Monday 29 January 2007

BPMN - A graphical notation for depicting business processes

The Object Management Group adopted the specification for

Business Process Modeling Notation (BPMN 1.0) in February 2006. This FAQ gives a useful introduction to this graphical notation.

A history and evolution of modelling with more questions than answers

Rob Vens blog posts (Clear the Fog) over the past year do seem slightly obtuse in some ways, but I guess that I am trying to dig into subjects that are not entirely mainstream, and architecture is already a rather ephemeral subject. Still, my eyes are apparently not open enough, as I see more fog than clarity.

However, there is one that sits squarely within my field of interest, Why use UML for business modelling, which is probably what led me to his blog in the first place. The article digs back into the history of modelling business activities, and looks at how they have resulted in the various languages, notations and methods that are being standardised more recently. His focus on object-orientation leads him to observe a number of drawbacks with the process-oriented approach, but unfortunately he fails to conclude with any solutions, or even avenues to explore. Perhaps there might be more clues if one shared his understanding of object orientation, as he summarises it in The Essence of OO

Interestingly, Rob suggests that process-orientation has been a drawback in effectively modelling businesses - he observes that this only came about because dataflow modelling was a common practice in IT, so the engineers could get away with applying those same tools to try and model business processes. I guess that I would contend that, from the point of view of wanting to make business support systems more efficient, it is crucial to look at the data flow that has been created, but I guess this comes down to an argument between purism and pragmatism - the desire to design a perfect system versus the concern with making step improvements in what has evolved up until now.

Wednesday 3 January 2007

See, I was right ... you CAN'T grasp a river!

I was starting to despair that my skipping through the blogosphere would have me only drinking from the fountains of hype-merchants and vendors (thus getting me drunk and loosing the plot). Then, as I was thrashing through tangled undergrowth of BPEL vs BPMS vs vendors' own viewpoints, I came across a calm stream of valid ideas from Keith Swenson in his Go Flow blog.



This is the first time, since starting to speak about the notion of mapping information flows in order to identify and remove the blockages, that I have found someone who appears to share a similar perspective (I though BPM Philosophy, not Technology was a particularly poignant post from Keith).



He may not be the most prolific blogger, but then its better to have monthly measures of substance than daily doses of dross. And like most good blogs it has connections to and from other valuable sources of information in the same space :-)

Don't use SharePoint for that!

Although Information Dynamics is my future, SharePoint technologies have been an important part of my past and present, so I am sharing this valuable article on when to prefer network shared file storage over SharePoint document libraries. Microsoft SharePoint Products and Technologies Team Blog : Is the File Server Dead?

I guess the next step would be to have the equivalent for structured data content (tables), so you can choose when is most sensible to store a table in an SPS list, and when its more appropriate to place it in an RDBMS like SQL

Tuesday 2 January 2007

First folders, then tags ... where will this end up

So you've realised that my thoughts are still all over the place on this - and that I have my head up in the clouds, but I do think my toes are still on the ground, so hear me out on this one. Also, in the hope that you'll forgive me for the link to yet another "solution without a problem", I'll try to justify myself from the outset.

Folders


So I discovered that folders were a great way to organize things into neat little pigeon holes - kept things from getting messy and I knew more-or-less where to find things.

Tags


Then I discovered tags, and realised this was a great way to overcome the limitations of grey, when subjects were too blurred to fit into a specific pigeon hole, or indeed belonged in more than one. Ok, so I first encountered them as keywords, and learnt the hard way how difficult it is to build up a valid taxonomy, never mind make sure that everything was properly marked up in a way that is visible and available in all situations.

Folders plus tags


Maybe I was lucky to have relied on folders for so long that tags never eclipsed their value totally. It did not take me too long to realise that rigid pigeon-holing still had value over and above an organic tagsonomy. Factors such as ownership or record typing still had to be black and white, with a clear and unarguable structure. After all, how can you wield ultimate power (or sly away from responsibility) if you have grey areas in ownership or unclear chains of command?

Missing the mark


And so the problems came in trying to find a good way of actually marking things up properly. I know that it kind of depends on record types and file formats, and we might never get to a perfect solution, but I was very interested to come across the idea of microformats. At first glance it seems like quite a strange and difficult idea to define (see microformats.org) but I think the article How microformats affect search engine optimization does a good job of explaining what they are and most importantly why they are useful.

I know there are a few too many buzzwords involved, but I have the horrible feeling that this going to be a common issue in the space I am currently navigating. I guess I'm going to have to get over that, but I'm not sure how easily you will.

powered by performancing firefox

A potential solution component

"I think that Microsoft's Windows Workflow Foundation (WWF) will prove to be a valuable tool in my armoury of technologies."

So why would I be interested in a solution already, if I have not yet got to the bottom of defining the problem?

Well, the simple answer is that I was impressed by what this vendor produced last time. As well as doing “big picture” architecture work, I have also been spending time doing some hands-on coding at the front line. I was tasked with teasing requirements out of business users and spinning them round to come up with quick prototypes of the application they desired. I discovered that what MS have delivered to coders in the ASP.NET 2.0 web controls makes it so quick to prototype and deliver neat applications (including code that is reusable and close to models).

Based on these results with the second version of the .NET framework, I am placing a bet on the 3.0 version. Yes I know that this version is just 2.0 with three new components bundled, but one of those components seems to fit squarely in my area of interest, where true solutions usually come with extremely large amounts on baggage. And here is one that will be "out-of-the box" on virtually every new PC shipped from the end of this month, and is already a free of charge download for a large proportion of the worlds PCs and a great number of servers too.

Here's a little glimpse for those of you out there who like pretty modelling tools (the Windows Workflow Foundation Rules Engine).

And here are some bloggers who periodically discuss WWF:

Matt Milner (trainer)
Dave Green (architect)
Paul Andrew (technical product manager)

Sunday 31 December 2006

Them's is fighting words, them's is!

Sorry, but I have come across an inordinate amount of waffle since starting to learn about this space (which I can't even define properly myself!) Everybody is touting the best thing since sliced bread, but nobody has yet managed to deliver any results. We would all be crying "what a bunch of HYPE-merchants" if a) it was popular enough to draw mass attention, or b) anybody knew what we we talking about or even cared, but nobody does so what kind of hype is THAT!

And then, amonst all of this noise, I hear a voice that seems to be offering clarity, substance and conviction - can it be true?

I suspect that only time will tell, but I am sufficiently inspired by what Ismael Ghalimi wrote in his article IT|Redux » BPM 2.0 (and others on his site) to start reading his forthcoming articles. Whether he is a mouthy self-server or visionary post-technologist, only time will tell - but he is pushing a very persuasive line where most people can't even manage to explain what on earth they are actually talking about, and for now that's very encouraging!