Friday, May 16, 2008

Dynamic documents as two-way end points help bind people and processes to SOA

Read the full paper. Listen to the podcast. Sponsor: JustSystems North America.

Making services oriented architecture (SOA) a fixture across larger swaths of enterprise IT and business processes has grown into a top goal. Finding additional innovation to amplify a SOA's value is therefore always welcome.

A separate but related trend in the field, of implementing managed XML-coupled dynamic documents via authoring and content governance, offers just such a high-impact SOA-enhancing value. Dynamic documents provide end points for SOA-delivered content and data, and deliver it into the formats and often required interfaces people -- meaning workers and managers -- need. Dynamic documents also offer many values around ease of language localization, automation of feed-delivered data, and centralized control over highly decentralized content.

Legions of those in the world that actually get things done -- the line of business personnel that must apply the digital world to the physical world -- are surrounded by documents. Documents, from spreadsheets to maintenance manuals, are the historical means through which people manage information. IT systems use documents to reach beyond their glass screens.

IT has not done away with documents, and it is not likely to. People at the end points of SOA-driven business activities will remain sort of like analog-to-digital converters, as well as digital-to-analog converters. They interface between SOAs and the real world, with documents as a bridge. We all do. And our mainstay interfaces consist mostly of static documents ... but increasingly those will be XML-enabled dynamic documents. How convenient!

That's why I found it fascinating to take on a research project to plumb the depths of how SOA and dynamic documents come together. The conclusion, contained in this report, is that those enterprises that implement dynamic documents capabilities can significantly leverage those investments by flexibly extending their SOA values out to those document end points.

The XML-enabled documents, in turn, can provide on-ramps and gathering points for more content and data to enrich the SOA activities. [Disclosure: The research report was sponsored by JustSystems North America, which is also a sponsor of BriefingsDirect podcasts.]

In other words, the trends around dynamic documents and the trends around SOA complement each other well. Architects and those departmental managers dealing with document overload and the need for better management, therefore, ought to be talking to each other. They may be able to help each other a lot.

Furthermore, the investments that organizations make in SOA can powerfully augment the value and utility of what they can do with dynamic documents authoring, management, and governance. It's also fascinating to consider how SOA-level governance and policies can play a role in how documents use and access -- down to a finely granular level -- can be managed and automated. Think of it as total process management -- from mainframe to everyman.

The result is that in the near future documents will behave a lot more like traditional applications, while traditional applications can behave more like SOA-driven processes. It's us to up to make the connections come around full circle.

Here are some excerpts from the report:
Combining the productivity enhancements of XML-based structured authoring and document management with the increasingly strategic benefits of SOAs is a next logical step. Embracing dynamic documents as SOA endpoints may also spur faster adoption of SOA principles and infrastructure.

If the accumulated business knowledge of individuals could better interface with services-enabled applications, organizations could combine the best of human experience with the new levels of IT interoperability. Any knowledge or semantic asset that can be identified, tagged, and contextually related to business functions should be made available to SOA composite applications as services.

This combination – SOA and easily authored dynamic documents – empowers line-of-business teams to innovate around how information is accessed, combined and presented. It allows organizations to improve the speed and efficiency of manual and disconnected document-centric processes, and to dramatically improve technology and knowledge transfer across lifecycles and value chains.

As XML dynamically updates data and content across myriad traditional documents, user benefits transcend the former static formats. Users can update documents, while their structure allows many others to access current data. Elevating workplace knowledge and data via the familiarity of documents -- and then extending that information across multiple business processes -- that’s what SOA is all about.

Companies with SOA projects should seek out documents as consumable resources – especially dynamic documents -- and then enlist them as resources for business-process benefit. Combined, SOA and user-friendly documents can substantially improve productivity, refine processes, integrate people and processes, as well as accelerate the financial payback from investments in both dynamic document publishing and SOA infrastructure.
Read the full paper. Listen to the podcast. Sponsor: JustSystems North America.

