Monday, May 09, 2005

Blogging from the Lawson user conference

I'm in San Diego today, attending the Lawson "CUE" user conference as a member of the press. I'll be updating this post through the day with observations. There is an announcement about to be made concerning a major change in the technology platform for Lawson, code-named "Landmark," and we're supposed to learn more in the keynote session that starts in a few minutes. In the meantime, read the press release on Lawson's website.

Lawson's Landmark business application platform
Dean Hager presented an overview of Landmark, and Richard Patton, Lawson's Chief Architect, provided more details in a press conference. Afterwards, for more answers, I interviewed Patton along with founder Richard Lawson and Pramod Mathur on the Landmark team. The rest of this section is going to be a bit more technical than I normally write, so be forewarned.

Landmark is Lawson's new technology platform that will, over time, replace the current Lawson "environment" that allows Lawson applications to run across multiple databases and operating systems. Landmark allows Lawson's business domain experts to specify applications in a high level domain-specific language (DSL) which then generates JAVA program code. Lawson's approach is based on "pattern language technology," including the work of architect Christopher Alexander in The Timeless Way of Building. Patterns have been a hot topic for a long time in the software engineering community, especially among practitioners of object oriented design and development, but I have not seen it discussed at all in the realm of enterprise application software.

When I read the press release, at first I was concerned that Lawson was heading down the path of developing a proprietary CASE tool or 4GL, a strategy that has failed most other vendors. But Lawson is not taking such an approach. Landmark is essentially a plug in for the open source Eclipse application development framework. The program code that it generates runs on any J2EE-compliant application server, such as the open source Tomcat or JBOSS, or proprietary servers such as IBM's, BEA's, or Oracle's. Software objects created by Landmark will be enabled for web services, allowing them to interoperate in a service oriented architecture.

Lawson believes that its approach allows it to focus on its domain expertise and what it calls its "signature features," such as "drill around," audit trails, and effective dating. Software quality should increase greatly as Landmark's code generation approach allows orders-of-magnitude reduction in the number of lines of program code. Patton quotes a 17x reduction in lines of code for Lawson's Vendor module, and a 41x reduction in Lawson's Employee module.

In terms of customer rollout, Lawson is emphasizing a co-existence approach: new Lawson modules will be written from this point forward under Landmark and will coexist and interoperate with existing Lawson modules. Existing modules will be "re-factored" in Landmark. Installed base clients can move incrementally to Landmark written versions.

Lawson's leadership is genuinely excited about the introduction of Landmark. Richard Lawson himself largely disappeared from public view for the last three years as he led a team of technology architects in crafting this vision and developing the basic framework of Landmark. What appeals to me most about Lawson's approach is that it may be leading edge, but it's not something created by Lawson in a vacuum. Domain specific languages (DSL) are commonly used by software developers in other domains. When I pressed for examples, Lawson pointed specifically to discussions they have had with Bell Labs and with another software developer that it was not free to name.

At this point, to Lawson's customers, Landmark just is a vision--although after three years in the lab, Landmark is much more than a statement of direction. The first Landmark application should be introduced within the next year and at that point we should have a good indication whether Lawson is able to deliver.

Offshore delivery of professional services
Lawson announced today that it has put together a "blended professional services" offering in partnership with Indian offshore services provider Xansa. Lawson was already using Xansa internally to provide some maintenance programming for existing Lawson products, leveraging the lower cost of labor offshore. Now Lawson will offer Xansa offshore resources to Lawson clients, to be managed by Lawson personnel onshore, to deliver customer projects such as integration and interface programming, data conversion, and development of custom forms and reports.

Lawson customers will appreciate the lower cost of delivery, as long as Lawson continues to step up to the role of being the "one throat to choke."

Update, May 11. Lawson and IBM have now announced that Landmark will be based on IBM's Websphere technology. I guessed this on Monday when I interviewed Lawson executives about Landmark, but they wouldn't confirm it at the time. I don't know why they had to hold back this piece of the puzzle for two days after the Landmark announcement, except that I suppose it gives them one more thing to announce during their user conference. Basing Landmark largely on IBM technology is a good thing. Application software vendors are notoriously poor at building proprietary development toolsets, and the more Lawson can incorporate publicly available components in its development platform, the better. IBM, which doesn't compete with ISVs, is a great choice.

Lawson says that during the first half of 2006, it will releases upgrades to its financials, SCM, HR, performance management, and procurement modules, with embedded core components of IBM's WebSphere, DB2, Rational, and Tivoli technology.

For more, check out Lawson's press release.

1 comment:

Anonymous said...
This comment has been removed by a blog administrator.