Talk:CA-Telon
![]() | This article is rated C-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | |||||||
|
The original draft of this article was created as part of a class project for the University of Washington. The following are transcripts of phone and email interviews conducted with some of the people who were involved (and are still involved) with the creation of TELON.
Interviews
[edit]Phone interviews with Don Christensen
[edit]Nov 4, 2006
[edit]Before Christensen Systems
1. Explain briefly what Telon is.
An application generator for COBOL & PL/I for IMS and CICS for transaction driven systems. Caveat: not quite an app gen because it didn’t gen the whole app. Primarily gen’d online interfaces for the apps and DB access. For 3270 only. Did not assist in building business rules; gen’s front and back end tiers in 3-tier arch. Assisted in prototyping/testing interfaces by running under TSO. IMS+CICS were not well designed for testing. Gen’d IMS+CICS control blocks. Wasn’t whole lifecycle. Missing front-end data design (E/R diagrams). Back end CASE tool.
2. What were you doing that led up to Telon?
At LM for 2 years starting in ’77 as an IBM consultant (Systems Engineer). Late 70’s IBM 3790 installed at Liberty Mutual in Boston and Christensen was involved in that upgrade.
3. In our class we’ve discussed how many innovations were created by someone who combined an understanding of the problem with technological experience. For example, Blaise Pascal watched his father’s duties as a tax collector and later built a mechanical calculator to help him. Is there a parallel with Telon? Did your work leading up to Telon give you an understanding of the problem that Telon tried to solve?
LM bought a app package for claims processing. [Chris would know the name.] To customize, several forms had to be completed for screen interaction (a form for input, one for output, etc.). Used awkward macro lang for the customization. Christensen wrote a single form using macro lang to gen package forms. Made the leap to generating COBOL apps as well. Christensen started thinking in terms of software engineering (how to automate software design) but began by solving the short-term problem of 3270 interfaces—the complexity of software was in the screen interfaces, not the business logic which was simple at the time. Telon would go on to solve the screen problem. Christensen didn’t think he understood the overall engineering problem as well as he would have liked. Telon allowed the programmer to design while thinking of screen flow instead of IMS/CICS transaction codes, which were cumbersome.
4. What/who were the competitors? How were other people trying to solve the problem?
IBM ADF - Informatics Mark IV – a report generator. Informatics Mark V – Online version of IV Telon had two advantages: 1 - Telon was more productive than the others. 2 - Telon generated COBOL, which made potential customers more willing to take a chance.
5. What were the first steps of developing Telon?
Christensen left IBM to become a private consultant to work at LM (’79) on insurance claims application. First 1.5 years was analysis/design phase of new claims app. During this project there were arguments over workflow. Christensen proposed that he could build an interactive prototype under TSO that would test screen workflow in a few weeks. Christensen, working on his own earlier, had created a painter written in assembler. The painter would allow the user to map out the description of the 3270 screens. The output of the painter was then consumed by another program written in IBM macro language that generated COBOL suitable for TSO. Christensen used his system to build the prototype. The prototype was tested by claims processors to verify the workflow and adjustments were made. This proved to be a very successful way for LM to design the app. After the prototype, LM put out a bid for the hardware and it was won by DEC so the generated COBOL system wasn’t put into use (although the screen flow was used.) Christensen then proposed using the generator on another project at LM but it was considered too risky by LM management. Christensen then went looking for another opportunity and was put in contact with Phil Stymfal who worked at The New England. [Check with Phil for position/title.] Christensen demo’d his new system to Stymfal who immediately saw the advantages and was willing to push the technology at The New England. Later on, as Telon became more accepted, Stymfal’s contacts through The New England Productivity Group [check with Phil] brought in more potential customers. Because LM had provided the computer time during the original creation of the painter/generator, Christensen had worked out a deal with LM that LM could use the product in the future for no charge. The first 4 companies who bought Telon (known as ADS – Application Development System) were The New England, Raytheon, John Hancock, and then LM. All in 1981 and all (except LM) for $30k. The original claims app that Christensen had prototyped at LM had failed on the DEC equipment and was eventually written using Telon. CSI was created after first sale to The New England.
6. Was there ever an “a-ha!” moment?
No particular moment, but of a gradual realization that it would work. The John Hancock sale was the culmination.
7. What technical decisions were made and why?
Mostly what Christensen was familiar with. The Macro lang was built for generating systems programs so it worked well for generating COBOL. It was an interpretive lang. TSO was used because it became the development environment and was a necessary precursor to Telon.