Thursday, May 15, 2008

BriefingsDirect Insights analysts probe future of online advertising and find transactional lucre lurking

Listen to the podcast. Read a full transcript of the podcast.

The future of online advertising captures the headlines and attention when the likes of Microsoft courts the likes of Yahoo! And Wall Street still has a hard time figuring out how much Google is worth, based on just those little text ads next to search results.

But the future of online business has a lot more in store than advertising as we know it. The cloud compute fabrics now being constructed can support a lot more finely tuned matching of buyers and sellers, for consumers and businesses alike.

In the latest BriefingsDirect Insights Edition, Vol. 29, our experts examine the future of online advertising, and how the gathering cloud of services hosts like Google, Yahoo, Microsoft and Amazon will fare in the next era. The consensus moves toward an algorithmic meta-data driven future in which the winners will likely be taking a piece of many online transactions. This real-time marketplace can scale up to global mass media, and down to the audience and location of one.

So join us for our latest BriefingsDirect discussion and dissection of software, services, SOA and compute cloud-related news and events, with a panel of IT analysts. In this episode, recorded May 9, 2008, we gather noted IT industry analysts and experts Joe McKendrick, an independent analyst and ZDNet blogger; Tony Baer, principal at OnStrategies and blogger, and Phil Wainewright, independent analyst, director at Procullux Ventures and ZDNet SaaS blogger. This discussion is hosted, produced and moderated by me.

Here are some excerpts:
I really think people have got this completely the wrong way around. To focus on advertising is just so "0.0," to coin a phrase. Advertising exists only because we don't have the Web. Advertising is something the B2B market has to use through magazines, TV shows, or whatever, because they couldn't reach the consumer directly.

Now, the Web enables people to reach potential consumers and business prospects directly, rather than having to go through this advertising. So, the idea that the software industry is going to get funded by advertising has got it completely the wrong way around. Actually, what is going to happen is that business is increasingly going to use software in order to get closer to its consumers and its prospects. It can actually skip having to spend the money on advertising in order to make that connection.

Let me explain how that might work, instead of running adverts on sites that host discussions about bookkeeping services for small companies, for example, or instead of paying for search ads that pop up when people are searching on the Internet for bookkeeping services for small companies. As a small company, if you are using a financial application to run your company and you want some bookkeeping services, a bookkeeping service might pop up as a menu option in the software. You can sign up for and use an outsourced service over the Internet.

Instead of the bookkeeping service actually having to advertise on the search engines, in the publications, the discussion forums, and the social networking sites, they just pay to have their service made available within a software package that relates directly to the service that they are offering.

Therefore, it's not really advertising any more. It's just product placement at a point where the consumer or the business, in this case, actually needs that service.

Now hold on. So, what we were saying is that business activities and consumer activities more and more move online. Not only will we be doing away with the on-premises software business to a significant extent, but we will be doing away with the advertising business to a significant extent. Then, no longer will the entertainment businesses be glossing themselves with adverts to support themselves, but, increasingly, we'll see placement of services in the context of an activity or process, be it for consumer, entertainment, or business, in the same way that we might go to a shopping mall. People pay rent to the mall organizer, which draws people in, to put their wares out on the doorstep in front of the glass pane, in order for people to pick and choose.

So we are moving from an advertising to a placement or even visibility value, and it becomes rent to those who can draw the people in.

I think that there are some indications that the bloom is off the rose of social networking, both as a significant revenue generator, as well as an application development platform, at least for one of the social networks to become a development platform. That's from some recent revenue indicators from Google that its relationship with MySpace has not proven to be as monitizable as they expected.

Some recent statistic show that the types of applications that have been generated on Facebook are very tenuous, very one-off or fun things that would appeal to teenagers, but not with any significant depth or business value. The amount of activity from developers on Facebook has been slacking off, or at least plateauing, which is not a good indicator.

I remember back in the Web 1.0 boom and the dot-com boom, one of the things that was interesting was the discussion sites were very bad at generating ad revenue, because people didn't click on the ads.

