Does It Compute?

Pushing the Envelope with Worldox

Many databases are structured to process large numbers of operations at once. Define what you want to do and the program repeats the operation until all items are processed, documents or emails created. Could be several hundred. Document Management Systems, in contrast, are set up to deal with one document at a time. So how do you create an equivalent to batch processing? Some operations are built in - moving or copying hundreds of documents at a time (assuming the system allows it, since this could be construed as a security risk).

This has come up several times recently with various clients, and here’s an outline of a process that could be modified as needed.

Step one. The firm receives hundreds of emails in connection with debt collection, real estate or other transactional operations. Worldox cannot scan the subject line or content of an email to identify particular client/matter numbers. However, these emails belong to one general operation, so you can create a specific “matter” to deal with them. Set up a rule in Outlook to move all such incoming emails (which typically come from or have a CC to a specific mailbox) to a particular Outlook folder. For a variety of reasons having to do with Outlook they frequently cannot be moved directly to Worldox.

Step Two. Using the enhanced Worldox email capacities introduced late last year, configure that sub-folder to move emails to a particular Worldox matter. You may have to create multiple matters by date depending on the volume. Worldox starts to have difficulty if you have more than 15,000 or so documents in a single folder.

You can then search that repository by matter number or client name depending on how your system is structured and if desired move specific emails to the main matter for a given client.

Worldox and IT Providers

As a consultant who does only Worldox, I rely on a firm’s IT provider to deal with server setup, general IT infrastructure, etc. Unfortunately many IT firms have some fundamental misconceptions as to how Worldox works, and this can lead to some nasty issues.

The main reason for this is that IT people are looking for an SQL or other server-based database. The key thing about Worldox is that it is NOT a SQL application. Nothing ever runs from the server. This has the advantage that it is extensively and quickly configurable and not very resource intensive.

The Worldox program sits on the same file server that contains your data. Client computers access it; they check for any updates or changes when the program opens, and save any user changes when the Worldox client closes. This means that once the program is installed locally, IT never has to touch end user computers for any updates or upgrades. The client install is run by a command line (something like w:\apps\worldox\wdmirror.exe /I). Worldox clients are NEVER (and in fact cannot be) installed from an *.msi file.

The piece that leads to the most confusion is the Worldox Indexer. Every time Worldox opens or saves a file it also puts metadata information (client, matter, etc.) into stub files in a location on the Worldox program directory. The indexer sits on a separate indexer work station (more about that later) and checks for any stub files every 10 or 15 seconds. When it finds any information it adds it to the main metadata index on the server. Every night the indexer rebuilds the indexes locally (for performance reasons) and then uploads them to the main index on the file server. Indexes are re-initialized weekly so that they are always “clean.”

While many IT firms set up the indexer to run a server OS, functionally it is NOT a server. What defines a server? Information or programs on a server are accessed by multiple users – thus Microsoft sells server licenses that allow 10, 50, 100 people to log into a server simultaneously. Administrators frequently use the indexer to troubleshoot potential Worldox issues, but no end users ever access the Worldox indexer. The indexer is a specialized workstation that logs in to the server (like other users) to perform certain tasks. While it may run server software, in fact it is just a tricked out workstation. So when IT starts talking about the “Worldox server” it is confusing because there is no such animal.

Understanding this structure will save a lot of headaches when setting up and configuring Worldox.

Are All Your Documents Searchable?

The goal of a document management system such as Worldox is to provide the firm with a single electronic repository of all its documents, emails, scanned documents (PDFs) that is completely searchable and indexed.

While such a system is known to break down in a number of ways – non-compliance by users, failure to store emails, etc., there is a “hidden” hole that is potentially much more serious. And that is the failure to OCR PDF documents that have been scanned or received from clients or other firms. When you scan a document, the resulting PDF is an “electronic xerox” – a picture of the original. Unlike a Word document, you cannot search for anything it in, it is just an image. The process of Optical Character Recognition (“OCR”) “reads” the document and creates a text overlay that enables the document to be searched and indexed. Many scanners are set not to OCR documents because vendors levy an additional per-page charge for that function. Studies have shown that for most firms, 20% to 30% of their PDFs are not OCR’d and hence not searchable. I recently ran diagnostics on some clients, and that number seems to be accurate. In some cases the number is dramatically higher.

Not only will users not be able to locate information they need, but in terms of reviewing your document store relating to discovery, records requests, subpoena responses, issues during corporate transactions, etc., the firm could face serious liability if it fails to locate and produce required documents or information. Sanctions due to incomplete/improper document production are every attorney’s nightmare.

There are a number of solutions to this issue, apart from paying your copier company extra money. Some PDF programs (such as Nuance) let you set up a “watch” directory. Scan documents to the watch directory and they are OCR’d.

The solution I prefer is a program called Symphony, from Trumpet, Inc. Symphony sits on your Worldox indexer and OCRs all documents stored in Worldox. It will also OCR your backlog – all those documents that already exist but were not previously text searchable. At $50/user/year (tied to the number of Worldox users) with no page count, the program is extremely affordable. There are also more advanced, somewhat more expensive, options that will greatly enhance your productivity when scanning documents. Check it out at www.trumpetinc.com.

One way or another, all firms need a solution to this issue.