Architecture gem

After a couple of months of forced inactivity I can now pick up blogging again. What better way to start again by writing about two of my favorite subjects:

• Architecture and its practioners, and
• Government IT.

Since I studied public administration back in the days, I occasionally read the magazines that are typically read by former public administration students. I came across an issue of "Digitaal Bestuur" (digital administration), of which the theme is "Government and ICT". This issue had an article by a lead architect (!) from a government body, with the illustruous title "IT is difficult. IT within government is extra difficult". Fascinated by this title, I started reading and came across the following excerpt:
"It is getting more and more difficult to control the increasing complexity of IT. Computers have become 1000 to 1500 times more powerful in the past 15 years, as a result the systems have become 1000 to 1500 times more complex, and to make matters worse computers nowadays all communicate with each other through networks."
After I had read the above excerpt 5 times (I really wanted to make sure that I really read what I thought I was reading), I finally understood why governments are struggling so badly with IT and why architects are still having trouble to get accepted or even taken seriously by other IT and enterprise disciplines.

Modernization 2.0

When I was stuck in traffic again this morning and listening to the radio, I heard about the news that Belgium is facing tremendous trouble in forming a new government after the most recent elections. The same has been true in the past for The Netherlands, where the results of the elections provided the image of a very divided country, with quite some votes for the more extreme left-wing and right-wing parties. It took quite a while to form a government, and the Dutch voters are now stuck with a government that nobody is really enthousiastic for (except maybe the families of the new ministers and the memebers of the three coalition partners).
But enough about politics, this blog is about perspective on IT. But I think this trend in politics could have something to do with a growing structural differentiation in society due to technology. Let me explain why.

Somehow, I really have the impression that in many aspects in society and our everyday life we are experiencing the consequences of what sociologists call "modernization". This refers to a concept that describes a process in which society goes through industrialization, urbanization and other social changes that completely transforms the lives of individuals. Key elements in modernization are structural differentiation and cultural generalization, or in plain English increasing individualism and mono-culture. Lately, this modernization appears to be happening at warp speed, not in the last place because of the explosive growth of technology usage in Western society. The use of technology and new (IT) concepts, especially those branded as Web 2.0, accelerate the processes of structural differentiation and cultural generalization.

I wrote about culutural generalization before in a reaction to Nick Carr's claim that Web 2.0 is amoral. Former Internet boy wonder Andrew Keen goes as far as saying that today's internet is even killing our culture. Although I have some second thoughts hearing this from somebody who made a fortune on the internet (that's not "eat your own dog food", but "spit in your own dog food"), I have ordered his book for the Christmas holidays that are coming up. It should be an interesting read.

On the other hand, there appears to be some evidence that the explosion of internet usage is also accelerating structural differentiation I mentioned earlier. According to Oxford University Press, societies are seen as moving from the simple to the complex via a process of social change based on structural differentiation. The process may be imagined, in its simplest form, as an amoeba dividing, redividing, then redividing again. Society is evolving from quite simple structures into seperate institutions of education, work, leisure, government, religion and social contacts. Most people do not stay with the same employer their entire working life anymore, some even work at 2 or 3 different employers, churches see a declining number of people, small political parties are attracting voters from the traditional larger parties, mass markets are being replaced by niche markets (Long Tail, anyone?). In fact even identities of single persons are differentiating: John Doe who works as a loyal clerk during day-time, may be SuperVixen666 in his favorite virtual world at night.

Media is already heavily influenced by the 2.0 phenomenon (serious newspapers publishing movies of incidents that were filmed by readers with their cameras on their mobile phones), politics arguably also is facing the consequences of the wisdom of the crowd (some might say the power of the mob) and the collective intelligence. It will be very difficult to predict where this is all leading to, but for sure society is dealing with Modernization 2.0.

Oracle makes offer for BEA Systems

Press release by Oracle:

"Oracle Corporation (NASDAQ: ORCL) today confirmed that it delivered a letter to the Board of Directors of BEA Systems, Inc. (NASDAQ: BEAS) on October 9 in which Oracle proposes to acquire BEA for $17.00 per share in cash. The $17.00 per share offer is a 25% premium over yesterday's closing price of $13.62."

If I did not know anything about acquisitions, I would think that this is Larry Ellison's reaction to the Business Objects acquisition by arch rival SAP (don't get mad.... get even). But as these acquisitions and offers usually take quite a long preparation time, I do not think this is a reaction to the SAP activities.

