MSA crosstable formats -- a dream thread

OK, let’s start this as a new topic, then.

For some time I’ve thought there could be many improvements in the ways MSA crosstables are displayed.

I’d like this to be a “Dream Thread”. In other words, I’d appreciate it if everybody could refrain from making comments like “Which of the dozens of important projects should be put on the back burner for this?” or “Don’t you realize how many changes would have to be made in how many places?”. Let’s dream a while. Dreams can become reality only when we start with questions like “What changes would we really like to see, if they were possible?”. The nuts and bolts can come later, like after about the 30th post (i.e. not until the top of page 3 or so) of this thread. Is everybody with me?

Let’s start with generalities. As soon as a viewer calls up an MSA crosstable, he should see the header information – tournament name and dates, location, number of sections, number of players, sponsoring affiliate, names of tournament staff – followed immediately by a set of radio buttons:

[b]Sections:

  • Open
  • Under 2000
  • Under 1800
  • Under 1600
  • Under 1400
  • Under 1200
  • display ALL sections

[/b]The above set of radio buttons could be omitted in the case of single-section events.

After the viewer makes his choice, the next set of radio buttons could be:

[b]Display Format:

  • Traditional display
  • Single-line display
  • Double-line display
    [/b]The default could be “traditional display”, i.e. the current format with lines of hyphens between each player and the next, and vertical bars separating the fields. We can discuss the other two (or more) options in following posts.

The third set of radio buttons could be along the following lines:

[b]Player Order:

  • in order of score, and within score by post-event rating
  • in order of score, and within score by tie-breaks furnished by organizer
  • in order of score, and within score by the ratings used for pairing and prize purposes
  • in alphabetical order by player name
    [/b]The first option could be the default, as it is now. The second option would be grayed out if the organizer did not furnish tie-break information. Similarly, the third option would be grayed out if the organizer did not furnish pre-event ratings.

Extra fields could be added for tie-breaks and ratings used for pairings and prizes. Initially, these fields could be added by hand by the organizer, into the U.S. Chess data entry software, after he has uploaded the “normal” fields from WinTD or SwisSys. Eventually, the pairing programs could be upgraded to populate these fields automatically. That way, features can be implemented in stages. They wouldn’t have to suddenly spring into existence all at once, from all sources.

How’s the above for a start? Still more features can be suggested in future posts.

Bill Smythe

Since we will be Re writing this area of our system over the next 24 months this is timely. Many more things will be possible.

I dream of a link (either using MSA or a scanned PDF) to every tournament ever rated.

Some of the historic tournament wallcharts found by Dave Hater in the Cleveland Library:

chesstour.com/nhs72.pdf

chesstour.com/ao72.pdf

chesstour.com/nyj72.pdf

chesstour.com/mwccl72.pdf

chesstour.com/snes72.pdf

chesstour.com/mat72.pdf

chesstour.com/esro72.pdf

More historic info can be found from earlier years at: chesstour.com/cross.html (near the bottom).

Fixed format crosstables with dashes to draw boxes are archaic. MSA should have been rewritten to use HTML-rendered boxes years ago.

That was one of the points in my post at the top of the page:

[b]Display Format:

  • Traditional display
  • Single-line display
  • Double-line display
    [/b]The first option could be kept for continuity, or for anybody who is using that format to create wonderful stuff in Excel. The second and third options should be HTML-based. More on that a little later, when I get a round tuit.

Bill Smythe

Are these in paper form, or electronic?

If the latter, some volunteer could modernize them, spruce them up, and submit them to U.S. Chess, who should be glad to accept them and display them.

EDIT: I just looked at one of those. OMG, it was almost 100 GB, and was simply an image of the handwritten wall chart, torn right off the wall at the tournament. OCR, even if it were reliable, wouldn’t help here. So forget about my suggestion in the above paragraph.

Back to the main dream – I hope.

Bill Smythe

US Chess has a number of crosstables from the 80’s and early 90’s in printed form in Crossville.

I tried to include scanning them in my budget a couple of times, to get shot down. I think Carol is sympathetic to the idea, it might even be one that could become a designated project for donors.

I doubt that records of most earlier events (back to the mid 70’s) still exist in paper form and events from before US Chess was computerized are going to be even rarer.

The crosstables in the Cleveland Public Library are in 31 boxes. They are the original rating reports submitted to USCF between 1972 and 1977.

I am currently scanning in all CCA events from these boxes. I can do approximately 1 box per day. To date I have been to the library on 8 different days and I am currently on box 7.

The next time I am going back will be November.

Every CCA tournament that I find eventually gets posted to the CCA website.

It is a labor intensive project. I have to individually scan each page. A page is about 10 players (i.e. 1 wallchart). That is the way CCA submitted events. Large tournaments take me over an hour to scan. Other organizers submitted rating reports. Those tournaments are obviously quicker to scan.

I agree US Chess should try to digitize these records. However, it is very labor intensive and not a fun project.

When I can these documents, I get a PDF file. I am not a computer guy and I have no idea how easy or difficult it would be to somehow get that info into MSA.

On the positive side, I am getting to see a lot of chess history. I am literally sifting through every rated event. I am currently scanning documents from 1973.

Dave Hater
Continental Chess Association

What does “display” mean to you? Just curious.

