Enterprise Architecture: From Incite comes Insight...

Today, I visited Bernies because they had a special sale for employees of large enterprises where we could buy anything with an additional 12% discount. I wonder if this model could also apply to software companies?



For books, The Pearson Group allows employees of large enterprises to buy books cheaper than found on Amazon. Sears, stays open a couple of times throughout the year and gives an additional 10% discount and even many of the local retailers provide various discounts in order to attract large groups.

The best discount though absolutely has to be Bob Stores where they give a huge 30% off all purchases three times a year to folks employed by large enterprises.

I wonder what would happen if say Symantec, Dell, Apple and Compusa where to give 20% discounts off current pricing to those employed by enterprises with more than 1,000 people? Would they gain more sales opportunities above and beyond the cost of the discounts?

Why aren't industry analysts providing creative suggestions to consumer-oriented software companies and how they can attract larger demographics?

Awhile back, I posted a methodology for industry analysts to bring integrity in determining truly what is the most used portal. Check out this deep analysis and comparison of a variety of enterprise portals...

Today, I took my son to register for Kindergarten in the fall. He will be attending the same school that I did thirty years ago...



It is kinda weird to step foot into a place where you haven't been in a long time. I remember the school being a lot bigger than it actually is. The very first room I entered was the school gym in which I have vivid memories of. I used to be pretty good at climbing ropes and noted that they were no longer there. I am not sure if I remember the gym because this is my first memory of competition in that there was only one other kid that was faster up the ropes than me in that he could climb without using his legs or was it the fact that one day I descended the rope to fast and broke a toe and was the first time I had an x-ray?

Anyway, my son preceeded down the hallway introducing himself to all the teachers who passed by. Upon entering the school's office, I noticed a familar face and I introduced myself. I was esctatic to see that the same gym teacher I had in school was still there (Hi Mr Everin). Other than a little bit of gray, he looks exactly the same.

I have heard horror stories from other coworkers and the stuff not taught in school nowadays especially in gym class. One coworker who is a Marine and wrestling coach at a high school in another town mentioned that their school system, kids don't do dodgeball or any other event where there is the potential outcome for a kid to feel like a loser. Instead they do inane events that allows all kids to participate and everyone can feel like a winner. I wish corporate America worked like this but it doesn't.

Anyway, seeing an old school gym teacher made me feel at ease. He told me that while they remodeled the gym, they still had the ropes and that they were simply placed on the stage. He keeps it real....

Nowadays, many of America's youth, especially those who immigrate from countries such as India don't know how to get busy. American kids are not only losing advantage when it comes to intellectual pursuits but also physical ones. My son, has mastered the Iron Fist technique and his younger brother has a pretty powerful Club Foot, so they know how to open can's of whoop ass for Bin Laden or anyone else that gets it twisted and set trips in America.

On first impression, it seems as if the school still has its act together. My wife and I debated the merits of attending a public school over choosing a private one like so many of my peers have. I have figured out that learning is neither reduced nor accelerated solely based on the credentials of the person teaching but on the student's desire to consume. My son will be savage in learning, whooping ass and most importantly interacting with folks who keep it real.

Next month, they will be having a special event for the parents to meet the teachers. I hope to share what I learn in this regard but got the sense he is on his way to somewhere where I could have only dreamed of when I was coming up...

Nick Malik has a wonderful post entitled: Help wanted: who pays to simplify your IT portfolio? and acknowledges that entropy creates IT portfolios. I recently read an article on Unilever where Accenture won a large-scale deal to convert over a thousand disparate IT systems down into one single instance of SAP. It would be interesting for any Accenture bloggers to provide insight into Nick's questioning as to what they believe the right answer to be?

The one thing that troubles me with how EA is currently practiced by the masses is the lack of acknowledgement of the architecture pendulum. Over time, we will go from glass house centralization to highly distributed back to vendor consolidation and then to distributed service-oriented architectures and back to some other paradigm. While it is IT's job to justify their existence, don't business folks who pay for our salaries simply acknowledge this as the pendulum effect? Why would they even participate or more importantly make long term committments to anything?

A little over thirty days ago, I made a challenge to the Ruby community in this blog entry and have finally been proven wrong. Now it is time to put my money where my mouth is...



Before I post facts, I wanted to thank the person who has taken the most adversarial position against my perspectives on Ruby which is David Heinemeier Hansson and his thoughtful response to a posting where I asked him challenging questions. The responses from James Robertson of Smalltalk fame have been equally enlightening.

I promised in an earlier blog to donate to charity. It is important though for me to explain what the real definition of charity is as most people don't practice it. I am firm in my belief that there is a distinction between charity and acts of giving money that are obligatory. Whether you are Catholic and believe in Tithing or are of the Islamic faith and give Zakat, this is not charity but a commandment from our creator and therefore is obligatory.