The cost per thousand (CPM) for discussion sites, or for the discussion area of a site, was always a lot lower than other types of sites that were more information heavy. So it's old news about kind of sites where people follow what other people are saying.

People start chasing page views without remembering the reason that they are chasing is to generate value for advertises. They think, "We've got lots of page views," but they don't think back to whether those page views are going to deliver value.

Another memo from Ray Ozzie surfaced a couple of weeks back. You may recall the memo back in 2005, the famous "turn the world upside down" memo that talked about the advertising support of the online model for software. He kind of reinforced that with his latest memo.

It wasn't saying, "We must offer software advertising to support software," but it was more of a discussion about the social mesh, the community, the social networking, a paradigm that's emerging.

It's going to be interesting, but I think it's going to leach into the enterprise over the next couple of decades as well. I'm talking years from now, but it's definitely a model that will be sustaining consumer computing. We are seeing that emerging on the social computing side.

You start looking at migration to digital broadcasting. At some point -- I don't know the exact technology mix involved -- combining that with the Internet, there will be some way of micro casting. There may be a large population segment watching a specific program, but you maybe identified in terms of which demographic you specifically are. It's almost sounding 1984-ish.

I think Google actually realizes that and understands that. Therefore what they are aiming to do is get into TV advertising and all these other sectors. These are vendors that enable this kind of personalization of the message, being a conduit between the prospects and the business that's trying to sell to that prospect, and using software automation to enable that.

They are thinking beyond the old model of advertising, and I think that's Microsoft's problem. Microsoft hasn't really understood this, is still thinking about online advertising as a segment, and is not looking beyond the wider opportunity to use the automation on the Web as a way of just bringing buyers and sellers close together.

This requires a tremendous amount of cloud compute to the same levels we have seen in matching search criteria to results and then matching that to advertising. That advertising is then bought through an auction bid process among those seeking the highest placement. So, if we take that same model and apply it to all sorts of different needs and wants of business, personal, entertainment, and luxury across the board, what do we call it? It's not really advertising.

So, we think that advertising is in the rear-view mirror. We're going to move to a new era of something different or better, perhaps subscription as a business model, where you, in a sense, rent digital assets.
Listen to the podcast. Read a full transcript of the podcast.

Wednesday, May 14, 2008

HP partners with Desktone to advance virtualized desktops as a service

Desktone, the desktop as a service (DaaS) provider, has lined up a powerful ally in Hewlett-Packard (HP), which has signed on as the first member of Desktone's partner program for desktop virtualization technology.

Desktone announced HP's involvement at the same time it unveiled its service provider partner program designed to enable service providers in the IT hosting, outsourcing and datacom businesses to offer DaaS to their clients. HP's Flexible Computing Services (FCS) will be the first participant. [Disclosure: HP is a sponsor of BriefingsDirect podcasts.]

HP, along with ClearCube Technology, also provides physical PC Blade Desktops. In this model, individual "client blade" PCs are used to host multiple independent user sessions, each one running on its own physical PC blade. In this case, it's possible to host as many client PC blades as you have rack space, power and data center space to accommodate, according to Wikipedia.

The Desktone partner program is aimed at service providers already in the hosting or outsourcing business and who want to leverage existing data center assets. Desktone said that partners who sign up in 2008 would have direct input into the company's Virtual-D platform product direction.

While many companies can benefit from virtualizing their desktops, building the infrastructure can be expensive, especially for small and medium-sized businesses. Acquiring the technology as a service, and paying for it as an operating expense can put the technology within the reach of many of those businesses.

For those who may be hazy on the concept of desktop virtualization, ZDNet blogger Dan Kusnetzky gave a short primer back in March on what desktop virtualization is and why you should care:

Desktop virtualization is encapsulating and delivering either access to an entire information system environment or the environment itself to a remote device. This device may be based upon an entirely different hardware architecture than that used by the projected desktop environment. It may also be based upon an entirely different operating system as well.

