[From] Robert Marzullo on Oct 22 1997 3:11PM EST
Steve , if you don't mind, can you gives us an insight of what you e-mailed Gary?
TIA
Bob
Bob,
I can do better than that. I can send you the original (edited slighly to make it clearer).
I am amused by the fact that I was virtually accused of withholding it from the SI thread, yet until your message today, no one ever asked me directly to post a copy.
I do not pretend to understand the fuss. There's not much new here, except perhaps for my uninformed and a priori comments re the Internet version of the telephony db Express-A-Bill product. The letter to Gary is mostly a summary of my earlier posts and opinions, all of which are available on the thread archives or through my profile history.
I certainly have no insights as to what's going on with this company today. I merely have some understanding of the telecommunications marketplace, and try to share that with others on the thread and via e-mail when I wish to make specific and/or personal comments re another's SI post.
I originally decided to respond to Gary's post #7112 to give him some background, but then decided to expand it as a way of tieing together much of what I've said earlier regarding the product and company. I've expanded it slightly for the SI thread. IMO, it's not especially interesting and borders upon but doesn't actually get technical, but for those of you who've missed it and somehow still care, here is my take on the db Express-a-bill product.
[Start of letter]
Gary,
Let me begin by taking one of your statements ("As I stated above, the best and most pre-tested technology and products have, historically, needed vast sums of cash to perfect them, after delivery.") as a point of departure.
You are generally correct, of course. I would like to posit some additional specific points in defense of your general argument. You may already know all of the following, in which case I apologize for wasting your time. However, it may be the case that one or more of my comments may add to your perspective, if nothing else.
Moreover, as background, I have evaluated the last beta/first general release (a misnomer, since the carrier-based product (e.g. a BT version) is carrier-specific and thus not consistent with a general-release product) of db Express-a-Bill, so I have some first-hand knowledge of the product.
I would like to be specific about one thing: I have no point in stating any of the following except to provide you with some additional background which in some small way may enable you to better continue your public SI thread course of rumination and dialogue.
My general points are these:
1. The historical db Express product is not the telephony product and the telephony product first announced is not the telephony product being promoted by the company today.
2. The Internet is adding product development demands (and costs) of its own, and will stretch delivery timeframes even further. Much of the latter part of this message treats this issue.
1. The historical db Express product is not the telephony product and the telephony product first announced is not the telephony product being hyped by the company today.
1A. The historical db Express product is a data visualization tool that works in conjunction with other databases. The db Express-A-Bill product is a call record and call pattern recognition tool that works with a specific long distance telephone company's specific billing format and specific set of telephone bill data (not all phone companies provide the same information and certainly not in the same format). It works with whatever basic database or sort capability the telephone company decides to supply with its call records. The database itself is inaccessible and not relevant to the db Express-A-Bill reporting function
1B. There is another version of the same product, with the same name (these people are not role models in either product planning or marketing) that is purportedly designed for the end-user market. I am not sanguine about its prospects and, indeed, it is not being discussed by the company anymore.
IMO, no one in the company's product management or sales team understands the end-user market and its different distribution channels, so it is just as well they are focused elsewhere. OTOH, again IMO, they are missing some incredibly large synergistic channel and Intranet opportunities.
1C. The telephony product first announced (in addition to the CPE version identified in 1B) was designed to be delivered to a customer on a diskette (or CD ROM), presumably by that customer's long distance company. That long distance company would then send the customer the latest month's billing on diskette or CD, and those records would then be retrieved and analyzed by using the db Express-A-Bill software
As an aside, if you knew anything about marketing would you call your product "db Express-A-Bill? And then release two different versions --three, if you include the client-server model identified in 1D-virtually simultaneously, give them all the same name, and then talk about them interchangeably in your press releases and backgrounders?.
This telephony product was the one that a number of rebillers (fulfillment companies to whom long distance companies turn over their long distance call data to be rebilled to customers in the respective LD company's name), including one named Profitec -- the subject of a major CCEE press release-- was to be using as its end-user package.
1D. A second telephony product was one (Lee had some acute observations about this -- it's more in his area, I think-- if I recall properly) that utilized client-server technology and ran on Windows NT. The premise here was that the long distance company would sell the client-server software to larger customers who would then put the full set of call records on a server and enable managers throughout the company to access the departmental information for which they had P&L responsibility over their company's LAN. I do not know what is being used for the database itself, but there are no sophisticated requirements here. The reason for the product to exist is because it takes advantage of db Express' data extraction and mining techniques (such as they are; I won't get into discussions about the accuracy of these terms), not because such software marks a departure from competitive offerings.