My family has a history of Diabetes, so giving to a "charity" that supports finding cures is not really charitable as I stand to benefit from the donation. Charity really should be about giving to causes in which the individual or their family members doesn't personally benefit and even should consider demographics outside of the one in which you belong. Imagine how much the world would improve if say folks of the Jewish faith donated to charities in Palestine or immigrants to the United States from India who now have high paying jobs were to donate to local homeless shelters and folks who aren't so lucky?



Since no one in the Ruby community ever indicated a preference as to once they prove me wrong, what charity they wanted me to contribute to, I have chosen two. I would have normally chosen a charity such as the American Family Association and their recent campaign to Boycott Ford but figured in this situation, I should do the exact opposite. For David, I will be making a contribution in his name to the Yes Institute. For James Robertson, I will contribute to Children of the Night. Hopefully, they will find these charities suitable.

Anyway, for the record, the Ruby Community actually didn't find a single fact to prove me wrong but at least it generated a lot of discussion. Hopefully, new insights will emerge and that they will consider alternative perspectives going forward...

How many things can IT align with at one time? The buzzword of the day is IT/Business alignment in which no one has yet developed any form of litmus test to know when the enterprise has arrived. Hype is the plague on the house of software...



If I am to align with the business, how does aligning with the principles of ITIL, COBIT, CMM help or hinder? Of course those who are busy making a career for themselves by studying these principles will have a canned partyline that can provide rationalization as to their efforts but many of us may find all of them suspect. The one thing that I haven't heard though in all of the industry rags on this topic is the notion of stewardship.

Most people in IT work in one of two roles. Either they are involved in the day-to-day delivery of IT services (more than likely operations-oriented). Or they are involved in IT strategy to support the strategic intent and direction of the business.

IT falls down on a pretty consistent basis in the chasm between these two roles. Some observations that I have seen are the lack of a chief security architect whose role is to ensure that IT systems are built in a compliance-oriented way. We understand that business provides business requirements and that IT architects think about system qualities but the notion of compliance really isn't either but it matters.

For example, when IT executives forget about the notion of a chief security architect, they find technical fixes difficult and the management of them ever more difficult. Compliance is not a day-to-day operations problem nor is it about strategies the business wants to undertake. Many enterprises fail at addressing this because they fall outside our normal processes for justifying and managing IT work.

Enterprise decision making in this camp is primarily based on a leap of faith which can lead to the wrong decisions. Even more sinister is the widely held belief that knowledge transfer is the problem in that one can distill down compliance principles into a one-pager to solve problems such as these. Maybe us enterprisey folk can start talking about how humans learn and how we have consistently failed in aligning humans with process!

Governance won't work, CMM, ITIL, etc won't work but maybe strong technical leadership and stewardship will...

Stephen O'Grady of Redmonk had hinted that productivity in large enterprises can also occur via the usage of business rules engines. I would like to extend his thought and ask the Ruby community whether there is merit in embedding the RETE algorithm into Ruby?



For those who are unfamiliar with the RETE algorithm, this blog does a wonderful job of explaining it...

Today's blog entry doesn't have a central theme and only contains random thoughts that I have had this week on IT oriented subjects...

• Too many folks are getting the notion of innovation twisted. Innovation is not really about technology but about societal issues that are transformational and impediments to it. I wish other bloggers could talk in these terms.


• Been thinking about GNU/Linux and its recent rapid adoption within the enterprise. Likewise have been thinking about OpenSolaris. Industry analysts aren't recommending one over the other. Curious in what situations they would?


• I wonder if there is a business opportunity for RedHat to also support OpenSolaris?


• Why does the marketplace believe that Microsoft is anti-open source? They are more community-oriented and provide more source code than any other vendor on the planet!


• Anyone know when Oracle will actually start embracing the open source movement? They seem to be the last big holdout


• VoIP is intriguing to me in large enterprises yet analysts again aren't doing their part. For example, if an enterprise were to consider Asterisk for call centers and regional offices, they could save tons of money. I have heard that a Dual Xeon can process 400 concurrent calls


• How come enterprise architects never seem to talk about the lack of strong technical leadership within their shops as the root cause to low morale, year-over-year budget increases and other IT ailments. We all know that the words management and leadership within our corridors is used interchangably but shouldn't, so what should we do about this?


• I wonder why David Heinemeier Hansson hasn't answered questions posted here. Is the Ruby community afraid to engage in an dialog with folks that may not 100% buy their value proposition?


• I wonder what it would take to get my favorite coworker blogger to discuss publicly what the five things he thinks enterprises do that are too enterprisey?


• I was thinking about emailing folks at Gartner to see if they would noodle an idea I had of changing the magic quadrant aways from listing vendors and instead to listing of products. That way, they could show 100% non-commercial open source products that provide benefit to large enterprises right next to closed-source commercial offerings. I wonder how successful I would be in convincing them? Maybe Larry at InformationWeek could discuss this concept in an upcoming article he is doing?


• What would happen if Jon Udell started his own analysis of the IT analyst industry and published it in an upcoming infoworld article? I bet that issue would sell more advertising for them.

