sic transit axis mundi

The winter solstice, or Dies Natalis Solis Invicti, usually offers small opportunities for contemplation. Or maybe it’s just the gradual fallout of colleagues whose good intentions got frozen in time for whatever reason... In that sense, working in IT is both a blessing and a curse. No other profession requires such level of continuous learning and the joy of working with smart people. Yet no other profession seems to be so overwhelmingly overhyped with new technologies, methodologies, paradigms shifts and whatever else. And like no other profession i’m seeing people’s smarts being used against them, by others or by themselves, and that seems mainly to do with a strange want that people are fundamentally rational.
After several years in the spotlight of our attention, SOA seems on it’s way out, at least on the popularity rankings. Feels a little like studying at university, that at the moment the whole landscape made sense and offered an abundance to make the right decisions, game over. Although considering tools like OpenESB the SOA approach is becoming undetectable mainstream and exceedingly powerful. Not that i’m going to miss the SOA pooha clutter, but it’s kind of typical for the state IT is in considering the addictive urge for things new.
Now, OpenESB touches on something. Having worked with it’s commercial predecessors for some 8 years i have always been a bit suspiscious about the promise of Open Source. As ‘The Cathedral and the Bazaar’ points out eloquently factors like instant ego gratification play an important role in Open Source communities, which is also it’s limitation. Now that might be old fashioned Open Source, but what i’m seeing with OpenESB is a rapidly accelerated iterative development process where openness allows for instant feedback, clarification and improvement. An assembly line modeled by M.C. Esher.. If this openness is a sign of the future, then i’m all for it. This is ‘free as in freedom’, http://www.faifzilla.org/ch00.html, in a new business model, a SOA organisation as Anne Thomas Manes so clearly foresees. It reminds me of what Taiichi Ohno meant for Toyota, which involved reverting back of the production process to it’s human measure. Simply put, if leaders merely "implement techniques" without fully developing people, their system has no heart. Having heart is rather important in Japan.. Heart is spirit and having that indicates a ‘system’ is viable.
Maybe this is what books such as Peopleware aim at, but i just haven’t had enough time to look into that one.. In recent years innovation has become a popular subject for study, and what is rather amazing is that there is a strong tendency away from the human factor to stress the ideas of collective intelligence. Of course the legend of the lone genius, flashes of insight and such are rather exaggerated, but might it just be there’s a mix between the context and the persons involved ? Maybe some people are more susceptible for contextual feedback and rise above the lowest common denominator to do grand things during the right circumstances. What proof is there collective intelligence will rise above the mean ? Amazon’s book suggestions surely don’t indicate such. Maybe some ants within the entire ant colony are called Einstein.. Just expecting the system to self-organize is as unrealistic as expecting the European Union would result in lower prices while willfully ignoring economical models such as ‘increasing returns’. But maybe it’s just the friction between control and trust that many businesses prefer to hear about abstract methodologies harvesting human potential which can rule out any dependency on key figures.
Anyways, back to OpenESB and Sun, if Sun manages to keep the current momentum in the software area while making a suitable transition away from hardware, maybe very soon we can speak about the very first real SOA company, a company that lives and organizes the same way as what the Java Enterprise System is foreseen to enable. Not just composite applications, mock-ups, facades, but federated dynamic assembly of semantic applications. Workflow-as-a-Service, on steroids. Let’s give it a go.
Even more so, let’s give it a go, so eventually i can work out some a-life ideas with the current illustrious codename of ‘cotropous gauge maze’, freely based on Hamilton’s quaternions... And on that note a couple of quotes for the new year from EWD who, knowingly or not, has worked on that wonderful area where mathematics and physics so deeply coincide and intertwine, IT. Bugger it if he and Godel had gotten together and formulated something Wolfram is point to..

John von Neumann draws attention to what seemed to him a contrast. He remarked that for simple mechanisms, it is often easier to describe how they work than what they do, while for more complicated mechanisms, it is usually the other way around.
Edsger Dijkstra, Trip Reports, 213
...Simplifications have had a much greater long-range scientific impact than individual feats of ingenuity. The opportunity for simplification is very encouraging, because in all examples that come to mind the simple and elegant systems tend to be easier and faster to design and get right, more efficient in execution, and much more reliable than the more contrived contraptions that have to be debugged into some degree of acceptability.... Simplicity and elegance are unpopular because they require hard work and discipline to achieve and education to be appreciated. Edsger Dijkstra, The Tide, not the waves; in Denning/Metcalfe: Beyond Calculation, Springer-Verlag 1997
... while we all know that unmastered complexity is at the root of the misery, we do not know what degree of simplicity can be obtained, nor to what extent the intrinsic complexity of the whole design has to show up in the interfaces.
We simply do not know yet the limits of disentanglement. We do not know yet whether intrinsic intricacy can be distinguished from accidental intricacy.
Edsger Dijkstra, Communications of the ACM, Mar 2001, Vol. 44, No. 3


Maurizio said...

I agree Sun could have the potential to model a different kind of company, a flexible service oriented system with the capacity to apply opensource concepts, extending them to the enterprise better than anybody else. But they need to change a lot in terms of operations and internal organization, as the company now is stuffed with too many hardware-oriented middle-managers in search of identity.

While I was still in Sun, discussing with my frustrated fellow software colleagues, the problem seemed clear to many of us and the solution evident: spin-off. The company as a whole cannot compete this way, there too many internal frictions, the single-country organization was ok ten years ago but now something more agile is necessary to resolve the present Babele. The Java and software services stuff should be a new and different company, maybe hiring younger people with software sensibility and strategic acumen. The lack of strategic thinking perceived during a Sun's internal meeting is incredible. One experienced Sun manager, while I was complaining against the last unfruitful and boring meeting, once told me: "you know, most of those guys were hired while selling for Sun meant just waiting in the office an order coming by fax from eager customers".

About Peopleware: I read the book and I know personally about a couple of companies which really applies Tom DeMarco's principles:



kukan said...

Ciao Maurizio,

Peopleware indeed looks interesting.. i'd like to take a step further towards even more openness as hopefully the little micro-rant below will clarify.

Fortunatelly i have never had the misfortune to share your experiences of working internally at Sun.
Need breaks laws, Sun needs high-quality software and service people, they need experienced individuals, and as these people are mostoften working for their own the only pragmatic viable way to do so is to partner up with them like no company has done in the past.
Let's give it a go, it's time for change.

Take care