The Virtual-D Platform enables service providers to offer hosted, subscription-based virtual desktops through a single, automated self-service platform. Enterprises can realize the full benefits of centralized virtual desktops without having to build and deploy the infrastructure internally. The Virtual-D Platform comprises two tiers, enterprise and service provider, which lets enterprises maintain ownership and control over their desktops while outsourcing the physical data center infrastructure powering those virtual desktops.

I saw the vast potential of DaaS nearly a year ago, when Desktone announced a big infusion of venture capital. At the time, I wrote:

The ability to deliver a PC operating environment in a way users are accustomed to via grid/utility efficiencies in a way that appeals to the realities of enterprise IT departments and needs may be a seed that has a long way to grow. But compelling economics and the movement generally to services delivery portends a fast-growing new market segment for home, SMB and large business users. Telcos and cable providers will need to provide these kinds of services, for sure.

Desktone is part of a burgeoning ecology of desktop virtualization providers, including Quest's Provision Networks, Citrix, VMware, WebGlobix and Ericom.

Tuesday, May 13, 2008

Combined HP-EDS can explore missing methodology around how to offload IT to the cloud(s)

HP's now official pending EDS buy for just shy of $14 billion positions the combined companies to organize and manage the hosted/on-premises mix to maximum efficiency and lowest TCO. It's a great goal to shoot for because all they have to do is beat IBM.

With this merger, the IT/business transformation second-source in the global market is a alive and well. There's always this: The better IBM does, the more need there is for an alternative.

HP with EDS has now clearly staked its future on the top prize in IT: next-generation IT operations efficiency, proper outsourcing methods, cloud computing services management, and high-level consulting as the onramp. This amounts to business transformation via IT transformation via IT multi-sourcing.

Both business and IT need to change, but with a hugely complex migration process in store over the next decade. The end goal is a symbiotic and ultimately fully aligned means to business agility, innovation and holistic change management. [Disclosure: HP is a sponsor of my BriefingsDirect podcasts.]

But there's a missing methodology in this migration process, sort like the "missing link" of how IT and business will evolve from lumbering and reactive gatherers into sharp-stick wielding, proactive hunters and inventors. That missing methodology is a tried and true way to determine -- enterprise by enterprise, unit by unit, department by department -- what elements of IT to offload to clouds and what to embed deeply into the core business as strategic assets. This is the bread and butter of HP and IBM for quite some time.

Most companies and IT strategists now recognize that some portion of what they now do for themselves in IT they ought to offload onto someone else -- or at least provide it as a service via some hybrid underlying support means. The cost efficiencies, utilization rates, flexibility, marketplace-driven productivity aspects of cloud computing are simply too wonderful to ignore. We simply should not have standalone email servers every 60 square yards inside of companies. It's foolish. Same with a lot of other applications. SOA can help use and extend those assets better, but we also need to take a look at offloading them all too.

At the same time that we recognize a milestone shift in how software and services are used and matured inside of businesses, the macro environment is driving the impetus for the same transformation. Perhaps more than ever, businesses need to not only to be efficient and seek to reduce recurring costs -- they need to be able to adapt as quickly as possible, and never stop.

The missing link methodology needs to enable companies to adjust to globalization, raw resources/commodities scarcity, dreadful energy costs, transnational labor use patterns, Internet time, social networks, transaction-driven business models, and massive upheavals in e-commerce, media, transportation, compliance, and the usual vagaries of competing against tough competitors springing up from who knows where next.

Companies clearly need to innovate better, and that innovation must use and leverage technology far better than in the past, and at lower total cost over time. Yet IT departments are not designed (if they ever were designed) to innovate at speed or scale. They are designed to carefully support the crystal and china setting upon the legions of racks, and to prevent any bulls from entering the closet -- lest the whole thing crash, and no fingers to point at the cure. There is a huge disconnect between what IT does and what businesses need to do. It's not IT's fault, it's just the way it's all developed over time ... but it's largely a dead-end.