8. What was the first application written using Telon?
Claims processing at The New England. [Check with Phil.]
Nov 6, 2006
[edit]CSI
1. What happened that made you decide to start your own company?
Decision was made to leave IBM (in 1979 after 12 years); some dissatisfaction with IBM. Starting a company was not the driving force. Wanted to pursue something that he enjoyed doing and leaving IBM was the way to do that. Incorporated in spring, 1981 when it looked like a deal was going to be made with New England.
2. Who was your first hire and what was his role?
Chris McNeil in summer of 1981. He continued to work on the app (painter and generator).
3. What was Phil Stymfal’s role?
Phil’s contacts were instrumental in the first few sales of Telon and was an important reference. Stymfal was hired in early 1982 as Director of Development. Bob Giel (classmate of Stymfal) was hired after Stymfal as Director of Administration(?) to make the business/sales decisions. Larry DeBoever was hired in early 1983 as VP of Sales/Marketing.
4. When did the product acquire the name Telon?
Early 1982; previously it had been called ADS (App Dev System). Greek words were preferred. Telos was the original idea but it was already taken by a software company in California. Telon was a contraction of Telos and Eon.
5. How was Telon expanding? What new problems was Telon trying to solve?
Added functionality. A PL/I version of the generator was created; there were a fairly large segment of IBM customers using PL/I. Creating PL/I macros was relatively easy, giving Telon basically a monopoly on PL/I shops. CICS generation support was added. 3 partnerships were created with European companies to sell Telon in Europe, which was quite successful (roughly half of the sales of Telon were international.) Telon Design Facility was created using Telon itself. Mediocre report writer created. Some development was done on New England’s computers, some time-share was bought from a route 128 company.
6. At what point did you realize that Telon was going to be successful?
The first 4 customers were very successful; all very satisfied. The sale to John Hancock was especially huge.
7. What were some major dates? (E.g. releases, customer contracts, new hires.)
Office opened around mid-1982.
8. What mistakes were made?
Because of Stymfal and luck, the first four sales of Telon were relatively easy. A false sense of the sales cycle caused Christensen to be less aggressive in going after new sales. Telon did not address enough of the development problem; it was a good tool for developing the interface but didn’t help enough in the overall business design.
Nov 7, 2006
[edit]Pansophic
1. What caused CSI to sell Telon to Pansophic?
Telon sold roughly $2 million in 1983 with 24 employees. It was determined that there wasn’t enough money to keep up with competitors [Who?]. There were also concerns that CSI did not have the management skills to make Telon successful. The decision was made jointly by Christensen, McNeil, Stymfal, and Giel. Pansophic bought Telon for $2 million and royalties.
2. When was the sale completed?
Fall of 1984.
3. What direction did Telon take? What was Pansophic’s influence?
Pansophic’s influence was mostly in sales and marketing. Al Seyler was a big component of that.
4. Were there any new competitors? What advantages did they have over Telon?
[Bob Giel would know best]. Some CASE tools running on the PC with graphical capabilities. [AGS?] Telon delivered better productivity. Generated COBOL, which was safer. Pansophic had the better sales teams.
5. What were the long-term weaknesses of Telon? Was it possible to correct them or was it necessary to start over?
It didn’t start the development process soon enough; it didn’t help enough with the analysis and design. It was addressed data in its physical form (IMS/DB, SQL, and network DBs). It didn’t abstract data to help with analysis. The key characteristic of Telon was both its key advantage and weakness—it generated solutions within the current programming physical data paradigm.
6. What were the weaknesses of the report generator?
It didn’t deliver the same level of productivity as the online product. [Phil might have a better response.]
7. What was your role at Pansophic?
Director of App Dev + Support. Making some directing in tech decisions. Technical support of sales/marketing. Traveling nationally and internationally presenting at seminars and talking to customers and prospects.
8. Why did you leave Pansophic?
The 3-year commitment expired. Pursue Zeidon.
Email Responses
[edit]An email was sent out to various people involved in the early days of TELON. The following are the various responses. No attempt was made to fix typos and errors.
Gig Kirk
[edit]November 10, 2006
[edit]Sounds like an interesting project. Good luck with it. Since Telon is still a going concern, you'll want to be careful when describing how Telon works and what it does. It's probably best to stick with whatever public collateral that CA makes available as any other information could be considered proprietary information or trade secrets.
I started in September of 82 (best guess) reporting to Phil. At that point, I believe only Don, Chris, Phil, and Bob were employees of the company. There was another employee that had come and gone by that time as I recall. I wish I had a picture of the office from that day as we had a large room furnished with two tables placed end-to-end and two 3270 monitors. As silly as it looked, it is still one of the most exciting times I recall.
