web 2.0, is about harnessing collective intelligence internally and externally. Typically this initiative is driven by IT, HR, Customer Support group or by a Knowledge Management group if it is in existence.
  2. Social MEDIA = Grabbing Attention: The second fork is about using Social Media channels for advertisement, brand building, product promotion, new product launch, campaign management etc. These initiatives are invariably driven by marketing department. The budget comes by cannibalizing other media channels.

The dynamics – the decision making – the budgeting process – the measurement criteria – the sponsors for the two forks are entirely different. Analysts have not yet started looking at the two forks separately. Look at the reports from McKinsey (The Rise of the Networked Enterprise), IDC (Predictions 2011: Welcome to the New Mainstream) & Gartner (Top 10 Strategic Technologies for 2010).

Now coming back to question 3 & 4 – [You need to read this post in conjunction with my earlier post where I had raised 4 questions and answered 2 of them].

3. If the current trend continues then where will it be in one year time?

  • By the end of next year this fork should be more visible to everybody.
  • Most organizations would have used some form web 2.0 technologies internally – would also have reported qualitative benefit – however, how to calculate quantitative benefit for internal deployment will remain a challenge.
  • Most portal platforms and knowledge management tool will incorporate Web 2.0 technologies (blog, wiki …).
  • Social MEDIA will dominate.
  • There will be a realization that the process of leveraging collective intelligence is not repeatable.

4. What happens if you take no action on the specific technology for next one year?

Sky will not fall on your head unless …

…you run a business where majority of your customer spends lot of time on Facebook.

Interesting statistics on Business Social Media Usage by Business.com

Let me give some of the highlights of the report (You can access the full report here or here).

Who is in-charge of the Social Media Initiative?

 

(Page 16)

  • Marketing = 66%
  • Customer Support = 23%
  • Others = 11%

On what measurable parameters is the impact of Social Media initiatives clearly visible?

 

(Page 19)

Maximum impact on: Brand awareness / reputation

 

Minimum impact on: Site traffic, useful product feedback

 

Will Web 2.0 work inside an enterprise?

I had written a post about one and a half year back (Why Web 2.0 will not work inside Enterprise) where I had identified 5 reasons. I was hoping that I would be proved wrong. However, available data tends to validate my fear.

McKinsey Report

First let us look at the report published by McKinsey (The Rise of the Networked Enterprise). Though it paints a very rosy picture of adoption and self reported benefit, one piece of hard data shows a slightly different picture. Have a look at their “Exhibit 3” on page 6. The correlation coefficient can be between 1 and -1 (1=highly related, 0=no relation & -1=inversely related).

What can we conclude from this exhibit?

  1. If you are networked with the external world then you are likely to gain market share.
  2. Using these tools internally may not give any significant or measurable financial benefit.
  3. Being too open can negatively impact your market leadership position. (this is the most disturbing part of the report which Mckinsey chose to gloss over)

List of 67 case studies that prove social media ROI

This list is compiled by Rob Petersen, founder of BarnRaisers in 2 parts (Part 1 of 34 cases & Part 2 of 33 cases). The interesting or disturbing part of this list is:

  1. Of the 67 cases only 2 (IBM & EMC) pertains to successful internal implementation of social media.
  2. Majority of the case studies (40) are purely marketing & sales initiative – it is about product promotion, brand building, product campaign etc.
  3. Most of the balance cases are about community building and leveraging the community.

 

Mobile Computing Trend – Minus the Hype

Yes, mobile computing is moving like an elephant and mowing down whatever comes in its path and there are many statistical indicators to support that hypothesis – as a result mobile channel is becoming must have rather than nice to have option. Hidden among this growth story is also a fundamental shift that is in progress which has gone under appreciated – the changed technology landscape has significantly simplified mobile application development and maintenance.

In Short – being present in mobile channel is becoming mandatory and affordable.

Now coming to question 3 & 4 – [You need to read this post in conjunction with my earlier post where I had raised 4 questions and answered 2 of them].

3. If the current trend continues then where will it be in one year time?

Android will continue to increase its market share

Also: Read this very interesting post on Darwinism Triumphs Over Intelligent Design by Jason Perlow.

Mobile browsing to account for more than 5% of all browsing.
 

