Elections, Part II

This mystifying set of images was also in the “Elections” folder. All it says on the back of the contact sheet is “Election.” My somewhat wild guess is that this was some kind of information gathering and disseminating site for an election in (another guess) the ’70s. I’m troubled, though, because I can’t come up with a theory that would explain why this would be happening at Rice. Note, however, that these images are a goldmine of obsolete technology.

Election maybe 70s

Election 70s

Elections nd possibly 1970s

What makes it even more confusing are a bunch of other pictures interspersed throughout the same contact sheet. It looks almost like a class. I confess I don’t know what to make of it.

Election 70

Bonus: Also found in the biology labs. This one I can’t identify.


This entry was posted in Uncategorized. Bookmark the permalink.

32 Responses to Elections, Part II

  1. Dagobert Brito says:

    Dear Melissa, that is a computer terminal. Bob

  2. Richard A. Schafer says:

    Melissa, during the 70s and sometime later, there was a press organization called Election Central that collected unofficial election returns and compiled them for radio/TV/newspapers in town. I’m not sure exactly how they connected to Rice, but we provided the processing for their work. Poll workers went to each polling place and collected the counts from the election judges and sent those counts to us, where they were keypunched (this was the 70s, after all) and compiled. At some point, Wm (Farrell will remember this in detail) produced Apple II software to generate Chiron feeds to let the TV stations put those result crawls on the bottom of the screen. We had terminals set up in every TV newsroom in town, with modems connecting them back to ICSA. The basement of Herman Brown was blocked off to anyone but the poll workers coming in with their data. Madeline Appell (now an assistant to the mayor) was one of the people who ran Election Central. I have vivid memories of certain rude candidates who would try to bully those of us working the system to give them results. The first picture appears to be a live remote broadcast from our machine room. That’s Jim Criswell on the right, who now I think of it might have been the connection between Election Central and Rice. The man in the suit facing the left was a local TV reporter, but I don’t remember his name. The second picture is one of the keypunch operators keypunching the data onto punch cards to be read. The third picture is a radio reporter, I suspect, calling in his report of the latest totals. For a while, we had these huge racks with coathangers and clips for holding the unprocessed (or processed) tally sheets. I still have some of those coathangers. I’m not sure about the fourth picture. I suspect it is a group of students going over some tally sheets, but the room doesn’t look like anyplace in the basement of Herman Brown. Carlyn Chatfield probably has some information on Election Central that she collected as part of the Centennial project.

    The terminal is a Tektronix 4006/1 graphics terminal from about 1975. See http://terminals.classiccmp.org/wiki/index.php/Tektronix_4006.

  3. effegee says:

    This is Election Central which for about a dozen years tabulated informal returns for the 5 Houston media outlets. It is at Rice because Dr. M. Stuart Lynn, Director of ICSA, hired the late James L. Criswell away from The Houston Post to manage the administrative details of ICSA after the late William Paling left the post of Assistant Director for Transco.

    Election Central was a consortium consisting of the The Houston Post, The Houston Chronicle, KHOU-TV, KPRC-TV, and KPRC-TV. Criswell had had an integral role in creating EC and convinced the media to move it to Rice which, at that time, sold computer services to outside organizations.

    In the days before electronic voting machines, reporting of official results was a slow process, often not completing in time for either the 10PM TV news or the morning edition of The Houston Post. (The Houston Chronicle’s premier edition was at that time in the evening.) Mechanical voting machines had to be opened at the close of the polls, each candidate’s vote read off, and the counts from each machine added up. The precinct judge then certified the results and drove the results to the courthouse. For primary elections with many candidates, precinct meetings, and twice the number of precincts, the results of close races were often not known before the next day.

    Election Central placed volunteers from groups such as the League of Women’s Voters in each polling place as “deputized” members of the press. The volunteers listened as each machine was opened and read out. They tallied the votes on precinct-specific tally sheets, added up the candidates votes, and phoned the results to EC. EC added up all the precincts and displayed the results. Prior to its coming to Rice, EC used “menu boards” to display results.

    After EC moved to Rice, phone calls were received in Jones and Brown by students waiting by their phones. They copied down the results from the field volunteers. (Each student receiving calls was assigned to a certain group of precincts.) The completed form was placed in the hall where a runner grabbed completed forms and took it downstairs to a motorcycle “runner” that took batches to Herman Brown Hall basement where the forms were keypunched and input into programs running on the IBM mainframe. Initially, printouts were distributed to press representatives waiting in the user work area of the basement.