One of my coworkers is always mentioning Clayton Christensen and he asked me to use my blog to help him find candidates for some positions he is having a hard time filling. Figured some background was necessary before I jumped into the details...



Clayton Christensen proposed that there are two major types of technologies, the usual sustaining technology and the less usual disruptive technology. A sustaining technology is basically one that makes doing-what-we're-already-doing cheaper, or better in some way that the existing market desires. A disruptive technology is one that is proportionally more expensive, doesn't work much better, and basically is not desired by the existing market. However, the disruptive technology also has other qualities that make it desirable in some way to a small minority market. Disruptive technologies become disruptive when they have been improved to the point where they now compete adequately, not with the existing technologies, but with what the market desires which is often considerably less than what the existing technology is capable of providing.

The vast majority of conversations with industry analysts with the exception of the folks at Redmonk tend to be about sustaining technology. Agile Manifesto and all the methodologies that adhere to it (FYI open source is an agile methodology) can be categorized as being disruptive according to Christensen as they displace traditional ceremonial hierarchical development methodologies.

Agile methods are disruptive in that they don't permit the illusion of management (distinct from leadership) control. The problem with agile methods is that the "customers" of agile are management and therefore the cost to adopt is higher. Management (distinct from leadership) in most but not all IT shops are currently getting it twisted by focusing on maximum control over deeply technical decisions in which they don't know anything about. Maybe the notion of strong technical leadership within enterprises will also become disruptive over the next several years...



I was going to put the HR description of the positions he was looking to fill here, but we all know how much insight that would provide...Here are characteristics of the right individual to fill his job openings:

• Wants the perceived safety of working for a large enterprise but never really had tolerance for all that enterprisey fluff.


• Has technologies on their resume other than the usual stuff found in enterprisey places such as Java, .NET, XML, J2EE, Mainframe COBOL, etc.


• Has a great ability to write and an even greater ability to present. The need to be published and present at conferences should be strong.


• Deep analytical capabilities, kinda like an industry analyst but actually do so in a credible way


• Desires flexible work arrangements and even the ability to sometimes work from home but won't take it so far as to desire to bring their dog to work or skateboard down the corridors.


• Wants to work for a boss that isn't an idiot and actually understands technology


• Wants to work for a boss that when he does something stupid, you can tell him to his face that he is a bonehead and he will actually appreciate it


• Wants it to be their choice as to whether they climb the ladder at a rapid pace or whether they are simply want to remain technical


• Doesn't mind periodic travel to cool universities to talk with professors doing meaningful research


• Get paid in full!


• Work with truly the best team of architects on the planet


• Is a US citizen as there is no visa sponsorship available

If you are interested or know of others, leave a comment...

Figured I would attempt to see things from the perspective of the outsider looking in...



Having spend more of my career working in small companies than large enterprises I still find it interesting how folks can read into what being enterprisey means. Anyway, here are some of the ways to tell if you are too enterprisey...

• Terms such as "division," "department," and "business unit" all have official meanings within your company, and all refer to different levels within the corporate structure.


• Each one of the above has their own separate logo


• It not only takes two weeks to plan for a one-hour task but you must also use the company-approved productivity tracking tool in order do so


• Your boss has more people over him (e.g. his boss, his bosses boss, etc) than under him.


• There is more than one employee with the title of "president."


• There is not only more than one CIO, but you run out of fingers counting them all.


• You must attend mandatory meetings in order to find out what you are quite capable of reading via email.


• You are required to change your password every few weeks even if this has been proved to decrease security.


• The folks on the support desk are measured on their ability to close trouble tickets than in solving problems (and it is more difficult to close a ticket than to solve a problem)


• You stamp everything "DRAFT" to cover your butt. NOTE: James McGovern doesn't do this and instead uses the phrase "for discussion purposes" which he observed by working with McKinsey consultants during the dot-com days and they were attached to a document labelled final specifications signoff.


• You need to spend more brain power trying to figure out how to explain sticky situations to managers than actually solving real problems.

What did I miss?

I was getting ready to address this topic when Nick Malik said it best here...

Figured I should provide a little amplification to this blog entry...

Was thinking about a trend that I haven't heard others talk about in a long time and was curious if anyone else is also observing the same thing...


At one time, use of lightweight directory-oriented products that leveraged the LDAP protocol was very popular for Internet applications. Nowadays, they seem to have taken a backseat and everyone is simply using relational databases. It seems as if the LDAP protocol is now stereotyped only for usage for lightweight identity stores and nothing more.

Industry analysts aren't even talking about this space and haven't produced "quadrants" in a long time. The guys over at Redmonk who provide wonderful insight into Sun products also never talk about Sun product offerings in this space. What's going on?



LDAP is highly optimized for read-intensive applications. An LDAP server that receives a request from an application takes responsibility for it. It may pass the request to other servers until it can be fulfilled, or it may refer the client to other servers, something which relational databases don't support until you start buying enterprise versions.