1E. CC then migrated the product to take advantage of Internet technology, never changing the name or widely announcing a new release number. Apparently, they think Microsoft makes money by selling DOS and Windows and not Dos 6.2 and Windows 95.
The Internet version is conceptually the same as the client-server version, except the call data remains on the telephone company's Internet server and is presumably accessed by using a standard browser, although I have never seen this version, so I am just making an reasonably informed guess. Access to the call data is then subject to whatever controls the subscriber's IT department puts upon employee Internet access. This is valuable and desirable to the degree that the number of call records is so large that obtaining them and manipulating them is problematical, which is certainly the case at large companies. What the product does is enable the individual managers to just access that part of the information (i.e. specific call records and data) in which they are interested at any particular time, and to produce standard, monthly reports.
[The following observation is new to this posted edition] There are two distinct conceptual advantages to an Internet version of the program as opposed to a client-server one. The first is the issue of scalability: the transfer of millions of call records and their manipulation in a client-server environment can cause difficulties when the LAN is already congested or when the client is at a remote site served by a relatively low speed WAN link. The second is the need to produce software for a variety of platforms in a client-server scenario versus the cross-platform nature of the Internet.
BTW, for those unfamiliar with developing products for the Internet, many of the tasks involved are not trivial. Not only must the developer be able to handle multiple browsers (Netscape and MS Explorer have the lion's share here, but not necessarily in Great Britain), but also multiple versions of browsers with widely differing capabilities. If you want to use frames, for example, that's OK, but you're stuck with the 3.0 versions of Netscape and Explorer. Java and Active-X present specific problems as well. They can be dealt with but the software has to interrogate each customer's PC to determine the browser and version being used, and then it must have been decided in advance how to deal with the particular configuration. Dealing with larger customers is easier because the IT department can set browser standards, but when one tries, as BT is presumably doing, to make the technology available to everyone, all the issues identified above and more arise and in spades.
1F. One main point of the foregoing is that these changes have taken place over the past two years, more or less, and that they have cost money. And the end result is that CC is dealing with a more sophisticated product with greater technical demands and a much higher cost factor. Many technology companies face this, but the successful ones are able to market product in volume while continuing product development. IMO, the result of CC management's failure to perform here is reflected in the current number of shares outstanding, the price of the stock today (reflecting financial performance to date), and the just-floated filing.
2. The Internet is adding product development demands (and costs) of its own, and will stretch delivery timeframes even further. Much of the rest of this message treats this.
2A. The Internet has not stood still. The recently-released Netscape Navigator 4.0 and the just-released Internet Explorer 4.0 again raise the ante in this unending technological poker game. The 4.0 browsers offer developers some unprecedented opportunities to facilitate communication and information, but require programming and data structure knowledge not necessarily possessed by today's static HTML programmers.
2B. BT could develop much of this capability (since it will be used on BT's web site, after all), but not all. Much of the server vs. client side implications of the 4.0s are not yet clear, but the concept of data binding, for example, is illustratively relevant. Data binding would allow a manager at a BT customer to create a report by performing an ad hoc query over the Internet, and while the report was being transmitted, the data behind just that one report could also be transmitted. Now, a great advantage of db Express is that it allows for multiple visualization of data (filescapes). If CC plays its cards correctly, db Express' previous strong point of being able to retrieve relevant data very quickly could be augmented and optimized by enabling data binding so that the full set of limited data required could be retrieved and transmitted while the manager was looking at the first report. The only thing better than fast retrieval is instantaneous retrieval and DB express in conjunction with data binding would allow this (making further queries upon the same set of data all non-Internet client-side functions) while still allowing db Express to retain its positioning by still being able to retrieve in desired report format any limited set of data faster than its competition.
2C. Will CC take advantage of this? Are they even aware of it? I dunno. It's hard to underestimate them. But my data binding example should show that the development game is perpetual and that it will take big bucks to play. Even if it is BT and not CC who does most of the development, both sides will have to invest something, and product delivery may take even longer as a result. If CC has not been able to make money in traditional software development, it will certainly need to suck up more of the green stuff now.
BTW, for those interested, I recently moderated a session on the two 4.0 browsers. See <http://www.ipn/org/programs> for details; the presentations (one Astound and two Powerpoint, I believe) should be available on-line by now.
Gary, thanks for reading through this. I hope you found something of use in the foregoing.
Steve Bergman |
