Business activity monitoring (BAM) is all about viewing real-time information and alerts about what's going on in your business, so that you can then take action if you see something happening that's going wrong, or which needs improving. The concept is closely allied with business process management, which in an ideal world gives you the ability to react by modifying processes in real time.
I was somewhat mystified, therefore, by the news the other week that webMethods and Informatica have teamed up to create a proposition called Business Activity Platform (BAP), which allows you to integrate processes using the two vendors' integration products, and then use Informatica's business intelligence tools to monitor events.
And what do you do then? Suppose you've implemented all this technology at an estimated cost of a quarter-million dollars for the software alone, according to internetnews.com's coverage and you gain the valuable insight that you could vastly improve efficiency if you tweaked several of your processes? After spending goodness knows how much hardwiring them together using these integration tools, are you really going to start dismantling everything again so you can act on your insights? The process infrastructure will be so brittle that you'll have no appetite at all for breaking it apart.
There are testing times ahead for webMethods as the shift to service-oriented architectures gathers pace, because somehow the company has to steer a path that migrates its EAI platforms and its customers towards SOA practices while at the same time preserving the viability of its business model.
At present it's so far off from achieving that objective that most of its product announcements come across as self-parody. First there was its brittle BAM proposition, which allows you to see exactly how much damage your integration project has done to your business while leaving you powerless to do anything about it. Then this week it began previewing a new version of its product that includes a free JBoss open-source application server, as if to remind customers of all the reasons they should be moving away from expensive, proprietary platforms like webMethods. The company will shortly be taking up its newly won place on the board of the Web Services Interoperability organization (WS-I). I can barely contain my anticipation.
posted by Phil Wainewright 12:14 PM (GMT) | comments | link
Thursday, April 24, 2003
Cutting out the fat
Systems integrators seem to be following one of two strategies as they adjust to the dual pincer of service-oriented architectures and the increasingly cautious economic environment. Both strategies come out very clearly in David Longworth's latest article here on Loosely Coupled, SIs adjust to living without integration.
Faced with a rapid draining away in demand for their complex integration skills, the established titans of the industry are aiming to seize the higher ground of business process consulting. Moving up to higher added-value offerings is often the preferred refuge of an industry under siege, but it's not always the best route. Remember, that's what the all big computer systems vendors did back in the mid-1980s in response to the emergence of commodity distributed computing platforms, and look what happened to them a bunch of now faded names like Data General, Honeywell, Burroughs and the like.
Meanwhile, smaller, less well-known SIs like Thoughtworks are heading in entirely the opposite direction, adopting the principles of agile development, which emphasizes small teams, tight budgets and rapid results. Even if the big SIs wanted to adopt such a strategy, they couldn't afford to. Their business model simply isn't lean enough to turn a profit on such meager returns.
There's an excellent, very approachable overview of agile development on the CIO Magazine website. It includes some wonderfully quotable comments from practitioners, such as Ken Moskowitz, CIO of Standard & Poor's in New York, who insists that all projects incorporate weekly builds of new software: "Go and build it, and we'll see if it was what I really needed in the first place.'" In another comment, Pat Morgan from Compaq's Enterprise Storage Group lets the cat out of the bag about the kind of unnecessary extra features that get built into traditional development projects, almost before anyone has noticed: "In our environment, you can burn a couple of million dollars in a month only to realize what you're doing isn't useful."