There are companies that specialized in digitizing paper records, I guarantee you they don’t do it all by hand.

Please let me know if you see any scholastics from 1974-77 in the Capital District of NY?

I’ve split a discussion of league play into a new topic, Website area to support league play.

I meant that, once the crosstables were converted to MSA style, they could be viewed through MSA just as all tournaments starting from late 1991 are.

However, I dropped that idea once I looked at one of the Cleveland Public Library crosstables and found that it was a huge PDF image file, not editable.

Bill Smythe

I just spoke to Judy at the office, she was not aware of the 31 boxes of crosstables from the early 70’s at the Cleveland Public Library. One possibility is that when Jim Meyer was at the office, he cleaned out a lot of old paperwork, he may have sent those boxes to them.

If they were all from events before computerization, there wouldn’t be any IDs on them, since those weren’t assigned until the first computer system was installed in around 1977.

In terms of a “computer system”, what did that mean in the context of 1977 technology? How was data inputted and extracted? How was the data stored?

I wasn’t there, but I think the original computer system was a mini-computer, probably DEC or Data General.

It would have had disk drives about the size of a large pizza box, which probably held somewhere around 5-10 megabytes of data each.

Input would have been via terminals that probably supported 24 lines of text, each 80 characters long. Printers were probably dot matrix printers using pin-feed continuous paper stock.

I know the original programming was done in COBOL.

Those of us who worked as programmers or systems analysts in that era learned to be parsimonious in our use of both expensive disk space and very limited computer memory. These days the art of efficient coding or data storage techniques aren’t even taught.

Thanks, Mike. I remember having to use punch cards to program information for a statistical package that analyzed voting data for a graduate level political science class in 1978. Ugh! Access to the computer was short and rigidly enforced. God forbid you dropped the cards or had one get jammed and torn to pieces. You always had to check for “hanging chads.”

OK, let’s get this thread back on topic.

At the top, I proposed three sets of radio buttons that would be presented to each viewer of an MSA crosstable. The first set of buttons would offer the choice of sections to view (Open, Under-1800, Reserve, etc). The second section would ask for the desired display format. The third would let the viewer choose (within reasonable limits) the order in which players would be listed.

In this post I’d like to expand on the second proposed set of radio buttons:

[b]Display Format:

  • Traditional display
  • Single-line display
  • Double-line display
    [/b]We all know what the traditional display looks like:

[size=150][b][code]

21 | ALBERT ALPHA |2.0 |W 23|W 17|L 6|
NY | 10123456 / R: 1941 ->1952 |N:2 |B |W |B |

22 | BEATRICE BETA |1.5 |D 7|W 19|L 23|
IL | 10234567 / R: 1833 ->1822 | |B |B |W |

23 | GEORGE GAMMA |1.0 |L 21|L 4|W 22|
TX | 10345678 / R: 1725 ->1721 | |W |B |W |

[/code][/b][/size]
But wouldn’t it be nice to display each player in a single line, with no rows of hyphens between players, in the manner you might once have expected in a paper magazine:

… … … … … … … … … … … … … … … … … … . pre . . .post … score. norm. colors. .rd1. .rd2. .rd3
21 Alpha, Albert . … .10123456 … NY . 1941 … 1952… …2.0… …N:2 … bwb … W23 W17. .L6
22 Beta, Beatrice . . 10234567 . . IL … 1833 … 1822… …1.5… … … … …bbw . . D7 .W19 . L23
23 Gamma, George…10345678 … TX … 1725 … 1721… …1.0… … … … …wbw …L21. .L4. . W22

Or, in the case of a dual-rated tournament – or just to reduce the width of the crosstable even if not dual-rated – a double-line format:

… … … … … … … … … … … …pre … … post… .score… …colors… .rd1… .rd2… .rd3
21 Alpha, Albert… … … R:1941… …1952… . 2.0 … … bwb… .W23… W17… .L6
… .10123456… …NY… …Q:1841 … 1852 … . N:2
22 Beta, Beatrice… … .R:1833… …1822… . 1.5 … … bbw… . .D7… .W19… L23
… .10234567… … IL… …Q:1733 … 1722
23 Gamma, George… .R:1725… …1721… . 1.0 … … wbw… . L21… …L4… W22
… .10345678… …TX… …Q:1725 … .1721

I apologize if those don’t quite look lined up. YMMV between one computer and another. Obviously, it could be done better in real life. It’s difficult to simulate HTML-like displays on this forum.

Bill Smythe

Quiz questions regarding the above post:

  1. Why did I lump all the colors together, e.g. wbw, instead of including each round’s color in with the result and opponent number?

  2. Why did I use lower case wbw instead of uppercase WBW?

  3. If the viewer chooses single-line format, but the tournament is dual-rated, where do all the ratings go?

  4. Should columns be suppressed if they are empty? For example, if no colors are listed for any games, should the “colors” column be suppressed?

  5. Should the viewer be allowed to suppress certain columns, even if they are not empty? Examples might include ID numbers, colors, norms, etc.

I’m hoping for some good answers from you guys.

Bill Smythe

COBOL is still my favorite language though I’m trying to pick up JavaScript and Python. I was taught how to write fast programs since my agency paid for computer time by the byte-second, but later structured programming was in vogue when the biggest expense was the salary of the programmers. Easier to maintain, but slower.