I think the real keys to Telon's success were timing and Pansophic investing a significant amount of capital in getting the message out. At that time, we were competing with fourth generation languages like Mark V and provided superior productivity without a 4GL runtime environment. The simplicity of the solution with minimal risk was an easy message to understand. Initially, the price of a Telon sale could be justified by a single project so it did not always need to be a strategic decision, though we tried to position it as a strategic product. When Pansophic tested the elasticity of the price, the market didn't even blink. I believe the price of the product was tripled or quadrupled in a very short period with no impediment to sales.
I believe that the risk involved purchasing a strategic product from a small company is a huge impediment. Finding champions to go out on a limb when the company might not be there tomorrow is too much to swallow for all but the most cavalier of people. Having been through this process three times, I still don't have a good answer for addressing this issue. While some initial sales were key to keeping the lights on, the Pansophic acquisition is probably the most noteworthy event to getting the message out and guaranteeing success.
Bob Giel
[edit]November 24, 2006
[edit]I hope the information in the attached word file helps.
I am leaving town tomorrow and will not be back for a week or so.
Off the top of my head, I could not remember the names of the two major TELON competitors and I could not remember the major IT consulting firm from the west coast that acquired Computer Partners. Also, I could not remember the last name of Bill, the Executive Vice President of Pansophic who convinced Don to sell, and who later became President.
I think everything else here is accurate. But it may be more detail than you want to use in a technical article on TELON.
Clearly my perspective is different than the developers. I believe that TELON was a market place success because it had a better Sales force. It worked well, it met a need, and we had a strong pre sales technical support staff. But two other products worked nearly as well, and for some customers better, than TELON. The difference was the superior distribution.
Clearly in the early days, the difference between TELON and the early competitors was a better product and a personal commitment by all of the Christensen Systems staff to do what ever we had to satisfy the customer. In return we got a level of customer loyalty and commitment that was necessary to get more customers and to sell the product internally in large companies like Goodyear, Baxter Travenol, US Steel, General Reinsurance, and other early adopters.
What previous technical or business experience helped you make decisions regarding TELON?
I had been the financial manager for Sales and Marketing groups at Data Terminal Systems, a fast growing cash register company, which had been spun off from the Digital Equipment Company. I had been modeling new product introduction along with instituting financial and budgeting controls in the Sales & Marketing departments.
When did you start working on TELON?
In January 2002, Phil asked me to meet with Don to discuss the future of Christensen Systems. Phil was trying to decide if the company was too risky to join. At the time, Don assumed that since the first three customers had purchased ADS (TELON) after a two or three month trial, that all customers would make a similar decision.
Don stated that Christensen Systems could grow from the $110,000 revenues in that first year to $7,000,000 (100 systems sold at $70,000 each) in the second and $70,000,000 (1,000 systems sold at $70,000 each) in the third. When I told Don that that was not realistic, and that if he got anywhere close to that rate of growth, that no one working in the company would have an intact marriage, Don stated that the company was “not worth even a single broken marriage” and he asked me to help him build a business plan.
I started evening meetings with Don in February, using the Apple II with extended memory at my DTS job to build the Visicalc models that became the basis of our financial model. By May, Don asked me to join Christensen Systems, and I agreed in July to join as the Director of Finance, when the company had enough revenue and cash flow to pay me.
Don hired a field support person in June, but when I first met him, it was clear that this person was expecting that he could shift to development after a few months in support, and that he would be given a substantial portion of stock as the fourth person in the company. When that field tech left within three months, Don asked me to take on all of the personnel responsibilities as well.
When I was able to arrange to be laid off by the cash register company in July, I started to work full time at Christensen Systems, and finally came on the payroll in October 2002, after Gig had joined us.
What decisions did you make and why? In hindsight, were they the right ones?
The first decision we had to make was creating a sales capability. We started to look for a VP of Sales and marketing, but it took more than a year to find a candidate that would accept an offer.
We used Phil’s network of Insurance Companies to secure trials at The Hartford and at Travelers Insurance. These started in the summer of 2002 and ended in sales for $50,000 each by September 2002. By the end of 2002 we had introduced the CICS product offering, and had secured several other $75,000 IMS sales for a total 2002 revenue of $850,000. Where we had a personal connection we were able to secure a trial, and we seemed to get a sale for every trial within 90 or 120 days. However we were unable to even get a hearing at any companies that did not know Phil. And we were hearing that many of these companies believed that TELON was a different class of product because it was so much less expensive than Mark V or the IBM offering. So to give TELON more credibility we decided to raise the price for the IMS version to $95,000.
