An Interview With Mark Herring  By Dan Snapper Mark Herring is the "father" of BBS offline mail reading and the creator of the .QWK message format. Back in April of 1988 he released the first offline mail reader and started the revolution that has enabled the amazing growth of internationally echoed message networks. Known as "Sparky" to many of us, Mark has become a very succesful shareware author, too. His current products include the Qmail 4.0 door, QNet 3.0 network messaging software, PCB/FaxMail system, The Pager Engine, DeLuxeý 1.25, and the recently released 1stReader 1.00. We asked Sparky to share a brief history on himself and the evolution of the offline mail reading system. We also asked him some questions about his experiences and views on shareware today... ŠSparky: I've been calling bulletin board systems since my first call to a Forum-80 back in March, 1979. Since then the systems, of course, have grown quite a lot. I called that BBS using a 300 baud 'drop-in' modem (called a CAT modem) and the BBS had all of 160k of disk storage (and that included the OS and programs to run it too!). Like I said, I've been an enthusiast since then writing terminal programs, one BBS system (for a TRS-80 Model III back in 1983) and having fun. In 1986, soon after PC-Pursuit was introduced, I wrote a series of programs that would keep a database of bulletin board systems around the country. I could select the BBS systems I wanted to call and the program would generate a ProComm script file that would handle the work. I could then go to bed and have the script call these systems late at night, on PC-Pursuit, to gather new messages (in a capture file), file lists and it could also upload and download files automatically. It even kept a ratio list so we knew where to send some files when the ratio got too much out of whack. The systems that I called were all PCBoard systems because the sysop couldn't change the prompts. That meant that all of the systems had the same prompts and the scripts would be reliable. That was THE big reason I started working closely with PCBoard, way back then. I gave another copy of this system (we called it 'Rape and Pillage') to Charles Grisamore and he and I would call 10-15 systems a night. Then, around 5:00am, his system would switch into a host mode where mine would call his, upload everything I collected, he would send me everything he collected, and then we'd hang up. My system would then call two local bulletin board systems (Radio Free Memphis and Unusual Situations) and his would call another friend of ours to send him the same collection of files. I wrote a little 'reader' program that would let us read the capture file and reply to messages. Then, the next time we called the system these messages would get uploaded to the system. So, in a sense, this was a 'prehistoric' Qmail system. In February of 1987 another friend of mine, Jim Key, was thinking of shutting down his PCBoard system so he could get a system that interfaced with FIDO. I remember calling him up one night and asking him if any software was available to let PCBoard echo messages. He wasn't aware of any - so I told him I'd try and modify our R&P program. I did and it worked, but it was VERY crude. Jim gave me a copy of a PCBoard message base and I reverse engineered the file format and then wrote a program that basically cloned ProComm's script language and provided echomail transfers. That was the birth of PCB-ECHO around the end of February. The system was setup between Radio Free Memphis and Unusual Situations, another PCBoard system based in Memphis and run by Jim Pottkotter. Everything worked well for about a week or so until I realized we needed to add a third system into the works. Echoing between two systems was easy but adding that third system took a bit of work. In March we added a system from California named 'Sleepy Hollow' to the group and things really started moving. Soon, systems like The Redundant Department of Redundancy (Utah), Execnet (NY), LANS (IN), The Sound of Music (NY), Farwest (British Columbia), Fleamarket (New Jersey), The Toolshop (Arizona), Qmodem (Iowa), Ed Hopper's PCBoard (Texas), Sparta PCBoard (New Jersey) all came together to create the PCB-ECHO network. It was all hubbed out of Memphis on Jim Key's 40 megabyte Compaq 286 system. So, from then on I was connected with Jim helping him manage PCB-ECHO. I fixed the bugs and helped a bit in the administration. Only Jim had the software that would let hub the network - the other systems had the 'node' software. So the network was a 'star' type of design where you would soon max out the number of systems available. That limit was reached around 50 systems, and it didn't take too long to reach it. In September of 1987 another friend of mine, Dan Mascheck, moved from Memphis to a little town named Wharton, TX just sixty miles outside of Houston, Texas. He kept calling back home to RFM (Radio Free Memphis) to keep up with the happenings and was starting to feel light in the pockets due to the long distance costs. One night he called me and asked if there was something I could do to help him out. Back in 1985, during a vacation to The Smokey Mountains, I had given some thought to what would happen to our hobby once measured service was installed throughout the country. I reasoned it was obvious that if calls were to be measured in units it was best to login, download as many messages and files as you could, then log off and read the messages at your leisure. Then, call back and quickly upload the replies. With this idea in mind I started creating The Qmail System for Dan. But I wasn't designing it for everyone, basically just for him to use. I took the PCBoard 12.x message base formats since I reasoned it was quicker to just grab the messages rather than translate them into another format. And I remember designing the format to the CONTROL.DAT file in about thirty minutes. At the time, I had NO IDEA that I was creating a standard. If I had known that, I would have given considerably more thought to what I was doing. But at the time I was helping a friend out with a problem and not thinking much past that stage. The original Qmail 'door' wasn't a door at all. It tied into PCBoard through the use of a download protocol. If you wanted to upload replies you would use the same protocol (maybe it was the 'Q' protocol) to upload the replies. I soon realized that a door was going to be required so Dan could configure conferences and his setup, so I had to figure out how to write a door. Well, as work progressed on the door and the original Qmail Reader a strange thing started happening. Sysops who belonged to PCB-ECHO started asking me about the system and told me they'd buy it as soon as it was finished. With echomail becoming more popular, these one or two node systems were getting overloaded by users calling in and reading mail 'online'. The other obvious advantage to the Qmail design (but one I had failed to realize when I started work on it) was that it could get local users in and out much quicker so more users could be serviced by a single node. It turned out that with Qmail, I was in the right place at the right time with the right idea. By the time the system was