    Within a couple of years, ICSA created electronic delivery to the media outlets (although printouts were still made and picked up by the media for distribution to reporters working on stories). The initial version of electronic distribution used stock CRT terminals for TV and IBM Selectric mechanism terminals for the print media. Custom software read files on the mainframe and displayed them on the terminal. The TV stations quickly went “on camera” with the CRTs. Eventually, a student, Wm. Leler, created a custom interface board to be placed in an Apple ][ computer that replaced the terminal and also connected directly to the stations’ title generators to produce broadcast-quality images (compared to the rasterized CRTs). This arrangement was used until the service was terminated after the advent of electronic voting machines.

    After Criswell became ill, and subsequently passed away in 1978, Rice alum Mary Burnside ran a couple of elections. Following her departure, Madeline Appel (now of Mayor Parker’s office), ran EC for several years until she began working in government.

    EC ended with a whimper. Electronic voting machines eliminated the reading of results in polling places and improved the speed of reporting of official results. After a couple of elections using a modified system the input official results into the distribution system. But the value added no longer justified the cost of even the stripped down system. Rice’s involvement ended in 1984 if I recall correctly.

  4. effegee says:

    The top picture is from 1976. I think it was the general election but it could have been the primary. There was only one time we allowed the press to go live from the computer room. The talent is a KHOU anchor — don’t recall his name either. I believe that’s a producer to his left. As Richard noted, Criswell is at the right edge.

    This particular election was unusual due to an IBM-induced bug in the mainframe’s operating system a few weeks before the election. While the bug didn’t surface during normal operations, the custom software tickled it repeatedly during the evening causing a loop in supervisor state the required a reboot each time it happened. (Mainframes at that time booted even slower than Windows Vista!) At roughly :28 and :58 all three TV stations were pounding the terminals with on-air reporting with 1,680-character blocks being sent to each station at about 3-5 second intervals. Because we have KHOU’s monitors in the computer room we were very aware that, every time Bill Balleza (then at KHOU) got on the air back at the station, he got at most 3 screens before the system locked up leaving him ad-libbing.

    Fortunately, IBM found and fixed the problem a few weeks later so this embarrassing occurrence was never repeated. However, the TV stations were very interested when Criswell carried Leler’s plan for the Apple ][ interface that would allow them to pre-record results into their character generators.

    The last election picture appears to be a recheck room. The EC software at Rice checked total precinct vote in each race against precinct registered voters and kicked out improbable votes for review. This looks like one of the rooms used for that… I think, perhaps the ICSA conference room based on the size of the table. I also think that that may be Mary Burnside at the left in this picture.

    • Richard A. Schafer says:

      I also think that’s Mary Burnside in the recheck room picture. Is that possibly Ron Cytron up against the blackboard on the left?

  5. effegee says:

    Sorry… I misspelled Madeleine Appel’s name in my first post. She is currently Deputy Chief of Staff for Administration for Mayor Parker. She left Rice to serve on Council Member Eleanor Tinsley’s staff.

  6. Melissa Kean says:

    First, Richard and Farrell, thanks very, very much. I truly appreciate it. Second, it’s interesting to me that something like this—which sounds like a fairly big deal—I’ve never come across before. I wonder if there’s more documentation in some of our manuscript collections. It’s entirely possible that I just didn’t know enough to recognize this when I saw it. I’m going to look around a little.

    • effegee says:

      Carlyn Chatfield in IT may have additional information turned up in her work for the Centennial. Madeleine and Jim’s widow were interviewed with several of us on Election Central. Any files that still exist are likely to be in IT’s possession. They may have thought of sending something to the archive but I doubt it. I know that I left Rice’s first network router (Proteon P-4200 serial #28, IIRC) in my office when I left Rice with a note attached that it should go to the Woodson. As far as I can tell, that never happened. 😦

    • Richard Miller (Hanszen '75 & '76) says:

      Also, this was not a volunteer effort on the part of the students. We got paid real money for our work. I worked for EC several times. In most cases it was at ICSA but once I was dispatched to the Galveston court house to get the election results for Galveston county. We did not use a runner as I recall but phoned the results in.

  7. Grungy says:

    My brother was one of the (more maniacal) motorcycle runners for EC, but his tales say that this took him far afield, not just across campus. IIRC, he was traveling between downtown and Rice.