Some people believe the biggest threat to IT services revenues is offshore outsourcing, but that's just one symptom of a much broader malaise. The majority of IT projects in the past simply haven't offered good value for money, and now that everyone is scrutinizing spending, there's an incredible amount of fat being identified that can be cut out to save costs. Offshore outsourcing helps to shave off some costs, but if money's being wasted on unnecessary features to start off with, the impact is hardly likely to be as dramatic as cutting out those features altogether. The real threat to established IT services revenues is this new awareness that previous methodologies have allowed a huge wastage of effort to go unnoticed. While offshore outsourcing provides a competitive threat, agile development completely rewrites the competitive rulebook for IT services companies, and poses a serious threat to their very survival.
posted by Phil Wainewright 1:15 PM (GMT) | comments | link
Wednesday, April 23, 2003
W3C should stick to its weaving
No one should be sorry that IBM and Microsoft have done such a good job of steamrolling the W3C over business process standards. If nothing else, you have to admire the panache with which they assembled such an overwhelming show of force in their submission of BPEL4WS 1.1 last week to OASIS, as I've described in my ASPnews column this week. But some people may, like ZDNet's David Berlind, be feeling rueful at seeing the W3C so ruthlessly sidelined. Don't be.
The W3C fulfils a vital role in setting and developing standards for the Web, just as the IETF in its turn looks after standards for the Internet. When there are already two separate standards bodies looking after these two complementary layers of the Web, why do we need a third? The answer's simple. Business process sits at its own, separate level, and just as the IETF and W3C each bring different skills and disciplines to bear on the policing and development of standards for their respective layers of the Web, so a third body is necessary to ensure that business process interaction gets a robust set of standards that will work. OASIS, having the explicit mission of promoting XML-based e-business, is the best organization to do that job (though it will have to prove that it is not the hapless puppet that critics like David Berlind claim it to be).
The W3C has enough to do anyway looking after XML and the various immensely important emerging standards relating to its use, such as XQuery, XPath, XHTML and others. The W3C excels at ensuring a level, standards-based infrastructure for information sharing over the Web, and that's what it should stick to. There's a huge amount of work to be done. In contrast, it has no track record of successfully establishing standards for business interactions, and has never shown any great enthusiasm for doing so, either. Gracefully yielding the field of business process co-ordination to another body would be an astute move, as well as ensuring that it gets involved in the standards-setting discussions as a constructively engaged interested party.
I notice, by the way, that we now have a new acronym to learn. Having tracked down the official OASIS notice that launched its new technical committee on business process last week, I see that it has been named the Web Services Business Process Execution Language Technical Committee (WSBPEL TC). This will go a long way towards standardizing BPEL as the accepted acronym, rather than the more challenging BPEL4WS favored by Microsoft and IBM. I have always been too impatient to spell out the latter in conversation, preferring to pronounce it 'bee-pel-for-whizz.' But I much prefer the sound of the new version from OASIS, which, rendered as 'wiz-b-pel', skips off the tongue very nicely indeed.
posted by Phil Wainewright 5:55 AM (GMT) | comments | link
Tuesday, April 22, 2003
Post your news to our RSS feed
In an ideal world, of course, everyone would publish press releases and other news items as an RSS feed, and all we would have to do is to collect them into an aggregated feed. But since Cape Clear is (to our knowledge) the only vendor in our neck of the industry that currently does that, we decided to shortcut the process and publish everyone's news anyway. Anyone can contribute a press release or news item to our feed, even if they've never heard of RSS and we've never heard of them, simply by submitting it using our online form. Provided the item meets our criteria (and we'll write and explain why if we feel it doesn't), our editors will add it to the feed.
The submission form and the archive listing can also be embedded on your own web page in the same way (they're hosted on a separate database server that will allow us to add some nifty extra features later on). So instead of having to remember to come to our site when you want to add or look up something, you can keep them where they're most convenient to you.
This is all a bit of a departure for a content site, which according to conventional wisdom ought to be trying to suck in as much traffic as possible to its own pages. But I beg to differ. I think it's wrong to think of a website as a static destination. Better to think of it as a delivery hub, the point from which you disseminate information and services over the network on demand. All of this is part of a philosophy that I like to call "content-as-a-service". The Loosely Coupled website will be elaborating on that philosophy in the coming weeks and months, and in the meantime we welcome your comments and feedback on our services as they evolve.
posted by Phil Wainewright 4:18 AM (GMT) | comments | link
Assembling on-demand services to automate business, commerce, and the sharing of knowledge