Leaves me wondering though what the rationale is for Oracle to make the offer for BEA. Oracle states in its press release that "the acquisition of BEA by Oracle will enable an increase in engineering resources that will in-turn accelerate the development of our [Oracle's, LB - sic] world-class suite of middleware". Which might mean that Oracle is facing some challenges, maybe even problems with its current Project Fusion program. Putting it all together is one of the key challenges Oracle faces in its Fusion endeavor, and the acquisition of the brightest engineers from BEA could be considered a way to face this challenge. This should not come as a surprise as I have said before that one of the key drivers for Oracle's acquisition wave was "acquiring the people who created the technology".

Also, the potential acquisition of BEA supports my prediction and observation that increasingly pure-play vendors are disappearing, as the functionality of their products is incorporated into enterprise applications, OS and hardware.
Maybe this is the moment to say: integration middleware is now officially commoditized.

Why SaaS will make and why ASP didn't

I picked up blogging again last week when I wrote that the acronym SaaS does not ring many bells yet with decision makers. Maybe ASP does, but that is something many decision makers do not want to get involved with as that is so 90's.
Still there are many people that think that SaaS is the same as ASP. Take this definition for instance which I found on the web site of the Dutch research firm that concluded that SaaS is not known by many decision makers:
"SaaS (Software as a Service) is software provided as a hosted service which is accessed over the Internet, and which is billed through a subscription structure. This is also called ASP(Application Service Providing)."

Let me explain why I think this makes no sense. To start with, the delivery model is completely different: with ASP, you pay a monthly fee for a fixed period (sometimes even for the duration of 3 to 5 years), while with SaaS you only pay for the usage ("pay as you go"). This may sound trivial, but it has consequences for the customer orientation of the provider: because a subscriber is not tied to a fixed (maybe long) period arranged by a contract, which is the case with ASP, it is easier to make the switch to another provider. This implies that the SaaS provider must do more for client retention, hence deliver great customer service.
The architecture of products is usually also very much different between an ASP and a SaaS provider, mainly because SaaS products have been developed from scratch to be used over the Internet (take Salesforce.com as an example). The ASP versions of products have usually been "web-enabled", but still have a typical client-server architecture. In the past, the ASP model therefore never fully delivered what it promised, because of the lack of bandwidth, or because of a performance degrade resulting from the web-enablement of the application.

In short: both the delivery model and the architecture of SaaS and ASP are different: SaaS solutions are generally speaking better prepared (optimized) for usage over the Internet, and its delivery model is more flexible than the ASP delivery model. So, we are dealing with different animals (that live in the same Zoo though, named Outsourcing). The key advantages of SaaS over ASP make that I believe that SaaS will succeed, as these are exactly the reasons why ASP did not succeed.

Are we dealing with GAF(general acronym fatigue)?

Two recent news items have made me wondering whether we are dealing with GAF: a General Acronym Fatigue.

The first news item is from a Dutch IT newspaper, which quotes research from a Dutch IT research organization: 73 percent of all IT decision makers do not know spontaneously what SaaS (Software as a Service) is. Only after an explanation of what is meant by SaaS (a formal definition), 69 percent say they know SaaS. Which leads to the conclusion that only 4% of IT decision makers really have no clue what SaaS is about, and 69 percent are just confused by the acronym / terminology. An ASP (Application Service Provider, not Active Server Pages) expert in The Netherlands says this is due to the fact that companies, vendors and end-user organizations all have different names for the phenomenon: "Web 2.o", "internet-accounting" or "online services". Some people have even called it "ASP 2.0", which is really silly as there are substantial differences in the delivery models as far as I am concerned.

The second news item has been not on a single source, but in fact has been heard among many blogs: more and more people are seriously questioning the ESB (Enterprise Service Bus) is worth using. I have answered this question already some time ago: it is not! It is just replacing one low viability solution (CORBA or brokers) with another (service buses). I have plead before to use Apache for instance as a service bus (or rather I would name it service intermediary), and obviously more people have come to this conclusion. But then of course, we are talking about the design pattern, and not about the commercial product type some people considered to be the next generation application server. I suggest to drop the name ESB, and instead to name it what it is: a service intermediary.

The future does not look toto bright for people confused with acronyms, in particular because the aaS letters are pasted to any random technology or product category you can imagine: PaaS (Platform as a Service), CaaS (CRM as a Service), AaaS (Accounting as a Service) well, you get the idea... the good thing is that we run out of options at 26 acronyms...

If we (yes, everyone in IT is responsible for this GAF situation) are unable to come up with good names for new concepts, patterns, technologies or products, we should not blame those poor IT decision makers for not understanding our value proposition. In 9 out of 10 times we are not dealing with an ignorant decision maker, but just with someone suffering from GAF.

OK, I am off for now, I am going for a BLT.