released on April 15th, 1988 we had 32 orders backlogged ready for shipping. I thought I'd be lucky to sell 100 doors in total. I had no idea that I'd still be at Qmail development over four years later with 1,500 sysops owning the system and over 13,500 copies of our offline mail reader registered. Soon afterwards I realized that the Qmail Door would make an excellent network hub so I created QNet software for other PCBoard systems. This introduced hierarchal networking and got us away from the 'star' cluster used in PCB-ECHO. Soon, other networks were starting up - built around Qmail Door and QNet. PCB-ECHO was transformed into the ILink network on September 22nd, 1988 when nine 'super regionals' got together and decided that Andy Keeves would assume the new network administrator from Jim Key. Since then, ILink has grown under Andy's management and other large networks have grown too. Jim Key continued to run Radio Free Memphis until this year. The system was closed in the spring. ŠSTARgazer: You've recently released a new mail reader called 1stReader. How does this product differ from other mail readers currently on the market? The other mail readers on the market today require that you combine other programs together with them in order to perform their full functions. In other words, if you were to take offline mail reader XYZ and give it to someone, they couldn't use it by itself. It requires a telecommunications program, a file compression system, text editor, file listing program (LIST) and other programs in order to work. 1stReader provides all of these programs in one easy-to-install and use package. The 'computer intelligence quotient' of the user does not need to be at the 'expert' level. Anyone who knows which serial port is connected to their modem knows everything that is required to use 1stReader. Since we also created The Qmail Door, both packages work together to make it easy for new users to download messages. We've received one report that a seven-year-old has successfully installed 1stReader and uses it to receive messages! Now, try that with any other offline mail package out there today. ŠSTARgazer: What method(s) do you currently use to market your products? ŠSparky: We use the 'demoware' concept. 1stReader normally handles up to 400 messages per conference. But, in it's 'demo' mode it will only let you read up to twenty messages per conference. We feel that 20 messages is more than enough to find out if you like the reader and how it operates. We do not disable any other feature in the 'demo' mode. Demoware seems to work pretty well for us. It lets anyone who is serious about registering the product evaluate the features of the system and yet provides an incentive for registration. I've seen where some sysops have objected to 1stReader's 'crippling' (as they call it) and will not let it be posted on their systems. I view the 'demoware' concept just like a subscription to a bulletin board system. If the caller likes the bulletin board system, after a brief evaluation period, they must pay money to have full access to the BBS. There is no difference between this method and 1stReader's 'demoware' concept. Qmail Door, when it's finally released, will also be available using the 'demoware' concept. QNet, on the other hand, uses traditional or 'classic' shareware methods. When you evaluate QNet you have the entire system at your disposal, just like a registered version. ŠSTARgazer: Why use different marketing methods for different products? Isn't this sort of inconsistent? ŠSparky: Not really. We consider QNet to be a 'leader' for sales of the door and reader. In other words, why purchase an offline mail door and/or offline mail reader if the volume of mail on the bulletin board system doesn't justify it's purchase? If we allow sysops to use QNet fully in a demo mode, then they will begin to build up large conferences on their own systems by connecting with other echomail networks. And that soon drives up the demand for an offline mail system. Our hope is that they remember where they received QNet from and come back to us for the offline mail products. ŠSTARgazer: You obviously feel that different products can require different marketing methods. What experience(s) have lead you to choose the method you're now using for distribution of DeLuxeý and 1stReader? ŠSparky: When we released the first Qmail Reader back in April 1988, it was released under 'classic' shareware. By June we had over 80 Qmail Doors sold but only FIFTEEN Qmail readers registered. I can still remember the day I made the decision - I was in Kennett, Missouri on business. There are two methods that can be used to provide incentives to register a product. The first method is the 'carrot' approach - when you register our product we give you a better one for your use. The second method is the 'stick'. This basically ends up with you policing conferences and finding people who continue to use the offline mail reader but do not register it. You then beat them over the head with your 'stick' getting them to register the product. Now, you'll catch a lot more flies if you use honey than vinegar goes the old saying, and I believe in that. So, we decided to create Qmail DeLuxe and give it to those who registered the shareware offline mail reader. Within three weeks of DeLuxe's availability we had over three hundred registrations! My philosophy is that you go with what works. In our case, we found that the incentive method worked well. You don't kill the goose that lays the golden eggs. Using this method we've registered over 13,500 copies of our offline mail reader. We've extended the concept even further with our new 1stReader product. When you are finished using the product in its demo mode, the reader asks the user if they wish to register the product right now using their credit card. They can then fill out an order form in the program and using the telecommunications module built into the reader, it calls a computer in our offices dedicated to nothing but processing orders. When they call our online order processing system, 1stReader uploads their order (encrypted in a zip file for protection) and then waits while our system unpacks the order and examines it. Then, the user waits about 30 seconds while the system calls out to the credit card clearing house on the second serial port's modem. This software checks to make sure the card number is not stolen and that there is sufficient credit available to complete the sale. If the sale was approved then the user can download the 1ST.KEY file so the can immediately use 1stReader (as a by-product, our checking account gets deposited with the money within 72 hours as well). This gives the user 'instant satisfaction'. For the one minute phone call they can begin using 1stReader immediately. We also send along a THANK.YOU file that thanks the user for the order. So far, the system has worked WELL BEYOND our expectations proving once again that if you make product registration simple and easy, you'll get a better return on the work invested. We also are about to institute dealer programs so sysops can sell 1stReader to their own users. We've come up with what I think is a unique sales method based on our online order system.