Meanwhile Computer Partners, which several years later merged into ???Systems, had seen TELON at Raytheon and at New England Life. They wanted to have the exclusive right to bring TELON into customer sites like the City of Boston as a development tool. They believed this would give them a competitive advantage when trying to sell their development and maintenance services. In exchange they agreed to help us sell TELON directly to their clients, after the first successful project, and they agreed to provide up to $100,000 working capital that was to be paid back in a year.
Because they had the perspective of a body shop, selling consulting services rather than Tools, there was always a difference between how Computer Partners wanted TELON to develop and how our other customers wanted from TELON. This tension crystallized first in Baxter Travenol, which was initially a Computer Partners client and which later decided to buy TELON directly. Computer Partners received a Sales commission on the sale, but was not happy that Baxter wanted the product to be used by other consultants after they purchased the license. Baxter wanted better documentation so short term consultants could use the product, while Computer Partners preferred that we put our efforts into other features.
With the tension in the Computer Associates partnership, we redoubled our efforts to find a Sales Vice President. Over that year I developed a relationship with State Street Bank that would result in a working capital loan. We secured the State Street Loan in 2003 just a few weeks before we had to repay the Computer Partners loan, so Computer Partners lost their security right to claim all of the Christensen Systems assets, including TELON.
By the time we finally hired Larry DeBover, we had been giving seminars on TELON around the North East. The Pittsburgh Seminar resulted in trials at US Steel and Goodyear during 2003. Larry brought on Kathy Harrison to manage customer support and two new sales people: Sue O’Brien and someone from Chicago. Sue was successful but the Chicago salesperson was not. Our sales in 2003 reached $1,800,000.
Larry was more of a marketing manager than a salesperson or sales manager. He worked on Marketing Communications, strategic marketing, developing International Distribution, Establishing an annual user conference and he focused on helping us find venture capital.
By fall of 2003 mostly through Larry’s efforts, we were negotiating with a group of four VCs who agreed to put up the financing we needed, if they could find a Boston based firm to manage their investment. Larry & I met with Bic Stevens of Ampersand, who was impressed with Larry’s aggressiveness. He later met with Don, and at the end of that meeting said that he would not invest. He told Don that the product and its customer acceptance looked great, but the biggest task was to build a distribution network for the product.
Bic had made a lot of money investing in Pansophic, which had gone public after the VC money was used to build their Sales network. He suggested that TELON would be better off with a company like Pansophic which already had an effective sales organization.
After Bic declined to invest, the rest of the Venture Capitalists dropped out, one by one. By January, all of the VCs were gone, and Larry asked Don to make him president and COO. When Chris and Phil heard that Don was seriously considering making Larry President, they told Don that they would leave. Don told Larry that he would not get the promotion, so he started to look elsewhere for a position where he could be president. By March Larry was gone, successfully securing the position he wanted, and we were ready to hire another person as Director of Sales. But we also were being pursued by Pansophic.
Pansophic had just acquired a 4th generation language in Salt Lake City, for which they had done a lot of Due Diligence. They were impressed by the enthusiasm that their customers expressed about TELON, and they knew that they wanted to be in the Application Generation Business. By this time Mark V had faded and the IBM 4th generation product was failing in the marketplace. By now the major TELON competitors were a new Application Generator from a well financed start up in Maryland and a product from France, which had a very effective sales manager in the US, named Art Latterman. Despite our initial misgivings, because of the bad publicity Pansophic had from their Salt Lake City acquisition, we agreed to talk with their executive Vice President. After that meeting, Don said that Bill ??? was the first person who he had met who seemed to understand the market and customers for TELON.
After 4 weeks of Due Diligence, Pansophic made an offer in April 2004, and the four of us had to choose between 1) continuing independently, with the new Director of Sales, or 2) accepting the Pansophic offer. Don and Chris voted to continue as an independent company, Phil and I voted to accept the Pansophic offer.
