Letters: SMBs & Open-Source Software; XBRL and MDM Defined

What's really holding back SMB's adoption of open source? One reader thinks it's not so much support as it is accessibility.

But Is It Accessible?

In the article "Headsup: SMBs Eschew Open-Source Software" (Dashboard, October 2006), it says: "What's holding back broader adoption? The main inhibitor is a perceived lack of support. 'Folks are saying that at the eleventh hour, when your business has been taken to its knees, there's no one you can call,' says Bob Anderson, Gartner vice president. 'You have to depend on this community to solve your problems. But if the community is asleep at that hour, what happens?' While support contracts are available, their expense often outweighs the benefits of the open-source model."

I suspect Anderson has not actually talked to SMBs about this. I don't think it's a support issue; it's an accessibility issue. Open-source tools can be hard to find, understand, deploy and use without in-house tech advocates pushing for them, and even then, it's not informal support mechanisms that deter business adoption.

The open-source platforms are generally not packaged and promoted (and perhaps not even developed) in a way that makes the business buyer comfortable, particularly with respect to usability. He doesn't have to get comfortable if you're talking about a Linux server (where I think adoption has been high across tiers). He does have to get comfortable if you're talking about desktop apps, portals, BPM tools, e-commerce and so on.

CMS Watch
[email protected]

XBRL and MDM Defined

As someone who has been involved in XBRL since its inception, it is gratifying to see the technology receive some attention from the analyst community ("XBRL or Master Data Management?"). However, there are still a few things to be straightened out about how XBRL and MDM fit together. XBRL is a technology and MDM is a business practice. MDM can be implemented with a variety of technologies, including XBRL. Whether a company should undertake explicit MDM initiatives and which technologies to use are decisions idiosyncratic to each company. Mr. Kugel is correctly vague on this point.

The author is also on target that XBRL scores better as an MDM technology when interoperability comes to dominate other factors. However, this should include intracompany, cross-business-unit, cross-geography interoperability as well as supply chain interoperability. Within the enterprise, another layer of data warehouses is not the answer to the MDM problem in many cases.

I recommend your readers look into the use of XBRL, specifically the XBRL GL taxonomy, which has already built in most of the features people are looking for in MDM solutions. XBRL is capable of supporting the sharing of data and metadata in an agile, service-oriented architecture.

New York
[email protected]

Put Your Brain Into It

Good advice in "Headsup: Should You Speed Up BI? Not So Fast!" (Dashboard, September 2006). It's nice to see recognition that you must think about what BI is trying to achieve, rather than just throwing hardware and software at it.