Replication of directory data is also required for high availability networks. Consider a directory supplying the authentication database for a business-critical Web application. If the directory is down, the Web application can't authenticate users, which means lost business. One way to reduce the risk is to make two or more copies of the directory data, each served by a separate machine. If one copy becomes inaccessible, the other can take over.

Microsoft, not only has built Active Directory and has exposed it via the LDAP protocol but also has released Windows Server 2003 Active Directory Application Mode (ADAM). For those in the open source community, there is also OpenLDAP.

Would love to gain insight as to why industry analysts are talking about this space and why more enterprises aren't considering as part of their architectural thinking? What am I missing?

I am known for expressing an opinion that goes against popular thought. Today is no different...



Stephen OGrady critiqued remarks made by Novell's President and COO Ron Hovsepians comments that attacks Sun's OpenSolaris project. I am of the belief folks have "read into" vs read what he actually said. The usual statements that he could have done a better job of articulating his position apply.

Anyway, figured I would take some of Stephen's comments and analyze them:

Given the costs associated with open sourcing software, this argument is not particularly supportable - the legal fees involved with open sourcing a piece of previously closed software alone would outweigh any potential "marketing" value.

I wonder what definition of open source is he using? Many of the larger analyst firms use a heavily weakened definition of open source to merely indicate that source code is available. We know that the folks at Redmonk hold themselves to a higher standard and aren't guilty of promoting ignorance (aka distillation) to the masses. The OSDL has a great definition of open source here.

If we were to use this definition and then compare GNU/Linux to OpenSolaris how would they align? For example, does Sun's license violate principle number 5: No discrimination against persons or groups? It seems to want to prevent sharing of ideas with folks in the Linux camp.

I remember an exchange awhile back on a posting as to my thinking on the LAMP stack. Maybe we could understand support for companies such as MySQL and EnterpriseDB and are they really compliant to the open source definition? Does the definition permit dual-licensing where some of the source is not available? As I understand EnterpriseDB doesn't make all of its code available either.

Would industry analysts have more integrity if they stopped recommending certain components in the LAMP stack and instead recommended not only enterprise-proven technology but also products that are 100% compliant to the definition? How come no one is talking about Ingres for example? Its the only DB that meets both requirements.

I am a big advocate of Liferay Enterprise Portal and will be shortly contributing XACML code to the base to further extend its leadership as the number one most secure portal on the planet. Do you think Liferay has in their licensing model and clauses that restricts say ExoPlatform from leveraging their original ideas or even borrowing code snippets? Of course not, because it is truly open. Would love to understand Stephen's thought as to whether it makes sense for Sun to become 100% compliant to the definition.

Don't tell me boring facts around how much Sun is spending in legal fees to convert to a new business model. Do tell me about how much they spend on contributing to open source projects in which they don't sell? For example, Sun has employees that contribute to the PostGres project but I had to learn this on my own. Likewise, tell me what the equivalent of this is in the Novell camp and let us downstream folks truly decide who is more open...

Figured I would spend time finding similarities in otherwise different stories...



Earlier in the day, I had the opportunity to speak to an industry analyst in the UK whom seemed to be incredibly insightful on the concerns that enterprise architects face. The funny thing is that I seem to always feel this way as most UK-based analysts tend not to talk about meaningless product/vendor babble like their US counterparts but instead focus on issues that actually matter. Would love to know why this is the case.

Awhile back, a coworker (Hi Missy) introduced me to a game affectionately labelled as "bullshit bingo" that she actually tested out without my knowing during one of the presentations I was giving. The intent of the game was to look for "phrases" that are often used by IT executives but have no longer any real meaning. Examples such as strategic, agile, synergies, getting ducks in order, etc all come to mind.

The industry analyst is focused on research related to IT/Business alignment and genuinely wanted to know my own thoughts on this topic. Nowadays, alignment has turned into capture somewhat questionable metrics and turning up a notch the notion of selling of ideas to non-technical audiences. The notion of distillation is starting to get popular in our EA vocabulary. In thinking about this problem-space I figured I would provide guidance not only to his research in my blog but teach other IT vendors who didn't get the memo you must stand on a soapbox and state the obvious with passion and that doing anything other than that like trying to focus on your core expertise is sometimes futile.

• Choose projects that provide the most business benefit!


• It is important to find an executive sponsor!


• It is cheaper to buy vs build!


• IT should be run like a business


• We need to do more with less

Vendors, don't actually worry whether your solution is best fit or even works at all. Say phrases such as these and many folks will start to fall in love with you. Of course, we understand that enterprises that have really achieved alignment aren't running around their hallways with cliche phrases and instead are focusing on what each individual truly has expertise in (Business folks focusing on the business and Architects focusing on technology) and both parties actually trust each other, but that will never be reality unless us Americans start reading the analysis of our UK counterparts.



