Enterprise Resource Planning

ERP Journal on Ulitzer

Subscribe to ERP Journal on Ulitzer: eMailAlertsEmail Alerts newslettersWeekly Newsletters
Get ERP Journal on Ulitzer: homepageHomepage mobileMobile rssRSS facebookFacebook twitterTwitter linkedinLinkedIn


ERP Journal Authors: Amy Eager, Jason Bloomberg, Automic Blog, Mehdi Daoudi, Steve Mordue

Related Topics: Artificial Intelligence Journal, XML Magazine, ERP Journal on Ulitzer

Artificial Intelligence: Article

Interview with Sandra Clark of Insight Technologies

Interview with Sandra Clark of Insight Technologies

XML-J: Can you give me an idea of how you got started and what the background is? Did the developer start off with XML or did he or she actually have some other technologies in place?
Clark:
No, in fact XMLMATE, the product we launched at XML DevCon 2000, originated after the development of one of our other products. That product, Intellego, is a full-blown application that automates the structuring and indexing of XML.

XML-J: The basic technology your company started out with was XML?
Clark:
No, COM/DCOM. A lot of our work was in COM/DCOM. We were struggling with some of the learning curves of XML. That's really how we developed XMLMATE - we needed rapid development tools to assist us internally. A lot of products originate this way; you develop them for a problem you've got. And we realized that there are a lot of developers out there who would want a similar tool.

XML-J: Are you going to continue in the other technology areas or are you going to concentrate on XML?
Clark:
We'll be in several areas, certainly COM/DCOM. We're in XML now, and we're looking at Java/CORBA as well.

XML-J: So what are you doing with COM/DCOM?
Clark:
We have a full-blown application, Intellego suite, that is an intelligent rules engine with support for conversion from various file formats. Using artificial intelligence, we can extract not only the inherent metadata from multiple file formats, but we can actually create additional metadata packaging by using artificial intelligence and applying business rules. We can then encapsulate an object and maybe attach a digital signature to it, route it into anything from a document management system or reference management system to a B2B- or B2C-type portal. We can export out, having created the metadata in any format such as XML, PDF or HTML. You name it.

XML-J: So you also do the PDF?
Clark:
Yes, we do PDF as well.

XML-J: Let's talk about XML technology, what you've started out with and where you plan to go in the long term.
Clark:
Certainly. XML is the emerging standard from our perspective and we feel we're in a very fortunate position to be at the start rather than the me-too environment. Certainly XMLMATE is an excellent order converter, a very good intelligent editor that allows rapid development for experienced developers, but it also helps shorten the learning cycle for those who are less experienced with XML. We present data in easily understandable forms, such as a table with full procedures, so it gives somebody a chance to see XML code and to utilize that and get up and running very fast.

There are interesting things like the XML binary object storage of any binary information, for example, images and a database inside of an XML document.

XML-J: Are you using the PC data, the XML data tech or are you doing something more than that?
Clark:
We're doing something more than that. I'm not deeply technical. You can tie me up in knots pretty fast.

XML-J: I was just wondering because XML does support a binary format that it can use to transfer data, and from what I've heard of their performance I was wondering if Insight had done anything in that direction?
Clark:
The performance for the transfer of binary object data within an XML object is very fast using XMLMATE, both composing and decomposing. You can send the binary format with an XML object anywhere, and you can restore it at the other end.

XML-J: This is in the XML editing area, right?
Clark:
Yes, XMLMATE is a powerful XML editor, but it's more than XML editing. The focus in XML editing is the structuring of the data. A lot of non-XML data formats can be well presented via XML. XMLMATE is an ideal tool for this research/development. We provide various types of conversion from non-XML into XML with an advanced editing facility. XMLMATE gives developers a quick start as an XML analysis tool. You can send an HTTP request to a Web server that returns XML, then copy the reply to the clipboard and paste it into XMLMATE for analysis of your ASP/CGI codes. You can run SQL query, get results into XML, view the structure and get your own ideas about how you would like this to be done. Or build an XML Document Type Definition (DTD) and use it as a template for a blank XML document. XMLMATE as an XML editor helps shorten the development cycle. It did for us. We had the same problems as everyone going to XML.

XML-J: Okay, the flagship product you have is XMLMATE?
Clark:
Yes. XMLMATE is the flagship product in our XML range. It's $149 for the professional version so it's geared toward the individual developer, but we also have corporate pricing. We're looking very much to the corporate server market.

XML-J: I was trying to lead to that. As an editor myself it will be hard to sell products. I was wondering who you are partnering with or planning to partner with?
Clark:
Our whole strategy as a company is to produce products suitable for OEM and as toolkits. First are XMLMATE and Intellego, these types of products. We started with Intellego and are in serious discussions with several major software corporations regarding this product. We've been approached for XMLMATE as well because obviously people are doing lots of things like file conversions, and they want a tool that they can OEM into their product.

XML-J: I can see you being approached by some of the XML server vendors and the apps server vendors, even as acquisition targets in the event you have something they could include in their own tool offering.
Clark:
Absolutely, and that's our aim. Trying to be a direct sales company is not the way to go in my view. Obviously, people will download it off the Web, but we are targeting the OEM area of the market also.

