Thursday, February 7, 2008

IT run-around crowd can't bypass the need for 'real' web services

I just finished reading a post by Patty Seybold in her Outside Innovation blog. In this post she bemoans the looming mashup hype, but notes that it's about time mashups enter the mainstream.

Too true.

Seybold goes on to say that mashup vendors are trying to convince potential customers that they don't need IT. In Seybold's words, these vendors are saying,

You don't have to wait for your IT department to wrap your back-end application functionality into real Web Services, you can "cheat" by extracting data in the form of reports and turning them into RSS feeds, which you can mash up with graphical user interfaces and tools.
She is implying, although she doesn't come out and say it outright, that this view of mashups is flawed. In this I completely agree.

Consider the sort of 'application' you can create with this approach. You can pull together data, you can bring in interesting visual elements, you can consolidate information, and you can look at it in a unified GUI. But you can't do anything. The data and visual elements aren't in the context of a business activity.

Clearly these mashups aren't going to be the killer apps Hinchcliffe notes are necessary for mashups to become a valuable enterprise tool.

Sorry, IT run-around crowd (Seybold's term) but any mashup vendor saying you can bypass IT and get what you need through RSS feeds, screen scraping and widgets isn't considering that you will need access to back-end enterprise services to get the most out of your mashup investment.

Let's consider an example. Let's say your sales manager wants to keep track of important news cross-referenced to either existing accounts, or opportunities in the pipeline. This could be put together by mashing up CRM data, available through reports and screen scraping, with RSS news feeds, all without the help of IT. That would give the manager some great data, but he can't do anything with it unless he hops out of the mashup and into the CRM system.

The real killer application would allow the sales manager to action the data within the mashup. So, for example, if the manager sees that a company in his pipeline just hired a new COO, the manager could push a button to create a TODO item in the CRM system requesting the account manager arrange an introductory call with the new executive. That sort of interaction isn't enabled by RSS feeds. It is enabled by services connected to back-end systems controlled and managed by IT.

What mashup platforms will do is reduce your dependence on IT to construct new applications. IT can take on the role of a trusted partner rather than gatekeeper, but you still need them. Any mashup vendor who says otherwise doesn't understand your business.

Wednesday, February 6, 2008

More 2008 predictions from the 'unnamed analyst'

Let me start out by congratulating Encanvas. They have joined JackBe in the illustrious group of mashup vendors who won’t let me review their mashup tools for this blog. Here’s what they had to say when I asked.
As Serena is a close competitor to Encanvas I’m not sure that would be appropriate but we have the greatest admiration for any organization that is encouraging agile computing and it’s great to have competition - so keep doing good things.

The funny thing is that as far as I can tell, Serena’s never been in a deal against Encanvas. Competition. It’s all in the mind.

And now for something completely different

The 2008 prediction machines are still at it, although I assume now that we are into February we will see them less often. One leading analyst firm, whose name I can’t give you since they would require I run this post through their vendor relations department, has made Web 2.0 predictions for 2008. Hint: They aren’t Gartner. As in the past, I’ll just call them the Unknown Analyst. (UA)

Here’s their 2008 mashup predictions.

  • Mashup vendors will start seeing a return on their investment in 2008, but at the expense of other Web 2.0 technologies.
Bloomberg at ZapThink already made a similar prediction. And just as I did before, I must disagree. 2009 will be the year of mashups rather than 2008. I hope to be wrong, but I don’t think I will be. Web 2.0 is just emerging into the enterprise, and mashups will be at the tail end of Web 2.0 spending. RSS, wikis, blogs and social networks are all going to be on the short list for 2008, with mashups entering the list in a big way in 2009.

  • Enterprise mashups will come out of the closet, and the vendors who influence standards boards and best articulate mashup value will set market expectations.
My take is that standards boards aren’t going to make a big difference. If mashups were stressing technology to IT customers, then standards would be important. But we need to remember that mashup customers will likely come from the business side of the house. Business users don’t give a hoot about mashup standards for security, widgets, REST contracts, etc. These are vendor problems the vendors need to solve.

However, while I disagree with the notion that standards board leadership will make much of a difference in market leadership, I completely agree that articulating mashup value will make a difference. If I can convince you that mashups will help you leverage your SOA investments to build those applications now languishing at the bottom of IT’s priority list, then I will likely get your business. If I can't convince you, then I won't.
  • Vendors from adjacent markets such as EAI and portals will enter the mashup fray in 2008. Pure-play Web 2.0 vendors will be the losers because big players like Microsoft, IBM, Oracle and HP will Web 2.0-enable their current offerings.
This is a very interesting prediction. In effect it says that stand-alone Web 2.0 technologies don’t matter as much as Web 2.0 technologies in the context of existing business problems. So will I need a social network solution in my company if my BI vendor puts one in the reporting portal? Do I need a stand-alone enterprise RSS solution if my existing reporting tools, Sharepoint and CRM vendors add the ability to export notifications as RSS or Atom feeds?

I think, for the most part, the answer is ‘no.’ Enterprises will be able to make do with Web 2.0 features added to existing vendor offereings. At least when combined with open source readers, wikis and blogs.

But mashups? I don’t think so.

Consider a scenario. Assume you want to pull your financial information together with lead data from CRM, marketing program plans from MRM and historical statistical information from BI. You also want to wrap a process around these data to calculate whether a program will generate the number of leads necessary, and alert Marketing to take action when it looks like a program will not meet expectations. If you were depending on one or the other of these tools to provide a mashup platform, you wouldn’t have a unified end-user experience, your process would be dictated by the host tool you happened to choose, and you would end up with the same point-to-point integration problems that have been causing headaches since the dawn of applications development.

In other words, you would have an integration, not a mashup.

So while I agree that some Web 2.0 technologies such as social networking, wikis and blogs may not be good long-term product strategies for independent pure-play vendors, I think mashups have a chance.

Right up to the point where they are purchased by one of those big guys, something not predicted by UA.