My view was that:
1) our only effective sales person, Sue O’Brien was now pregnant, 2) State Street Bank was getting out of Small Business lending, and had told us that they would be pulling the loan, 3) now we were facing better financed competitors who had product capabilities that might match TELON, 4) we could look at the sale as our best way to build a distribution network for TELON. Pansophic had agreed to create a dedicated Sales force and Support organization for TELON, using their existing Sales management systems, and Regional Offices. This Group would report dotted line to Al Seyler, the former Pansophic VP of North American Sales, who also had sold Mark IV before he joined Pansophic. (The 9 to 12 month Sales cycle for TELON was much longer than the 2 to 4 month cycle for Pansophic’s other products like Panvalet and their report writer, which were sold without a trial. As such it required different selling and support skills)
We accepted the Pansophic offer, with some sweetener for existing employees. But because they had been burned in the Salt Lake City deal, they were unwilling to actually acquire us without first unwinding the agreements with Computer Partners and some of the international distributors.
Instead they invested $400,000 in Christensen Systems for 14% of the company, and a 6 month option to purchase the rest of the stock. After renegotiating the Distribution agreements, they purchased Christensen Systems in November 2004. A number of customers who delayed purchasing TELON from Christensen Systems until the Pansophic acquisition, then bought in November and December. TELON sales for 2004 were about $3,000,000.
In retrospect, our decision to accept the Pansophic offer was a lucky one, because during the summer of 2004 Chris came down with Mono, and he was unable to work for about 3 months.
After a 4 month period to shift Christensen Systems’ Personnel, Payroll and Accounts Payable over to Pansophic, I was asked to create a position for myself by Al Seyler. Pansophic had a strong Selling organization, but weak marketing. Though I had no skills in Marketing communication, or Strategic marketing, I suggested that I could help by doing competitive analysis. So I stayed on for another three years, working with Al, the Sales force, and the developers to understand TELON’s position and potential in the market.
Because TELON was priced significantly below the two new competitors, our first effort was to raise the price from $95,000 to $225,000 for IMS and from $50,000 to $160,000 for CICS. This gave TELON the credibility to get into head to head trials against the two competitors, and TELON won about half of them. Within the year Pansophic had built a 6 person sales force for TELON and also approached Art Latterman to hire him away from the competitor. After a 10 month delay before his first TELON sale, Art became our highest producing Sales person, selling to MCI and a number of clients in the mid Atlantic region.
As I recall, TELON annual sales grew to about $6,000,000 for 1985, about $12,000,000 for 1986, about 18,000,000 for 1987, about $28,000,000 for 198, and over $34,000,000 for 1989.
What outside influences were there and how did they affect TELON? (i.e. who were the competitors?)
The early Competitors were IBM and Informatics. The later competitors who appeared before Pansophic acquired TELON were ??? from Maryland and ????from France.
What made TELON effective, especially when compared to the competition?
Don was most confident after John Hancock selected ADS after it had benchmarked ADS against Mark V, a product from Informatics and The IBM Application Generator. Mark V was an enhancement to the successful Mark IV report writer, but it was expensive (~$100,000), cumbersome to use, and generated less of the application than ADS. The IBM competitor required a run time component and would be much more expensive to own over the life of an application. Even though ADS was a product from an unproven company, the customer risk was low because it generated stand alone code. And ADS was clearly the most productive for the developers in the head to head trials at Hancock, Raytheon, and Travelers Insurance.
After the Pansophic acquisition and especially after we hired Art Latterman away from the French company, TELON had the superior sales and distribution network. The Maryland company was building its Distribution capability around its sole product. Even with a dedicated Sales and Support staff, TELON enjoyed the Corporate and Field office support of the rest of Pansophic. TELON no longer won every Head to Head trial. But we did learn where TELON was actually a better solution, than its two competitors, and where we should cede the customer to the competitors. Pansophic had the resources to roll TELON out to international markets (except France) much faster than either of our competitors.
This sustained TELON when we got caught in a longer than anticipated development cycle for Release X.0. This release delay may have contributed to Pansophic’s decision to terminate Al Seyler (who had Sale management, not Product Development experience), and to the revolving door of senior managers that we faced in my last 18 months at Pansophic. Each time we brought one up to speed, he was replaced, and none understood the TELON market or TELON customers as well as Al. We almost squandered out market lead, with the late release and the uncertain management. Indirectly, the revolving management led to my departure in 1988. I gave a one year notice in late 1987, and trained my replacement. Within a few weeks after I left, she was reassigned by the newest managers to sell another product.
November 24, 2006
[edit]I remembered Bill Nelson's name as I drove to work today. But I still do not remember the name of the French Product that Art Latterman used to sell or the name of the other competitor we had at Pansophic. Also I do not remember what CSC stood for. Was it Computer Sciences Corporation? I believe that was the west coast company that later acquired Computer Partners.
Chris McNeil
[edit]December 4, 2006
[edit]What was your role at Pansophic?