XML-J: Are you making any headway there?
Clark:
I think it's too early to name any names. However, we're very pleased with the quality of companies that we're talking to. Obviously, we will be very grateful for any extra help we can get from anybody. It's all about networking and the show actually was an excellent forum for that.

XML-J: Yes, I think so.
Clark:
We certainly are going to do some advertising as well.

XML-J: That's why we're trying to do two or three interviews a month so that vendors can get some exposure through the magazine.
Clark:
It's certainly very appreciated. We're very excited. When you launch a new product you're never sure how well you'll be received, but we had several hundred people actually have demonstrations and the response from those people was excellent, so we were very pleased.

XML-J: I received an e-mail with a link about www.XMLBoutique.com. How did that name come about?
Clark:
We were looking for a good domain name. I don't need to tell you how hard it is to find a dot-com name that will indicate your subject matter, and we didn't want to limit it just to XMLMATE. We feel XML Boutique is a Web site where other XML products would be available. And that is the intention - a boutique, a store. That's really how the name came about. So we see XMLMATE as the first tool set.

XML-J: I understand this is an editing and conversion tool. Do you see this spreading into overall Enterprise applications, to B2B applications?
Clark:
In B2B applications?

XML-J: Someone like WebMatters, for example, because they have the adaptors to extract data from ERP resources. Where would you see yourselves playing in that area or, for that matter, eXcelon or Orion? Some of the XML server vendors that specialize in extracting the data as opposed to creating, you are creating the tools but obviously that's not the area in which other vendors are concentrating, to be able to process that data. Where do you see your product fitting in?
Clark:
Certainly Intellego is probably well suited for that because that's what it does. It takes any data source and parses the data through several times. It has an inference engine and uses artificial intelligence, and then you can route back the extracted data into any application. It's excellent for taking any form of data and applying more intelligence to it, then passing it through to another application or directly into a Web publishing system, or whatever you want to do. Our system is different.

The strength is in the fact that Intellego utilizes a declarative rules engine that allows more than one solution as the output of a business rule. The declarative nature of the engine enables it to read, load and apply rules to incoming data, generate new rules or conclusions based upon processed information, take necessary actions and output the result into external applications. When we looked at the marketplace, there are people who are doing niche parts of what Intellego does, like e-mail extraction.

Companies out there are looking at the vertical market, such as the legal market. We've looked at it as a big picture. Again, as with most development, it usually results from a customer requirement. However, we built the product because we have a lot of experience in software development, and there's nothing worse than developing a product around a custom application and trying to "productize" it as an afterthought.

We built it from the ground up with the knowledge that we would then take this product to market. And that is very, very key.

XML-J: Tell me about this rules engine now. What kind of capability does it have and how do you see it being used in a business application?
Clark:
Intellego Suite is an intelligent rules engine with built-in support for various data conversions. It was based upon a logical model of information feeders and we consider it an intelligent information filter that enables the supply of filtered information to the search engines and record management/document management systems. The artificial intelligence lives in the rule processor, which is based upon Prolog language and language extensions, developed by Insight. One of the language extensions is XMLProlog, which allows the handling of structured data in an efficient way on a symbolic level. XMLMATE was built as a tool to help with XML data processing inside XMLProlog.

The Intellego rules engine has built-in functionality to parse various data types into XML and output results into a variety of file formats, including XML. Most organizations are having problems with applications like document management, record management and knowledge management systems. They've got portals, but what they don't have is the ability to create the metadata automatically and to apply intelligence to that.

XML-J: You know that most of the vendors that actually have more products, they usually have that on their tool and others so how do you feel this is distinguished from these tools?
Clark:
I think it's the fact that from our research we developed a piece of technology that's a missing link that other vendors don't have and we've been looking at some very complex applications that are starting to come through. As people become more knowledgeable, applications and requirements get more complex; for example, the longevity of a record is now a key issue. We can encapsulate a record with metadata to create an object with business rules attached to it. The object encapsulated may contain thousands of documents. We can then attach a digital signature to it for long-term preservation of the record, independent from any application. This would be a typical application. Intellego takes data from a variety of feeds, so you could have a news feed coming in and you could have legacy database information, Lotus Notes. It's across the whole basis, as opposed to people who just specialize in Web documents or Microsoft Exchange.

So we're looking at a much bigger picture. And that, we believe, is our strength and certainly, from all the research we've done, nobody is doing quite what we are doing. This is what makes it very attractive.

XML-J: About standards. You said Insight was also thinking of moving into that world. Do you know what kind of migration you plan to do there?
Clark:
We've built our software to adhere to industry standards in such a way that it will enable us to move the base architecture into other environments. From my perspective in Insight, I'm making sure we're going to support the most important environments. Therefore, the whole Microsoft COM/DCOM is one thing, but a lot of people out there are going the Java/CORBA route. Therefore we want to be able to offer our applications for that environment also.

XML-J: When do you plan to do that? What is the road map?
Clark:
Time scale. We'll be looking at Java/CORBA within the next couple of months. We're probably looking to release the CORBA version of Intellego toward the end of the year and futher versions of XMLMATE over the next few months.

More Stories By XML News Desk

The XML-Journal News Desk monitors the world of XML and SOA /Web services to present IT professionals with updates on technology advances and business trends, as well as new products and standards.

Comments (0)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.