During lunch I learned of a protest by the Falun Gong of which I have the utmost respect for during the speech of Chinese President Hu Jintao and thought to myself how this really correlates to IT/Business alignment. For example, if Kathy Gifford and I wanted to start an apparel business in one of the Central American countries using cheap labor there would be mass protest but if I wanted to hire cheap illegal labor here I would get a different result.

Many folks in the past attempted to fight China of its Most Favored Nation status citing political abuse, conflicts such as communism, and child labor laws but in the end none of this really mattered to the leadership as economics is king. Doesn't this feel kinda like enterprise architecture and IT/Business alignment?

Isn't IT/Business alignment really a funciton of governance? Isn't governance really about changing behavior models and not financial controls? Maybe several bloggers can start noodling what governance 2.0 looks like?

Came across an interesting book entitled: Workforce Crisis that provides case studies and "best practices" for undertaking workforce analysis...



The current shortage of IT graduates is getting worse in America. The perfect storm of baby boomer retirement and a deficit of young workers means the situation is poised to go from bad to worse unless enterprise architects start thinking on better ways to solve this problem.

Right now, enterprise architects who aren't capable of thinking but only in following have embraced the notion of outsourcing where the idea is to seize the opportunity to capture young folks in other countries. This approach is doomed to fail in the long run and is somewhat racist in its approach.

Thinking back to when I was 21 and all of the wonderful things I wanted to do, having the opportunity to choose between working for a global consulting firm travelling to see lots of different businesses, working for an Internet startup where I could bring my dog to work, working for a large software company such as Microsoft, or working for a large Wall Street firm where I could get started on a career path where I could become VP by 26 all seemed appealing. Many Fortune enterprises though have never really indexed themselves against the opportunities offered by others and only compare themselves to their verticals.

Is it really sustainable if your particular vertical can't hire locally yet also offers work that is analogous to watching paint dry to offshore workers? Sure, us enterprise architects have a duty to keep up morale and make things sound more attractive than they really are but do we also have a duty to index ourselves and the career opportunities provided against other verticals?



I think there is another answer that needs to be considered. I previously mentioned that several IT professionals from local companies after hours tutor local kids from the community on various computer skills. One student in particular (his name is Charles) is in tenth grade and knows Linux cold. He knows how to write C code, compile kernels, etc. The main problem though is that a tenth grader doesn't have a resume. I wonder what would happen if the HR folks didn't screen on resumes but instead simply allowed him to interview?

Taking this thought one step further, what if HR folks didn't sit on their butts waiting for resumes to cross their desk and instead actively recruited IT folks before they became hot. In our town, the Warhawks Football program is a dynasty and makes it to the State tornament pretty much every single year for the last fifteen. There are college scouts checking out our Freshman program. Why can't HR work in this fashion?

The enterprise that figures this out first wins...

Last week, at work several folks made interesting comments to me in passing about blogging which I would have otherwise ignored but figured there was an opportunity to learn other perspectives...



For the record, I never blog at work nor about work unless I specifically state so. Some folks will get it twisted and read into what I am saying instead of simply reading what I said.

One perspective that was shared that made me laugh is that the rationale for my blogging is so that I can send secret messages to others. For other EA bloggers out there, do your co-workers think that is the reason you blog? How do you manage this particular aspect?

Usually when I blog, it is early in the morning and I am sitting in my basement in my underwear. The thought of me thinking of any of my peers simply doesn't enter my mind. I really hope that they also aren't thinking about me whenever they are in the same setting.



Knowledge Crisis and I have the privelege of working together and have pretty open dialogs not only at work but in the blogosphere. I would like to think of myself as being pretty transparent with very little to hide. I have been known to share thoughts publicly even when they go against the grain of the community yet folks still sometimes think that I have some alternative agenda.

JT has told me on several occasions that I have the uncanny ability to torque a room full of people without expressing an intent to do so. Sometimes directness and cutting to the chase has this effect. Maybe I am guilty of not being "sensitive" to others thoughts.

I would love to hear from others that don't know me personally but have read my blog for an extended period of time, what they have "read into" my thoughts on any of the following areas:

• Politics: Am I democratic or republican


• Religion: What religion am I


• Agile: What do I believe to be wrong with current approaches


• Open Source: Why do I really support this notion


• Industry Analysts: Which ones do I like, which one's I don't like and why


• Work: Am I really blogging about work but saying I am not just to avoid bureacracy


• Conferences: Why do I value speaking at events

Any insights on the above or anything else that comes to mind you can provide to me on myself are greatly appreciated...

Every Monday and Wednesday night, folks from several Fortune enterprises volunteer their time to teach local inner-city children basic computer skills. I am currently teaching the fundamentals of relational databases and in the future I will be teaching Java. The one problem I have noticed is that learning is slower when these kids don't have the ability to practice what they have learned at home. I was thinking there are some industry analysts that can help me solve for this problem...