The Second Life world is flat

Some folks have gone really excited about the "news" that Linden Labs is going to open source the back-end of Second Life. The amount of attention this non-news has gained, confirms my observation that Second Life is at the peak of the hype. Why? Because there really is nothing new or spectacular here in the story most bloggers and authors refer to. It comes down to Joe Miller, VP for platform and technology development at Linden Lab stating the following at VW07:

• "We’ll be open-sourcing the back end so sims can run anywhere on any machine whether trusted by us or not."

OK. When? Under which conditions? How? Just this statement is not truly shocking if you ask me.

• "We’ll be delivering assets in a totally different method that won’t be such a burden on the simulators."

If I understand correctly, the move is primarily driven by the lack of capacity for the back end run by Linden Lab.

• "Very soon we’ll be updating simulators to support multiple versions so that we don’t have to update the entire Grid at once."

Again my question: when? When is very soon?

• "We’ll be using open protocols."

Which ones? I heard rumors that IBM is having talks with Linden Labs on protocol developnment. With the WS-* disaster in the back of our minds, of which IBM was one of the key creators, we should not expect any added value really here. In fact if I were in a very cynical mood I'd say this is the kiss of death for Second Life, but it's Saturday so I am not :-).

• "SL cannot truly succeed as long as one company controls the Grid."

The bottom line, although it should read that SL cannot survive as long as one company controls the Grid (please note the uppercase G here. As if we are talking about the Matrix. Would have been even more striking if also an uppercase T was used).

Dana Blankenhorn wonders out loud if open source can save Second Life. He invites people to share their insights. My 2 cents is that Second Life is nothing more than a finger exercise for really successful 3D virtual communities, which will not be mainstream before 2009-2012. Linden Labs have done some pioneer work in the area, and they and the users of Second Life (I refuse to use the term resident), including the big companies that have jumped on the phenomenon will have learned for their experiences. But they will come (or already have come) to the conclusion that the Second Life world is flat, meaning they can fall off at any moment.

Beyond Kathy Sierra

Quite frankly I never heard of Kathy Sierra before last week. I could say here thatg I have been a loyal reader of her blog for a long long time, but reality is that I first heard about her when all the blog postings came in about the death threats aimed at her. And obviously a lot of people are interested in this subject:
Most reactions throughout the blogosphere vary from supportive, to shocked, and in fact James McGovern was one of the first who had a somewhat skeptical reaction to Kathy's story. I think this was quite brave by James, and although his later postings about the subject were not as good in my opinion (but I suspect James is going quickly for the 1000th blog posting so he can retire as a blogger as he announced), he has brought up some very interesting points, that even go beyond the whole Kathy Sierra case.
I do not know Kathy Sierra so I do not have any opinion on how she has handled this whole situation, although generally I do feel that you should never bend to those that use violence. That was my opinion last year with all the fuss about the cartoons in the Danish newspaper, and that is my opinion now. But maybe it is different if it happens to you, I don't know. Nevertheless, I wonder what the motive is of people threatening a blogger. It wouldn't surprise me if we are "just" dealing with some bored teenagers who just want some attention, kind of the same breed as those people that write computer viruses (still can't think what's so gratifying about that).
The key about the whole discussion goes beyond Kathy Sierra as I said earlier. James pointed in his post to the single largest threat of the Internet in general:
"In cyberspace it is very easy to become someone else and very difficult at times to prove that you aren't really you"
This has proved to be very difficult in the past and present (phishing, threats on discussion boards, electronic voting etc.), and it will become even a bigger issue in the future. With the advancement and expansion of new forms of collaboration, social contacts and transactions that use the Internet, identity is becoming the key issue. Obviously there is a growing number of people that have a need to take on another identity when being on the Internet. How often do you see nick names, avatars or email aliases that reflect exactly the name or appearance of its owner? Even with some very popular blogs, the writers chose to use a nick name, probably because they were writing about quite some confidential information on their company.
Anonymity / namelessness is very valuable to many people on the Internet, but it will become a complicating factor in the evolvement of the Internet. Especially as it appears as though increasingly virtual spaces / worlds and the real world are blending, this becomes increasingly problematic. Particularly because legislation is still in progress and absent in this area. In fact some people will even argue that the Internet should not be subject to legislation from any country (tell that to China...), and should be subject to the self-cleaning capacity of the World Wide Web and its users.
Which leaves us with the very interesting observation that on one hand the current evolvement of the Internet has the promise of providing maximum transparency, but on the other hand it is threatened by the obscurity of its users. Information becomes transparant, users don't. All the ingredients are there for a dialectic process of change.