Want Fries With Your Change(s)?

The other day. I unexpectedly ran into an old, familiar, and spot-on concept originated by the science fiction author and futurist Arthur C. Clarke: people overestimate the short-term impact of technology, but underestimate its long term impacts.

Some Web 2.0 proponents need a reminder of this timeless axiom, given their postings of late. For example, let's start with Dion Hinchcliffe:

As for other significant Web 2.0 trends, Web 2.0's techniques are starting to bleed into the enterprise, something I call Enterprise Web 2.0. The heavyweight and ponderous techniques for enterprise architecture and even SOA are just not anywhere near as vibrant as very similar approaches out in the wild.  The mashup community on the Web is extremely active, even though still in its infancy.  It won't be long before you see a lot of the lightweight Web 2.0 development techniques and tools, like Ruby on Rails, become mainstream in corporate software development.  We are seeing surprisingly active interest in the conference circuit, with a almost surprising number of sessions on SOA, Ajax, and Web 2.0 in the enterprise in the next few months.  Gartner has even coined the phrase for a SOA model that is compatible with the Web 2.0 world: Web-Oriented Architecture or WOA.

Dion may go to a lot of conferences, but it appears that he hasn't set foot inside a major corporate or government IT organization in quite some time. Yes, emerging technologies like Ruby, Ajax, and others are making their way into IT development in larger organizations. Will it replace existing tools and operational systems completely in the near future? I wouldn't bet on it, and neither will the CIO's and CTO's I know who get paid to keep the operational "lights on" in their organizations. Why? I can think of three things right off the bat: 1) legacy systems; 2) massive operational systems such as ERP, CRM, and supply chain software; and 3) costs and investment already sunk into operational systems.

37signals software is very impressive - for workgroups, Would I deploy their wares today on an enterprise scale with tens of thousands of users? Not at the moment. And I'm not saying that they and others will never get to that point, but they haven't yet.

Sorry Dion, but I have to question how a programming language and framework, by itself, alters the corporate development landscape significantly in the short-term. The fact is, it doesn't. Because it cannot without a whole slew of different elements coming together with it on an enterprise scale. If any government or Fortune 1000 IT shop is in the midst of such a metamorphosis, I would love to hear all about it, as would just about everyone else in our corner of the blogosphere.

Later on in his post, we get this as one of four "Apparent Web 2.0 Trends:

Traditional Software Vendors Will Struggle in a Web 2.0 World - Microsoft and Google will likely figure it out, though it's not a sure thing either.  Microsoft has serious product line baggage and Google has healthy challenges in managing its growth and maintaining a sharp focus on strategy.  Google's latest products don't seem to have their famous edge, for example.  The smaller, nimbler Web 2.0 startups might continue to be a great source of innovation but it might make sense for Google to acquire startups and  immedatiely spin it off to avoid the "big company effect."

Really. You missed a few other substantial players here, such as Oracle, SAS, and SAP, who aren't exactly losing money at the moment and aren't having their systems shut off and replaced by some mashup with a MySQL backend database. Last I checked, Microsoft still makes very healthy profits from 'product line baggage' such as Windows and Office packages. Is that supposed to end next week? Next month? Next year? And who, praytell, is stepping up to the plate as a plausible replacement? Writely? Please....

The traditional packaged software vendors that 'get it,' and the firms I mentioned above do to a certain degree, are well aware of the shifts taking place. Oracle wouldn't be buying, or attempting to buy, open source outfits if they didn't see something coming. Microsoft and Google have always been purchasers of smaller, more technically nimble outfits - even before all of the Web 2.0 mania exploded onto the landscape. I don't see that as "struggling" or reactionary, I see it as astute business.

Finally, we get to the issue of timing. To hear pundits like Dion and others tell it, this metamorphosis should be complete in a very short timeframe. There is no question that the new technologies and the changes that they enable will occur, its a question of how fast. Large corporations and governments aren't known for agility in these matters when it comes to IT, and I don't see how these technologies are disruptive enough to change that mindset anytime soon.

Unless its disruptive enough to significantly alter their cash flow and profits downward very quickly, which isn't going to happen. Heck, even GM could screw up for decades in the face of far better Japanese vehicles before having to face some serious survival music...

Comments

[Dion Hinchcliffe]

Hi Robert,

You wrote quite a detailed analysis though I disagree with many of your points of course.

I do actually work with very large enterprise clients all the time and I do find that CTOs and CIOs do indeed track much of this, which surprised me as well. But then again, they wouldn't be much good if they didn't.

I would argue that Ruby and RoR will actually be disruptive in the near term. My recent fandom has actually been driven by watching corporations begin adopting it with surprising speed. Interestingly, most of the adoptions I'm seeing are coming in the back door first since it's the developers that often long to access the radical simplicity and ease of use that RoR and the like offer. But then it stays because of the results that start materializing.

Finally, I'm witnessing a big backlash towards the complexity of many enterprise technologies. J2EE is widely considered to be a significant mistake and most of its central features are actively avoided. This is all while WS-* is having enormous difficulties with implementation and adoption. Last data point: We recently had a large client that asked us to wholesale convert their SOAP Web services to RSS since none of their content management tools spoke SOAP but they all could handle RSS.

Genuine simplicity in EA is coming strongly into vogue, even though the enterprise has become more complex than ever. Coincidence? I don't think so.

In the end, people at the top want results. Tools and techniques that deliver those results quickly will float up rapidly. And since I'm actually seeing it happen, that's why it gets my attention and my respect.

Just because this won't unseat the big players any time soon doesn't mean folks shouldn't start looking at it as approriate, and as soon as possible really. With the upside becoming as obvious as it is, I would argue that anything else could be considered irresponsible.

Best,

Dion

[Bob McIlree]

Hi Dion,

Actually, I think we're at odds in only one aspect: the rate of assimilation of newer technologies in large enterprises. I'm saying essentially two things here: 1) that assimilation of this scale takes years, not weeks or months; and 2) a similar amount of time is usually indicated to dislodge huge industry players, if it occurs at all.

Large enterprises and governments appear to be risk-adverse because of scale and contraints/non-functional requirements germane to them. Those are the primary drivers of technology assimilation from my perspective. Scale can be satisfactorily handled in many ways, but the constraints and non-functional requirements are, and will continue to be, problematic.

Regards,
Bob