Director of Development. I was involved with R&D and new product development.
Explain briefly what Telon is.
An application development system for COBOL & PL/I under IMS and CICS mainframe systems. One of the first RAD tools. Generates programs to run under IMS and CICS. It also generates the code to interface to IMS/DB, CICS/VSAM, DB2 and other relational DB systems using 3270 dumb terminals only. Provides tools for prototyping and testing with an interactive screen designer and debugger. Also provides non graphical development environment on a PC under DOS before the days of Windows and networks. Please see the CA website for the current definition.
What were you doing that led up to Telon?
I was employed by Liberty Mutual Insurance starting in 1974. In 1976 I was assigned to a project to develop an insurance claims processing system(CAPS). I was the lead designer and technical implementer of the project. I worked with the business user, George Bean, to design the system. I met Don Christensen on this project in '77. He was an IBM Tech rep assigned to provide support for IBM's new 3790 mini computer which was the front end piece of the system.
In our class we’ve discussed how many innovations were created by someone who combined an understanding of the problem with technological experience. For example, Blaise Pascal watched his father’s duties as a tax collector and later built a mechanical calculator to help him. Is there a parallel with Telon? Did your work leading up to Telon give you an understanding of the problem that Telon tried to solve?
The experience on the Claims Processing System(CAPS) lead directly to Telon. Telon was designed to automate the development process that George and I had gone through to implement the CAPS application.
From a design perspective: As the designer I would sit with George and we would layout screen formats on a 3270 terminal and print them out. We would then take those print outs and notate the business requirements. It was a crude form of prototyping which worked very well. It was the precursor to the Telon Design Facility.
I designed a standard template of screens which could be used to search/add/update/delete an entity. It laid out a flow of screens that the designer could use to standardize the system design. It removed the "creativity" from screen flow which removed a lot of complexity. Each entity had it's own series of screens which adhered to the same formats and screen flow. It simplified the design process for screen layout and system design. From this experience came the Telon standard "template" for system and program design. It was called the "Circle Flow" design technique as each entity had a number of screens which all flowed together.
From a technical perspective as the lead implementer: Liberty Mutual had bought a generic claims system from Insurance Systems of America. The ISA package was a table driven transaction processing system that had been ported to run under IMS/DB which is also a transaction processing system. The implementation required coding numerous tables for screen mapping and data base mapping for the ISA package as well as for IMS/DB. It was further complicated by the coding of screen formats and edit rules on the IBM 3790 mini computers. The business problem we were trying to solve was simply to capture a lot of data with a bit of processing thrown in. The effort to code a single screen was quite large and error prone. The problems were further exacerbated by the fact that the 3790 was not ready for prime time. The buzz word at the time was "distributed processing" but the reality was "distributed headaches".
Don wrote a set of macros that would generate the tables required by all three components which greatly simplified the implementation process. Nevertheless, trying to get the 3790, the ISA package and IMS to all talk to each other was a difficult task. Testing and debugging was a nightmare considering the mixed environment and the lack of any interactive debugging tools.
One day in a fit of frustration, I turned around to Don who sat behind me in the office and said "Why don't we just generate Cobol programs to run natively under IMS? It would be so much simpler and actually more effective then the current kludge." We had discussed it before but this time it struck a chord. We started to work together in our free time to build such a product that would support prototyping, program generation and interactive debugging. Don subsequently left his job at IBM to continue work on this new product. I designed the "template" Cobol program and screen flows which would run under TSO/IMS. Don went to work coding the macros to generate the Cobol code. We met after hours at Liberty or his house to work on the new product.
Shortly after we began work, Liberty moved to New Hampshire. I wanted to stay in the Boston area so I left my job. Don and I continued to work on the software. We had planned to start a company but before we actually did I decided to take a break from the software business and ended up moving to Cape Cod in late '79 to study classical guitar. Don took the bull by the horns and carried on the development as a consultant at Liberty Mutual in New Hampshire.
Fast forward 1.3 years .... I had learned to play classical guitar and had run out of money in the process. I called Don in early '81 to ask him for a recommendation as I was planning to get a part time software job on the Cape. Don by this time had hooked up with Phil Stymfal and was being very successful in implementing the first incarnation of Telon(ADS at that time) at New England Life. Don asked if I wanted to work together again. I told him I wanted to work part time. Don asked if I could work full time with a six month commitment. I ended up working full time, 60-80 hours a week, for the next 7 years.
Was there ever an “a-ha!” moment?
