The Inovis Blog

Being a software developer who tries to stay ahead of the bleeding edge (presumably caused by paper cuts), I’m always keeping my eyes peeled and an ear to the ground to keep up with the latest of the latest trends. Admittedly, it leads to bad posture (common to many a software developer), but it has served me well over my something-mumble-something years in the industry. A lot of what I hear, of course, is rumor and a large percentage of that is quite spurious and not worth repeating. However, I’ve recently heard not one, not two, but three different rumors all of which seemed very exciting to me (and at my age, that’s saying something—I’m not sure what, but something).

As some of you might know, I’m an active participant in the X12C subcommittee. Someone recently let the cat out of the bag concerning the last trimester meeting and discussions on future directions, but even in such meetings, the official discussions may not be the whole story. (Reminder: ordinarily I don’t repeat rumors but this was too intriguing to suppress.) While picking up coffee during a break, I found myself in conversation with a colleague who let slip that a “skunkworks” project is underway to produce color-coded EDI (or ccEDI). The most obvious way to use ccEDI would be to color code individual data elements to provide further clarification (particularly in the area of semantics) and, of course, improve readability all without increasing the size of the data. In addition, ccEDI color coding can be used to personalize interchanges and, probably in its second release, permit the use of wallpaper patterns. Proof positive that EDI not only lives, but lives well.

On the XML front, I recently heard about the latest attempt to make data transmission more compact. The Small Minded Thinkers Society (SMTS) is planning to propose a new project for SML (Small Markup Language), a new XML-based vocabulary that would restrict all tags and data values to a single character (not counting whitespace). This is an example of a purchase order in SML:

<P>
<O>1</O>
<L>
<I>a</I>
<Q>7</Q>
</L>
<L>
<I>b</I>
<Q>2</Q>
</L>
<S>S</S>
<V>b</V>
<E>5</E>
<T>9</T>
</P>

Having also participated in UN/EDIFACT’s Joint Technical Assessment Group in the past, I am aware of a similar project that was attempted a number of years ago by that group. Sadly, that effort fell victim to internal political problems.

Turning back to X12, I’ve recently learned of a proposal to merge the best features of EDI and XML. “It was one of those bop-on-the-head wow-I-could-have-had-a-V-8 things,” an involved X12 member told me (on condition of anonymity). The new project, as yet unnamed, will provide the compactness of EDI with the meaningful tag naming capability of XML. The following is an example of how the data would fit together:

<ediInterchange>
ISA*00* *00* *ZZ*EXAMPLESENDER01*ZZ*SAMPLERECEIVER1*080401*0247*U*00402*000087592*0*P*^~
GS*TX*STM*STM*20041019*0247*57578*X*004030~
ST*864*00057708~
BMG*00*00057708~
N1*UC*XDTS~
MIT*sdm01000*stmdat~
MSG*This is an example of~
MSG*the rumored functionality~
SE*6*00057708~
GE*1*57578~
IEA*1*000087592~
</ediInterchange>

My source implied that there was a similar development effort underway within UN/CEFACT, but word is that they plan to use a different naming scheme for the XML tags. No word as to whether this XML-based EDI will be able to support ccEDI.

To reiterate one last time, I don’t normally repeat rumors, but I think, if they are true, these represent some pretty significant developments and I had to share them. I’m very interested to hear what others think.