I posted a call to several newsgroups and many wonderful folks have stepped up to assist. Java Developers Journal Editor Yakov Fain has agreed to supply copies of his wonderful book: The Java Tutorial for the Real World which is targeted at younger folks. If you have kids, I would seriously picking up a copy for them.

The folks over at OReilly have also volunteered several books to assist in our effort. Likewise, Microsoft has been generous in providing copies of software for our usage as well.

So, we got instructors, books, software but no computers for these students to take home with them. Of course, all of us instructors can drop out of our desire to give back to the community and spend time writing grant proposals but this doesn't feel agile nor does it leverage our strengths.

I wonder what would happen if we simply asked industry analysts that happen to read this blog for them to ping their respective hardware vendors and see if they would be interested in making a donation, what would happen?

Was noodling about how the Wisdom of the crowds applied to the discipline of enterprise architecture...



What key criteria separate wise crowds from irrational ones?

• Diversity of opinion: Each person should have private information even if it's just an eccentric interpretation of the known facts.


• Independence: People's opinions aren't determined by the opinions of those around them.


• Decentralization: People are able to specialize and draw on local knowledge.


• Aggregation: Some mechanism exists for turning private judgments into a collective decision.

Lao-tzu declared that those who declared that those who justify themselves do not convince that to know truth one must get rid of knowledge and that nothing is more powerful and creative than emptiness - from which men shrink. Here, then, could one aim of enterprise architecture is to show backwards-fashion what must be destroyed.

Many folks attempt to codify acquired software development wisdom by describing a software development process (aka governance and reference architectures) kinda like how to bake a cake. Wisdom of the crowds may suggest that enterprise architects think of this as "backward" law. When you try to stay on the surface of the water, you sink; but when you try to sink you float. When you hold your breath you lose it - which immediately calls to mind and an ancient and much neglected saying, "Whosoever would save his soul shall lose it."

Came across an interesting blog entry that us enterprisey folks should think about...

Figured I would ask him seven questions in hopes that he would respond...



1. Isaac Gouy left an interesting comment in this blog stating that it is easier to ridicule someone that to show they are wrong on their own terms. This begs the question of when will an honest, open dialog occur between two communities who obviously have different perspectives? Hopefully you are still not stuck on a clerical error I made and admitted to by confusing in previous blog entry Ruby vs. other dynamic languages?

2. Would love to understand your perspective on productivity in the following context. As you may be aware, many large enterprises in the United States and Europe have been exploring outsourcing where they are taking highly productive local talent and eschewing them for less productive but cheaper folks in other countries. The rationale is that if you can get resources for twenty cents on the dollar then that beats any productivity gains otherwise afforded by local folks any day. Not saying that I believe this to be true, but would love your perspective.

3. Are you willing to publicly acknowledge that Java was not chosen by large companies for productivity (without getting it twisted that Java is highly unproductive) and that the real driver for its choice was that at the time, most enterprises were doing either mainframe development using COBOL or some flavor of Client/Server such as Powerbuilder and that they wanted a new capability to get on the web which their current languages couldn't provide? If you acknowledge publicly that there is at least some truth to the statement, are you willing to then acknowledge that Ruby may not be as successful?

4. Do you really believe all us enterprisey folk prefer large highly inefficient ceremonial insultingconsulting firms or that we are simply making the best out of the cards we are dealt? Why do you think we wouldn't really want to hire the best talent regardless of what firm they happen to be employed by?

5. There are other enterprisey folks in the blogosphere, in which each brings their own perspective. It would be useful for outsiders looking in to engage in a dialog where new insights emerge. Would you mind providing your own critique of the following bloggers in an upcoming blog entry:

• John Gotze on Successful Enterprise Architecture


• Charles Betz on A story of too many tools


• Pavi Agrawal on I want to be CEO


• John Wu on The Profession of Enterprise Architect


• Tom Olzak on Manage your outsourcing outcomes


• Luis Suarez on Is Web 2.0 enterprise-ready?


• John Wu on EA is all about seeing the big picture

6. Do you generally believe that it is true that for each 25% increase in problem complexity, there is a 100% increase in solution complexity. If so, complexity doesn't really start in IT but in the business models that IT targets?

7. Do you generally believe that 80% of software work is intellectual. Some is creative. Little is clerical. If so, what should us enterprisey folks focus on for opportunities to optimize?

Ran across this blog entry and felt it was worth a little amplification...

Us enterprisey folks tend to do stupid things on purpose in the name of governance, to get "buy-in" or some other Rationalization. I am of the belief that the Agile Manifesto provides a philosophy for avoiding stupidity...



On the way home yesterday, a peer of mines made an interesting comment regarding my strong beliefs in the benefits of being agile. Essentially, he recognized the fact that agile methods for software development could provide value to any large enterprise application effort by reducing complexity, increasing productivity, predicting higher quality and most importantly increasing morale. He also indicated a thought process that was right on the money.