The day at Liberty Mutual when I realized we could generate native Cobol programs to solve the business problem. That day I turned around to Don who sat behind me at Liberty Mutual and said "Why don't we just generate the native Cobol code to create the business applications? He said "That's a great idea, let's do it!" That was the start of Telon. A second a-ha moment was the realization that prototyping was a much more effective process than the then current "paper" design techniques. Another "a-ha" was the realization that we could create a system to prototype as well as debug interactively.
What technical decisions were made and why?
We wrote to the technical environment of our customers which was also what we were familiar with. We were very customer driven in our product development and improving the productivity of our users was utmost. As an example, the IMS/DB environment is a very difficult environment to prototype and test under. To prototype an IMS/DB data base it is necessary to have someone in the systems group create/update the database schema within the IMS/DB environment. It can really slow down the development process. The systems group also has to install the IMS transactions so that the programmer can test their programs. Since the programmer developed under TSO, we decided to write our own version of IMS/DB which ran under TSO and supported all the features of the DL/I language called "Data Language Simulator"(DLS). It also included an interactive debugger for Cobol and PL/I. This allowed our users to quickly generate Telon "DLS" databases and programs and to debug the application interactively under TSO. Once the application was ready to move into production, Telon generated the necessary tables and programs to implement the "real" DL/I database and transactions in the IMS/DB environment.
When where you hired?
I rejoined Don in the summer of 1981. I was the initial software designer/developer/customer support person. It actually was an easy transition back into the software development world as Don had kept to the design of the template programs I had designed when we first began work on Telon.
When did the product acquire the name Telon?
Early 1982 I was discussing product names with a Greek friend. I thought Telon sounded like a good name. My friend constructed a plausible etymology for "Telon" as the combination of Telos "productivity" and Eon "forever". I designed the Telon logo as 8 circles around the name Telon. The eight circles represent the Octave as well as the steps of evolutionary development. We had a graphics design person draw it.
At what point did you realize that Telon was going to be successful?
To be honest, I believed from the initial idea that Telon would be successful. The first four sales were important to allow us to grow the company. The early hiring of Phil Stymfal and Gig Kirk were crucial moves. The later hiring of Sue O'Brien as a Salesperson and Kathy Harrison as Customer Support Manager were also instrumental.
What mistakes were made?
I think the most crucial mistake is that we did not develop a sales force. Early on we were very successful as the market was wide open and we had an "in" at companies that Phil had ties to. The sales cycle was stretched out after the initial sales which left us cash poor. We were not able to raise venture capital which made it difficult to hire new people. The search for venture hurt us as Don and Larry DeBoever(Director of Sales and Marketing) had to spend so much time with the venture people that sales suffered. At that time Don was our sales evangelist and the driving force behind our sales. We had hired only one salesperson after two years in business. I view the sale to Pansophic as a move that allowed us to acquire a sales force. Pansophic had a great sales force and Al Syler was a very effective manager.
A thing that really hurt us a while after selling to Pansophic was that the development team lost contact with our customers. Pansophic's policy was developers and customers never interact. All communication came through the Sales and Customer Support personnel. Unfortunately the Pansophic Customer Support personnel were not equipped to carry out this role. We lost our greatest advantage which was the direct feedback from our customers.
What were the long-term weaknesses of Telon? Was it possible to correct them or was it necessary to start over?
We created a software product to solve the particular problem presented with the technology of the day. It was very successful for it's time and in it's market. It is still very successful as a legacy product. The difficulty was that the market's move to PC's and networks was a quantum leap which created a brand new market which required all new tools. Telon became a legacy product.
Why did you leave Pansophic?
The 3-year commitment expired.
Additional Comments
We built Telon to solve real life problems for real life users. Since we had developed business applications in the past, we had great insight into our customers needs. We were very customer driven.
I think the methodology we sold with the software was very helpful in making our customers more successful in designing systems. Telon presented a standard system/program design that all systems utilized. Basically if you saw one Telon system you saw them all. This was a real advantage for our customers in designing and maintaining systems. I really like the K.I.S.S principal when it comes to software systems.
Telon focused on prototyping. In those days the typical design methodology generated tons of paper to support a design. Basically every line of code had to be speced out in some way before any code was written. There were also endless "code reviews". Using our approach of real time prototyping, the system was being designed and built at the same time. Instead of endlessly documenting a system before starting to implement a system, the user actually built a basic self documenting prototype which was further refined as the system was built. The need for "code reviews" was greatly simplified as there was no longer a need to review the generated program logic. The reviews could focus on the custom business logic.