  8. Keith Cooper says:

    The third shot is a nice photo of an ICSA study carrel, complete with a vintage dial phone. Some of your readers will recognize these as part time sleeping quarters; students were often sacked out underneath the carrels late at night — occassionally with a pile of printout for a pillow.

    • Richard A. Schafer says:

      And some may remember that Bill Paling hated seeing people sleeping under those carrels and would waken them with his foot as he walked by.

      • effegee says:

        Yeah. Scott Warren can attest to that!

        BTW, the dial phones on the carrels were plugged in only for election nights. There was a single dial wallphone on the column next to the most northwest carrel near the door to the user-access keypunch room (picture #2).

    • Richard Miller (Hanszen '75 & '76) says:

      This was especially true during the final weeks of the semester when turn-around times started to increase almost exponentially. At this time, terminals had really not arrived. You could use them for APL and limited TSO but the large projects such as compiler parsers still used card decks which were so large they had to be read in through the ‘big’ card reader in the machine room. By my senior year (I was working for ICSA by then) we had a semi-working HASP interface that would allow you to submit a job to the internal reader via TSO and avoid the lines either at the small 2501 card reader or the wait at dispatch. We would work in our shared office using TSO to edit the card deck we had read into a file, submit it to the internal reader, wait for the printout to come back. (We also slept in the office)

  9. dingbat says:

    I was one of the motorcycle runners, as Grungy mentions. My runs were between the county tax office downtown and Brown, then shifting to between HB and Hanszen. I wasn’t told specifically to travel ‘maniacally’, only to get there as quick as I could, so I went only “ludicrous speed.” Going Main Street all the way was much too slow. 15-25 minutes if you made the lights. I believe the land speed record for the two routes was set and broken multiple times those nights. Main to Prairie to Bagby to Dallas, over the bridge and then immediately left down the freeway embankment onto the underground section of I-45 that becomes the Pierce Elevated, jet off down that and onto 288, exit at McGregor, go right at Cambridge and enter campus at Lovett, straight across campus past Willy on the sidewalk and make the zig-zag around Chem Lec to the front door of HB. Empty backpack. Repeat. Had the delivery trip down to about 8 minutes or so by breaking Every Single Traffic Law. Great fun. Going across campus was tougher. The bike was WAY too loud. Someone came out of Hanszen as I roared by and tried to strangle me. In an effort to be nice, I sacrificed a little speed by killing the engine at Weiss and coasting to the South doors of Hanszen commons for the handoff. A kid was waiting there to take the papers up to a room and use their phone. It was during break, and they were camped out in a few different rooms to distribute the calls. I pushed the bike out of earshot and idled my way back to HB. Ah, good times.

  10. I hate to think what would have happened if you had been specifically directed to travel maniacally. After all, you were on a mission from God, or from the county election commissioners, or something. Motorcycle across the quad on the sidewalk: wow, that was a different time.

  11. R Kimbro says:

    I love everything about this post.

  12. Jerry Outlaw says:

    Bert- You were always such a literalist!

  13. mattnoall says:

    EC ran starting before 74 but I am not sure of how much before. Perhaps Farrell can recall

    The Tek graphic terminal was a bit rare In those days. There was one in the terminal room used by many as an alternative to the Decwriter paper output terminals (yes, young ones that was state of the art in the early 70’s). About 75 or so one of the students wrote a Star Trek game for the Tek terminal that was a fairly popular way of taking a small break. I think it was Bill Paseman, but perhaps Keith can rember

    • Richard Miller (Hanszen '75 & '76) says:

      As I recall, the terminal room originally had either Teletype ASR33s (110 Baud) or IBM Selectric terminals (134.5). There was the one Tek for the graphics people and sometime in that era we started getting some 300 baud CRT screen terminals. All of these were hard wired to the IBM 3750 communications controller. You could do TSO on any of them and with the right typeball you could do APL on the IBMs. I remember the game. I do not think it was written here but was freeware (may have been modified by someone a Tektronics). I think there was a version for the CRTs and even the IBMs. My memory is users were allowed to play unless someone needed a terminal for ‘real’ work

      • Mark Williamson says:

        Minor corrections for posterity:
        The terminal room had Teletype model ASR33 terminals before the IBM 2741 Selectric-based terminals were added for APL. The ASR33s had been used with the previous mainframe, a Burroughs B5500.
        The IBM communications controller model number was 3705.