2011 will be the year of tablet.
 

The market right now is dominated by iPad but in the coming year will see many alternatives Android based, Windows based and even Chrome based. All the analysts agree that the sales will boom:

Also: See the graphical representation of the prediction

HTML5 

  • Offline working
  • Local storage
  • Access to Geo Location
  • Canvas & Video
There will be no opposition to HTML5 – most Smartphone will support it

There are not many things where Apple, Google & Microsoft agree on. HTML5 is one of them!

However: Here is a different view point from Forrester (paid report).

4. What happens if you take no action on the specific technology for next one year?

You can ignore the mobile platform if you …

…are not into B2C business

…do not see and significant growth in your web site access from mobile browser

…do not have any field staff that comes into regular contact with customer

Otherwise, you better start preparing plan for your mobile web site – it can be an extension to your existing web site.

How mobile development is becoming affordable

Traditionally, mobile platform fragmentation was considered the biggest challenge for enterprise to invest in building a mobile channel for customers. Not only where the OS and the programming platform different, even the support for HTML was not uniform. Some handsets did not support HTML and you needed to create the pages in other markup languages like XML. So, it was necessary to use some sort of a translator which where you would create the page definition using a special markup language and the tool will translate it to handset specific markup.

Off course, if you application needed to access some of the features of the phone you had no choice but to write a native installable application. To get reasonable market coverage you needed to write the application in almost half-a-dozen platform. J2ME was an option but it could not utilize full features of the phone.

Year 2010- all that has changed. Three things have happened.

  1. Not only are smart phones are slowly becoming ubiquitous – statistics prove that access to web sites from smart phone is much more frequent compared to its actual market share
  2. Majority of the Smartphone browsers now have their core engine as Webkit – not only iPhone, Android but also Blackberry (OS 6) & Symbian (S60 onwards)
  3. Adoption of HTML5 by Apple, Google and even Microsoft paving the way for mobile web application which will be able to use handset features (like GPS location), use local storage, work offline etc.

Therefore, for most applications, it is sufficient to create a mobile web site using a tool like GWT and cover majority of the users. For an enterprise, this means creating a mobile channel is becoming both imperative and cost effective.

Cloud Computing Trend – Minus the Hype

If you subtract the hype, cloud computing is moving as a tortoise – slow and steady, more slow than steady. Will it win the race? May be; but not next year. The question is where we will be one year down the line.

[You need to read this post in conjunction with my earlier post where I had raised 4 questions and answered 2 of them. Here is the answer to the question 3 & 4]

3. If the current trend continues then where will it be in one year time?

  • Basic premise of (1) economy of scale, (2) pay what you use and (3) better utilization through sharing will remain intact – though some reports challenging the extent of cost saving will emerge.
  • Amazon will extend its lead over others with the most comprehensive offering on IaaS – competitors will try to carve out their own niche.
  • Google will not make much headway in the enterprise segment – perpetual beta does not gel with enterprise.
  • Microsoft will do just enough on office suite to keep competition at bay – but not too much to cannibalize its core office business.
  • Same will happen with major ERP vendors – they will make just enough noise but stop short on cannibalizing their core business.
  • Every vendor will look for a pie in the private & hybrid cloud – but the actual adoption will be very low the talk will shift to governance being the key.
  • Critical concerns (both real and perceived) like (1) security, (2) privacy, (3) SLA and (4) compliance also remain – like credit card usage on net objections will slowly go away – but tipping point will not be 2011.

Established software vendors find themselves at the crossroad – not sure if they want to welcome cloud or shun it. This is what Forrester has to say – “…the vendors realize that license revenue models are no longer a reliable engine of growth, particularly in a mature segment such as ERP. The enterprise applications vendors, therefore, must evolve to recurring revenue models and do so fairly quickly (three to five years) …” [Reading between the lines – fairly quick (3 to 5 years) = nothing next year & recurring revenue models = as a user you pay more in the long run] (see this & this)

What is the chance of a “Black Swan” appearing in 2011? I think it is low. In 2012 or 2013 – may be.