We promoted evolutionary application development so that the user could really understand what they needed by using the actual system they were building early on in the development process. We recommended building the system with many cycles of design and implementation. This was quite different from the "design it all" then "build it all" technique of the time. The resulting systems were much more likely to meet the user's needs.
It is interesting to note that there was some resistance to standardizing and simplifying the system/program design. The resistance was presented by programmers who were more interested in programming for themselves as opposed to programming to solve the needs of their company. The idea that the goal of a software developer is to make the company more successful is an important lesson to learn as a software developer. Fortunately we had a great group of early adopters who provided great feedback and allowed us to build a system that satisfied their needs and thus make both CSI and our customers successful..
It is fitting that the original "technically kludged" CAPS system I had worked on in the '70's at Liberty Mutual, which was my raison d'etre for Telon, was subsequently rewritten years later using Telon. It was a good day when I heard that news! Interestingly enough, the system that Don had worked on at Liberty Mutual in New Hampshire also ended up being rewritten in Telon.
The corporate ideals of CSI were "Excellence, Integrity and Respect for the Individual". I think these ideals went a long way to promote a great culture. My personal goal was to build a company that I would be happy and proud to work for. We had a great group of people.
Fair use rationale for Image:TelonLogo.jpg
[edit]![](http://upload.wikimedia.org/wikipedia/en/thumb/f/f7/Nuvola_apps_important.svg/70px-Nuvola_apps_important.svg.png)
Image:TelonLogo.jpg is being used on this article. I notice the image page specifies that the image is being used under fair use but there is no explanation or rationale as to why its use in this Wikipedia article constitutes fair use. In addition to the boilerplate fair use template, you must also write out on the image description page a specific explanation or rationale for why using this image in each article is consistent with fair use.
Please go to the image description page and edit it to include a fair use rationale. Using one of the templates at Wikipedia:Fair use rationale guideline is an easy way to insure that your image is in compliance with Wikipedia policy, but remember that you must complete the template. Do not simply insert a blank template on an image page.
If there is other fair use media, consider checking that you have specified the fair use rationale on the other images used on this page. Note that any fair use images lacking such an explanation can be deleted one week after being tagged, as described on criteria for speedy deletion. If you have any questions please ask them at the Media copyright questions page. Thank you.
BetacommandBot (talk) 08:23, 15 January 2008 (UTC)
I don't pretend to know whether this has any legal relevance, but the logo has not been used since about 1985. John Blackwell (talk) 01:55, 5 October 2008 (UTC)
proprietary runtime
[edit]While Christiansen Systems and Pansophic owned Telon, it is true that the runtime components were considered public domain. Howaver, since CA took over, they have included terms in their maintenance contracts that customers may not continue to use the runtime components (merge routines, field edits, abend handlers, etc.) after their maintenance contracts expire. Smaller software vendors have had some success in selling replacement routines to allow customers to escape this trap.
John Blackwell (talk) 12:14, 20 May 2011 (UTC)
Blacklisted Links Found on the Main Page
[edit]Cyberbot II has detected that page contains external links that have either been globally or locally blacklisted. Links tend to be blacklisted because they have a history of being spammed, or are highly innappropriate for Wikipedia. This, however, doesn't necessarily mean it's spam, or not a good link. If the link is a good link, you may wish to request whitelisting by going to the request page for whitelisting. If you feel the link being caught by the blacklist is a false positive, or no longer needed on the blacklist, you may request the regex be removed or altered at the blacklist request page. If the link is blacklisted globally and you feel the above applies you may request to whitelist it using the before mentioned request page, or request it's removal, or alteration, at the request page on meta. When requesting whitelisting, be sure to supply the link to be whitelisted and wrap the link in nowiki tags. The whitelisting process can take its time so once a request has been filled out, you may set the invisible parameter on the tag to true. Please be aware that the bot will replace removed tags, and will remove misplaced tags regularly.
Below is a list of links that were found on the main page:
- http://www.cbronline.com/article_cg.asp?guid=A832A908-0B08-4F5B-BCCD-8A2BDB7C3EDE
- Triggered by
\bcbronline\.com\b
on the local blacklist
- Triggered by
- http://www.cbronline.com/news/pansophic_to_see_telon_38
- Triggered by
\bcbronline\.com\b
on the local blacklist
- Triggered by
If you would like me to provide more information on the talk page, contact User:Cyberpower678 and ask him to program me with more info.
From your friendly hard working bot.—cyberbot II NotifyOnline 18:07, 8 December 2013 (UTC)