  14. effegee says:

    The first EC was strictly hard copy printouts delivered in the HBH basement. I am pretty sure that this was the 1972 general election.

    In October 1973, EC processed the first transit referendum (HARTA http://en.wikipedia.org/wiki/Metropolitan_Transit_Authority_of_Harris_County) delivering results to the sponsoring media (papers + TV) on IBM Selectric-like terminals. The system was written using the IBM APL interactive system. This was also the first election that I worked; I operated the terminal at KTRK.

    The HARTA experience whetted the sponsors’ desire for on site delivery. However, the APL system had limitations of workspace size that precluded its use for primary and general elections. (There may have been a effort to do so; Richard or Mark may know.). We began exploring ways of using the general time sharing system, TSO, to deliver results. TSO really only tolerated ASCII terminals having been intended to use IBM’s >$5K a seat 3270 display stations. Even if we had the 3270 model that was deplorable at remote locations, the cost and effort to do so several times a year would have been prohibitive.

    I can’t recall the exact sequence now but between October 1973 and the general election in November 1974, we created online delivery on TSO. That entailed modifying the batch EC programs to create files to be displayed online. Initially, I think these were simply typed at the terminals using the system LIST command wrapped in a macro (“CLIST”). This was slow and clunky. On top of that the TV stations quickly saw the potential for displaying results on the air and didn’t like the previous page scrolling up the screen.

    By the 1974 general election, as a newly hired full-time employee, I had written assembler programs using a low-level interface that wrote an entire screen in a single I/O, pausing for the terminal operator to signal to continue by pressing “return”. For TV stations, it was expected that they would record a sequence of displays and play them back as needed. To my surprise, when I responded to a connectivity problem at KTRK, they had removed the cover from the CRT. Put the terminal in a dark corner with a camera about 18 inches from it, and were using it live. The terminal operator crouched under the camera to press the return key. The other stations followed suit. There was a separate version of the display software for the newspapers which printed an entire race at a time. Both versions had the ability to move from race to race by entering a race number at the prompt allowing direct access to “interesting” races.

    The TV stations weren’t happy using the raster CRT displays which didn’t produce “broadcast quality” results. After the 1976 “TCAM loop” debacle described above, EC gave the green light to the Apple ][ project that had been proposed by undergraduate Wm. Leler. Leler constructed wire-wrapped interface boards that connected to the mainframe as a terminal on one side while connecting to a station’s broadcast titling system on the other. He wrote software that allowed the Apple operator to fetch a race and the record it to the titling system. Because the three stations used two different titling systems, there were two radically different models of boards and software. There were no spares…and Leler left to go to graduate school in North Carolina long before EC folded its tent! Oy! I have no idea what became of the boards. They were boxed up and “stored” with the rest of the EC files in the mid-1980s.

    • mattnoall says:

      Are you sure the TSO wanted ASCII? Given the mainframe was an IBM 370 I would expect EBCDIC

      • effegee says:

        The video terminals were definitely ASCII. The later dot-matrix printers were also ASCII.

        You are correct that the Selectric mechanism terminals were EBCDIC. We were looking for replacements for them during the development of the online distribution system in order to have higher speed hard copy terminals for day-to-day operations. The problem was that the hard copy terminals also had to support APL which limited the choices because APL was -Er- a niche market. The DECwriter finally provided the right functionality.

  15. effegee says:

    Forgot to mention the terminals themselves.

    The first video terminal were made by Teleray. They were connected at TV stations by modem at 300 bps. They were replaced by the Apple][c microcomputers,mi believe at the same speed.

    The initial print terminals for HARTA were a mixture of IBM Selectric mechanism terminals — IBM 2741 and a compatible made by Novar. These ran at 134.5bps. The Novars continued to be used with the TSO print element at the newspapers. The papers eventually migrated to a dot-matrix similar to Decwriters at 300bps.

    I don’t recall using any speed higher than 300bps at the remote locations.

  16. Deborah Gronke Bennett (BSEE Hanszen 1982) says:

    Melissa, are you lost in the technical jargon back there somewhere? 🙂

    • Melissa Kean says:

      Oh my, yes. But the important thing is that it all gets written down and saved. It doesn’t really matter that I don’t understand it. Although I do understand some of it!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s