As a result, total business innovation must seek alternatives to just transforming internal IT capabilities and practices alone. Fortunately they seek these alternatives at just the time when those alternatives are increasingly available and viable. Choice on IT and business services off of the wire is entering a fertile and impressive stage. There will be lots to choose from. Choosing right is a big deal for the next decade.

But how to move best on this momentous opportunity? This is the question that HP-EDS can answer as the driver to their businesses growth. Only through deep, consultative partnership can huge enterprises undertake internal IT transformation while making the essential decisions about what to keep inside, and what to seek as the best services alternatives. At the same time, they need to build and adjust continually the business processes that are supported by these services from many sources. And they must position their abilities with multi-source IT with their current and future business requirements and goals.

HP's services units have been diligent about establishing meta methods that allow for both efficiency improvements, and transformation. HP's software and hardware units have been diligent about business technology optimization (BTO) and high-efficiency/high-availability computing. HP's acquisitions have given it an arsenal through which to operate data centers at peak efficiency and top operational integrity.

Adding EDS to the mix to tackle the definition of and implementation of the missing methodologies to take IT functionally to a multi-source level that actually enables businesses at the strategic level seems a very strong fit indeed.

Monday, May 12, 2008

SOA Software acquires respository and governance vendor LogicLibrary

SOA Software, a provider of governance solutions for services-oriented architecture (SOA), has acquired LogicLibrary, a leading SOA repository and governance vendor.

The acquisition of the Pittsburgh, Pa.-based LogicLibrary by Los Angeles-based SOA Software creates a more comprehensive SOA governance and automation solution, said the companies. The goal is to allow companies to accelerate their full adoption of SOA and rapidly deliver services for distributed and mainframe environments.

The merger underscores not only the SOA vendor consolidation trend (ongoing), but also highlights the market driver of more end-to-end governance and management aspects of SOA deployments. HP and TIBCO also had recent announcements that point up a wide and more automated approach to SOA governance/management.

We're increasingly seeing the means to relate the design time aspects of SOA with the runtime, or operational, aspects. This will no doubt be a big topic at the upcoming IBM Rational Developers Conference.

What's more, I expect to see more of this "total management" approach to SOA coming from the open source SOA infrastructure providers, too. The juxtaposition of SOA and cloud computing and wider use of server virtualization will also drive the need for better total management.

LogicLibrary's technology will extend its integration capabilities across both governed development platforms and governed service platforms. LogicLibrary provides a set of features with reporting and analytics capability focused on SOA development governance. Its products include an enterprise repository providing broad support and governance for development assets/services, along with deep integration and federation with IDEs and application-development point solutions.

The prevalence of services, both internal and external, in enterprise applications now requires companies to have an enterprise-wide SOA governance solution to ensure the integrity of their policies, the companies said. According to Alan Himler, chief executive officer and chairman of LogicLibrary:
“The combination of SOA Software’s governance products, with LogicLibrary’s strategy to provide federation with other leading repositories, creates a single solution that provides unparalleled lifecycle and policy governance across all major platforms.”
A year and a half ago, I blogged about the consolidation trend in SOA governance, and I raised the question of who would be next? While I listed the candidates in what I said was no particular order, SOA Software and LogicLibrary were in the top two spots.
So who’s next in the buy-or-be-bought sweepstakes? Likely candidates (in no particular order) include SOA Software, LogicLibrary, Progress, IBM, Novell, IONA, Red Hat/JBoss, HP, Cape Clear, Mind Reef, Rogue Wave, Cisco, Sybase, TIBCO, BMC, Borland, AmberPoint, Software AG, Composite Software, CA, Above All, Adobe, Oracle, SAP, Sun Microsystems, among others.
There are still some names here that may need dance partners. Fortunately, the music has not stopped yet.

Software AG's Mik0 Matsumura has some more thoughts.

Service Oriented Enterprise also reports on the merger.