4. What happens if you take no action on the specific technology for next one year?

You can happily ignore the cloud unless you …

  1. …have a compute intensive application needs to run once in a while – cost saving potentials are great (unless you have to upload/download large amount of data)
  2. …are an internet startup – cloud will save you from the dilemma of how much to invest on hardware
  3. …are planning to revamp your CRM solution – after all this is one of the trigger point of the cloud hype
  4. …expect large number of tablets (iPad and others) getting used in your organization – you will need an alternate to Microsoft Office
  5. …are considering a new email solution – there are published reports showing cost effectiveness of cloud over in premise
  6. …need to overhaul you data centre – virtualization and related management technologies can improve utilization

However, there is no harm in experimenting with different technology options available.

Here are some more details on specific categories of cloud which may interest you:

Category Interesting tidbits
Email For hosted email solution, here is an analysis from Forrester which claims that email in the cloud is cheaper not only for small but also for midsized companies. Additionally, Google offering comes out much cheaper than the others. This is a good report to read if you are contemplating new email solution. 

What it does not talk about is when it will make sense to chuck your on-premise email for cloud based one.

CRM Most of the adoptions in this category are on and around CRM solutions. Everybody knows that Salesforce.com has been the leader in this category. There are more than 300 case studies listed in their side but most are around CRM solutions. Most users while evaluating a CRM solution will definitely look at the hosted offerings. 

Have a look at the HP case study which claims 40% saving.

ERP SAP had launched “Business by Design” on 2007. However, if you go through the case studies (Select Category = “360 Degree Customer Video” & Solutions = “SAP Business ByDesign”) you will find on eight of them. 

There are others who offer hosted ERP solutions. Here is a list of eight others. Out of these, only NetSuite and Plex Online talk about the hosted solution on the home page. Also, here is an offering from India: “Ramco – OnDemand ERP”. They have eleven case studies listed on the site.

If you are planning to evaluate new ERP solutions see if you can defer the decision by one year – situation may be different.

Office Suit Though Google launched Google Apps in February, 2007 – it has not made much headway (see this). Same goes with the other interesting hosted office suite from Zoho. The main thrust of Google has been on the ability to collaborative editing of documents. Unfortunately, there are not many scenario where that becomes a killer feature. 

On the other hand Microsoft has recently released Office 365 (see this review & this) which is quiet well received. This game is for Microsoft to lose. So far, they have been playing it right – releasing just enough features in the cloud to be ahead of competitor – but not too much to cannibalizing its existing business. (this will through more light on their strategy)

2011 may be the year of tablets with iPad and many other competitors available in the market. Not many of them will run windows hence will not be able to run Microsoft Office. This will provide a fillip to cloud based office offerings and people will seriously start looking at them.

PaaS There are 3 major platforms but none of them have taken off. Each will cater to its own niche and limitation. 

  • GAE for creating APIs and Add-ons,
  • Force.com for customizing Salesforce.com and
  • Azure for extending existing .Net applications.

You can look at them only to supplement any other move to cloud.

Azure has close to 50 case studies, Force.com one example of bespoke ERP solution.

VMWare is attempting to standardize Java for cloud (see this). Heroku may be the dark horse with a platform for Ruby.

Though I am a fan of Google I cannot help but refer to the post where one user has recorded all his cribs (here is the summary). He has been advised that he should have gone through the well documented limitations. [BTW: If you are looking for workarounds – have a look at this]

IaaS Amazon, by far, has the most comprehensive and usable offering – in terms of ranges of CPU size, types of storage etc. Majority of the cloud tool makers ensure compatibility with AWS. 

Contrary to popular perception, Microsoft Azure can also be used just as a windows machine running any software that runs on Hyper-V. However, their niche seems to be tight integration with Visual Studio, especially for .Net applications and developer productivity. (here is an interesting deal)

Rackspace is trying to drive adoption by open sourcing their cloud platform so that organizations can make use of that platform for private cloud.

Private Cloud There seem to be enough evidence to show that virtualization can improve server utilization. However, managing a large virtual infrastructure can become very complex. Here is a post from CA – the last para talks their products but rest of the post gives a good outline of the challenges involved. 

But, have you ever wondered why this class of technologies gets grouped under Cloud Computing? After all, this technology does not meet 2 of the 3 requirements of cloud computing (economy of scale & pay as you use). It is only about better utilization through sharing.