Essentially, the vast majority of IT executives can't seem to get a consistent message from American developers as to how to meet the goals of the organization and therefore they have turned to their favorite insulting firms to provide them with the sage wisdom they seek. The vast majority of consulting firms that get the ear of IT executives tend to eschew agile approaches in favor of ceremonial CMM process-oriented maturity approaches. The one thing that has occured is that all of these consulting firms are consistent with their message.



I have always been of the belief that agile will stagnate unless it is allowed to grow beyond its original founding fathers. The vast majority of practitioners of Six Sigma and CMM more than likely couldn't even tell you much about the founding members and it has grown by leaps and bounds. Agile approaches are the only way that America and IT doesn't go down the toilet, so the community at large if they are interested in saving America must do their part to find out its root cause as to why it isn't being adopted faster especially when the value proposition is so obvious.

For agilists in the blogosphere, I would love for you folks to provide your own perspective on the following items:

• The founding members of the agile community have done a wonderful job at incubation. What group of individuals would make greats leaders to take it to the next level?


• The folks over at InformationWeek pinged me several days ago, wanting to know my opinion for upcoming magazine articles. Could you folks drop a note to the editors (I have been emailing Larry) to publish more in-depth coverage on this important topic?


• There seems to be only one industry analyst blogger that even talks about agile approaches. His name is Michael Cote. Could you provide some "amplification" whenever he blogs insights into agile approaches? Could you also add him to your blogroll?


• As far as enterprise architects in the blogosphere, other than myself, Scott Mark is the only other blogger than talks about agile methods. Could you not only amplify his thoughts but also encourage the other EA bloggers to start talking about more important topics?

Anyway, I think my takeway from the conversation with my industry peer yesterday was that it is futile for me to champion agile approaches as IT executives are already assimilated into heavyweight processes and I would have a better chance of success using a metal detector to find unicorns in my sock drawer than getting agile implemented on a large scale within any Fortune enterprise...

Many have stated that us enterprisey folks take simple things and make them complex while others have suggested that enterprisey folks don't take advantage of opportunities to learn from what Internet startups are doing. There is some truth to both...



The first fact is that many enterprises often overprocure in the name of "enterprise" class solutions when workgroup-oriented products will do the job. The main reason for usually doing this though really has nothing to do with the differences in features between versions but more in the way that folks have to think about budgeting. In many cultures, if you have a budget you are forced to consume, kinda the use it or lose it proposition. No one has really figured out a way to compensate folks in the enterprise for true cost savings.

You may read in many magazines about lots of projects done within enterprises that have had huge ROIs yet the annual IT budget still increases. Another behavior that plagues many enterprises is that folks simply get promoted based on how much money they spend. For example, If I were to champion use of an open source Enterprise Service Bus such as ServiceMix and figured out a way to keep support internally while another colleague say championed purchase of a $5 million closed source Service Bus, which one do you think will get promoted quicker?

Another evil that is starting to grow is the effect that Sarbanes Oxley is having on enterprises and their IT systems. It wouldn't take a rocket scientist to figure out that the vast majority of employees of auditing firms do not really have any business certifying IT systems and practices yet that is what they do. Audit firms are not employing CS majors but do employ a lot of accounting majors. Let's say the architect team at CSFB found a way to make all of their IT systems 100% SoX compliant using a couple of things freely downloadable from the Internet and spend a total of say $1,000, do you think that these accounting folks would actually publicly certify their solution or do you think they would need to see larger numbers and bigger solutions?



Several years ago, I did one thing that was not too enterprisey but was well received. We invited several architects into a closed room and played Diablo on company time which for the most part would have challenged lots of HR rules. The rationale for this wasn't about gaming but was about our ability to understand other architectural styles. We had a briefing with an architect who worked on the design of Battle.net which supports at peak 100,000 users concurrently with subsecond latency and wanted to change our perspective as to how extremely scalable applications are really written. It would have been wonderful to get that type of analysis from folks like Gartner, Forrester or even Redmonk but they simply aren't talking about architecture but are somewhat still talking about products and vendors. Of course, I am hoping to work with all of these firms to see if they could provide alternative perspectives that us enterprisey folks can more easily consume.

For bloggers that have read my blogs, may have figured out that I am of the belief that I can learn a lot about content management from Internet startups as well. I am of the belief that are content management strategy and thinking while is more mature than other enterprisey folks elsewhere it is still junior to say what Playboy has already done in many aspects. Imagine the attendance if I invited architects into a closed room and used the same approach as I did with a game? Now for reality, this simply isn't going to happen. I would be ecstatic though if some industry analyst firm were to dig deeper into their implementation and create a case study in which we could consume and learn from.

I have also had conversations with an architect I know of that writes gambling software and their architectures. They currently run their data centers outside of the United States and have a slick way of load balancing across countries. Gambling sites have higher availability than most corporate systems and never lose transactions. Do you think anyone in the analyst community ever talks about architecture in this form?



