Data Management

CAD Manager's Newsletter #113 (August 26, 2004)

25 Aug, 2004 By: Robert Green

Way back in 1990, I was working with a client company that started a U.S. office and immediately became overwhelmed by DWG files provided by customers, contractors, and vendors. The deluge of electronic files -- actually, no more than 25,000 in this customer's case -- was a coordination nightmare as users just copied files into folders and zipped them to tame the file-management beast. Using the mighty 386-20 computers with a whopping 4MB of RAM, we continued to create drawings using DOS-based AutoCAD Release 10 and continued to crank out more drawings that we simply couldn't keep track of.

It became clear to me then that there should be a way to manage CAD files that was network based, reasonably automated, and wouldn't get in the CAD user's way. In the early days, we had only 10-CAD (an early precursor to AutoEDMS) and Cyco's AutoBase (an extension of its AutoManager viewer) as possible solutions. Even in 1990, it was clear to me that document management for CAD would become ever more required to do business in an electronic world.

Fast forward 14 years, and we see that the amount of CAD data we can produce today dwarfs that initial experience, yet document management is still very absent from the landscape. We can now move files easily, yet we frequently don't know which files we need to move or who has the most recent versions. We've got better computer platforms and tools, yet we still dump files into folders and ZIP them just like we did when computer networks were new.

Have we really progressed at all?

If your company hasn't implemented some sort of engineering document management system, you're simply running on borrowed time. So many technology and business factors are converging in today's market that a CAD manager can no longer "know where everything is."

Allow me to justify my statement by outlining some of the key technology and business drivers I see that make document management critical:

* Branch/remote office work sharing
* Telecommuting workforce expansion
* Remote access to/from vendors and subcontractors
* Real-time data sharing for traveling staff
* Digital manufacturing/prototyping file sharing

These trends have been taking shape for the past five years, but are now receiving renewed interest because technology investment is up, globalization of businesses is on the rise, and communications bandwidth is increasingly cheap. Every one of these trends shares two components that point to the need for document management:

* The ability to move data to other businesses, branches, or suppliers without data reentry, which makes operations cheaper.

* The ability to work with a smaller core staff, using less office space, thus radically reducing staff-related costs.

Trust me when I say that business trends that save money and make staffs smaller are not going to go away. Also believe me when I say that as core staffs get smaller and remote staffs spread out more widely that your document-management burden goes up exponentially while your proximity to the documents continually decreases.

The trends all point to an inescapable conclusion: You simply can't dodge the need for document management.

Document management should accomplish these minimum objectives:

* Organize your files (names, paths)
* Secure your files (against loss or unauthorized access)
* Build a searchable history (archiving, locating, printing)
* Control work processes (approvals, revisions, and so forth)
* Extend the reach of your files (to those outside your walls)

Document-management applications are typically a collection of software applications that allow you to customize a system for your particular needs while meeting the above objectives. Because document-management systems integrate with networks, Web servers, and software applications, you might guess that they are complex, expensive, and require a lot of implementation effort -- and you'd be largely correct in assuming so.

Given the complexity and expense of implementing a document-management system, you may be tempted to just continue trying to manage files using network directories and manual procedures. Chances are you've tried to organize your files before and found it wasn't always easy to get people to follow the rules, right? You may also have procedures for revising/approving files, archiving old jobs, and managing markups -- all of which present the same problems, right?

The bottom line always seems to be that manually controlling electronic documents is fraught with danger in the best of circumstances and can be almost impossible as the number of remote factors in the management equation increase.

Before we can analyze the available document-management systems, we have to define some terms that will allow us to make intelligent comparisons. Because there really isn't any industry-standard terminology for document management, I'll begin my list of terms here and finish them in the next newsletter.

Database engine. Every document-management system I've ever worked with has at its core a database. The database simply contains information about the documents in the system and facilitates database-intensive functions such as searching, report generation, and transaction logging. Database technology in current systems ranges from the old DBF file format to relational engines like SQL or Oracle, or can even be proprietary databases that are not industry standard at all.

Note: The database engine is the component of the document management system that your IT staff will care about the most!

Software interface. How will your users interact with the document-management program? Via a web browser, a proprietary software package, a Windows Explorer-style interface? These are the normal options for available software, and each has its own strengths and weaknesses.

View/Print/Redline tools. These utilities are almost always available with document-management systems but can sometimes cost extra. Functionality, especially viewer functionality, can vary widely from package to package. In CAD-heavy environments, proper viewing and printing of files is critical and merits detailed testing.

Collaborative tools. Document approvals, document review lists, email support for messaging, and even document check-out and check-in capabilities are becoming increasingly common in document-management software platforms. Collaboration tools vary widely from system to system and there's no market unity regarding which tools should be standard, so you must shop carefully to locate the right features for your company.

Customization tools. Typically, integration with Visual Basic or some form of scripting language allows the user to customize the system in varying degrees. Again, there's no such thing as a standard in this area, so you must carefully evaluate each system's customization tools.

I would like to thank the 23 readers who sent me detailed proposals on additional topics for the CAD manager surveys, as requested in my August 12 column. I received quite a bit of general feedback, but the highly detailed suggestions from CAD managers in the trenches have proved extremely valuable.

I will be developing and testing the CAD Manager 2004 Survey over the next few weeks and will give detailed instructions and a Web link to the survey in the next issue of the newsletter, due out September 9. In that same issue, I'll continue my coverage of document management systems. Until then!