I had the opportunity to visit a data center in Antigua where many sites that are unmentionable run. They had technology that is more modern than data centers here in the United States. Do you think they get any coverage or will show up on a quadrant? This data centers had the most mature implementation of enterprise guest management I have ever seen. They literally are the reference implementation as to how things should be done. Anyone game for a roadtrip?

I have been noodling several things said by David Heinemeier Hansson and James Robertson who are co-Presidents of my fan club and their interesting outsider looking in perspectives on the enterprise. Felt it was not only important to respond to their insular thoughts but to encourage others to start having a dialog on how folks within enterprises truly think...



Folks such as Dan Jacob, Mark Watson, Charlie Savage, Danny Lagrouw, Jeff Schneider, Robert McIlree and Chad Dickerson and all provided pretty reasonable perspectives on how enterprises operate from their own viewpoint, but really didn't dig deep enough into the conversation in order to uncover new insight.

I am still wondering what it means to be too enterprisey? I know that 99.9999% of all folks that are employed in corporate America do not blog. The number of folks that are enterprise architects from this demographic who blog I can count on one hand. Since I am responding to your blog from my blog which I have been doing for over three years now, think it doesn't make me too enterprisey? I wonder if folks have noticed that I encourage other enterprise architects to also blog?



Most folks in enterprises also for the most part have seemed to miss the whole point behind the open source movement and think it is all about free (in terms of price) software. I wonder if I am too enterprisey if I have been savage in getting folks to understand alternative perspectives on open source.

Likewise, most folks in enterprises have no sense of community whatsoever and never really share what they learn with others. The hoarding of knowledge within enterprises is rampant but not where I work. I guess I am evil because I have written several books to date of which many of them have the title enterprise within them. I wonder if folks in the blogosphere have actually read one book entitled: A Practical Guide to Enterprise Architecture that they would call me too enterprisey after reading it? You may have noticed that one of the co-authors happens to be very prominent in the agile community? You may have also noticed in terms of the table of contents the desire that Agile Manifesto may be a great foundation to EA practices...



Maybe I have it twisted in that most enterprises nowadays really aren't writing lots of appications from scratch reducing the potential for Ruby to be adopted and that their real problem is in building the extended enterprise where they need to make one system communicate with another either internally or with their business partners. Feels like a conversation around Enterprise Service Oriented Architectures may be useful for folks who disagree with me to start noodling.

Hoarding of knowledge also comes about in that most folks in other enterprises work in walled gardens where they are not allowed to communicate with outsiders. I am the polar opposite. Would the community consider engaging in a face-to-face conversation at any of the events I plan on speaking at? You may find it interesting that I will be on a panel regarding web 2.0 at the Infoworld conference hosted by Jon Udell where I hope to represent and encourage others within the enterprise to not eschew but to embrace web 2.0.

I wonder if an insider starting calling out the flaws in so-called best practices for the enterprise and started to encourage others to think differently would this be too enterprisey? It may be intriguing to get an outsiders perspective on enterprise architecture so that you can help shape
Enterprise Architecture 2.0



The one thing I think that makes me too enterprisey is that I have strong opinions on something that are usually backed up with facts. Sorry to bust folks bubble in that Ruby will never follow the path that Java has taken. Don't get it twisted as this is not about Ruby hatin but more about making sure folks understand the dynamics of what really matters. Productivity is somewhat important but there are simply things more important than the current productivity conversation within most enterprises.

Enterprises several years ago adopted Java because the business side of the house wanted to participate in getting their business on the web. This couldn't have been accomplished using mainframe COBOL nor any of the client/server development tools of the time such as Powerbuilder, so they had to look at a new language that provided them with capability they previously didn't have. Java wasn't selected for productivity reasons.

Enterprises in the past have always eschewed for the most part serious use of dynamic languages and only used them for lightweight tasks. Rexx is one of the more popular dynamic languages used on the mainframe and is primarily used by the systems guys for utility oriented tasks but it never really made it into business-oriented application development. Enterprises though are finally starting to wake up to the notion of dynamic languages and more importantly the notion of declarative programming. This movement of dynamism within the enterprise isn't happening via approaches such as Smalltalk nor Ruby but is happening in terms of what is known as Business Rules Engines.

One of the sad things about dynamic languages and declarative programming via business rules engines is that enterprises have embraced them yet folks in Infoworld almost never talk about them nor do industry analysts ever provide meaningful coverage. I wonder if I can ask Stephen O'Grady and Brenda Michelson to start talking about business rules in upcoming blog entries? Would be interesting to see if they are of the belief that business rules engines have the potential to provide even more productivity lift for an enterprise over simply switching away from Java to Ruby.

As an individual, I tend to have tons of outstanding questions and haven't hesitated to seek out alternative perspectives. Maybe it would be useful for those who are outsiders looking in to pose several questions in their own blog to me? If your blog has trackback enabled I will surely make my best effort in providing answers to whatever your heart desires.

What is it you truly always wanted to know about us folks in large enterprises but were always afraid to ask?