I. GENERAL INFORMATION
Prepared and supplied by Harrison Eiteljorg, II
Date: December 2003 to January 2004
Title: OprxivDoc.doc (or OprxivDoc.txt)
The older propylon on the Athenian Acropolis was investigated by Harrison Eiteljorg, II, in 1975 and again in 1987 and 1989, thanks to the permission granted by the Greek authorities and the good offices of the American School of Classical Studies in Athens which assisted in obtaining permission. Changes in the files included with this document continued to occur until January of 2004. The initial work was funded personally by Eiteljorg. The preparation of the material for archival preservation was funded by the Center for the Study of Architecture.
The remains examined were those related to the Acropolis entrance between the end of the Bronze Age -- the end of the thirteenth century B.C.E. -- and 437 B.C.E. when the Propylaea was begun. However, the only studied material relevant to the period prior to the middle of the sixth century was the fortification wall, and the portion studied was that part of the wall that survived in the entrances of the sixth and fifth centuries and remains standing today.
The project was an excavation in the sense that the materials composing the bulk of the older entrance had to be uncovered. However, the area had been excavated at least twice before and probably three times. A portion of the area had been covered with a pavement to keep run-off water from reaching the foundations of the Propylaea; so that part of the site could not be examined. As it happened, the previous excavations had not reached bedrock in one crucial area.
The older propylon was also studied thoroughly by William Bell Dinsmoor, Jr. over a long period of time. The Eiteljorg publication, The Entrance to the Athenian Acropolis Before Mnesicles, appeared in 1994, and the Dinsmoor publication, The Propylaia I: The Predecessors, in 1991. To the extent that terminology is important with any of the materials here, the Eiteljorg work should be consulted.
All the documentary evidence from Eiteljorg's work has been assembled by him for archiving by the Archaeological Research Institute (ARI) at Arizona State University. Please note. The archival materials have been moved to the Archaeology Data Service, located in York, England. The ADS' website is at http://www.archaeologydataservice.ac.uk. Files concerning the older propylon project may now be found there and are no longer at the Archaeological Research Institute. That evidence is now entirely in computer form, though the original work was done without anticipation of the use of computers. The computer files generated in the course of the work and others generated for archival storage are defined and described in this document. All were made by Eiteljorg or under his direction. They include scans of drawings and notes made on site, scans of slides taken on site at various times, a CAD model which was constructed over time, beginning in 1985 and only concluded in 2003 (with the addition of links to external data), and data tables constructed to augment the CAD file. There are also some black-and-white negatives that have been scanned; they were used for photogrammetry work on the Mycenaean wall in 1989. Finally, there is a data table to identify the scanned photographs. Of course, this document is also a part of the archival data.
An explicit statement of copyright has been placed in the CAD model by CSA and should show upon opening the model. It reads,
"Copyright, Center for the Study of Architecture, 2004
"CSA holds the copyright to this document/model. The right of any scholar to access and use the model for true scholarly purposes is expressly granted. In addition, CSA expressly grants to the Archaeological Research Institute (ARI) at Arizona State University the unlimited right to use the model for true scholarly purposes and to permit other scholars to access the model for scholarly use. Commercial use of any kind is only permitted with the written permission of CSA; commercial use includes any use for publication or public lectures that are not free to all."
Should CSA cease to exist, the copyright and all rights to limit or control usage of the model will be vested in ARI.
Data tables are free for use without restrictions. Images are restricted by the terms stated above.
The first use of computers for the work was the creation of a CAD model. The survey work on the older propylon was accomplished with tape measures, line levels, carpenter's squares, and plumb bobs. A transit was used to take levels. No other sophisticated survey instruments were used. (The material south of the SW wing of the Propylaea was not independently surveyed by Eiteljorg; material there was included in the CAD model through the use of Dinsmoor's drawings.) As a result, data entry was primarily by using simple geometric shapes and altering them as required by the measurements.
In 1989 photogrammetry was used to survey some of the stones of the Mycenaean fortification wall. The data from that survey work was put into the model with explicit coordinates. Holes and cracks in the upper courtyard of the older propylon were entered into the model with the use of AutoCAD's plane transformation process.
All measurements are in meters. All the blocks of the primary study area -- the area lying within the external corner of the Propylaea formed by the southern wall of the central building and the eastern wall of the SW wing -- were measured to the nearest mm., and dimensions obtained from the model should be taken to be precise to the mm. Dimensions of the blocks of the Mycenaean wall, obtained via photogrammetry, should be taken to be precise to the nearest cm., as should dimensions and positions of cracks and holes, which were placed with the aid of AutoCAD's plane-transformation process. The material in the secondary study area -- south of the SW wing of the Propylaea -- was not fully measured by Eiteljorg. Dinsmoor's drawings and notes were used to create the portions of the model pertaining to that area. Precision should not be assumed to be higher than the nearest 3 cm. because of the dependence upon scaled drawings for data transfer.
There is a possibility that Eiteljorg will continue to modify some of the files remaining in his possession. If he does so, the archive will be provided with updated information. Interested scholars, however, may wish to contact Eiteljorg (see http://csanet.org to find his address or write c/o CSA, P.O. Box 60, Bryn Mawr, PA 19010) to learn if there have been updates not yet communicated to the archive. Other scholars who use and augment any of the materials should also provide the new or changed data to the archive.
The CSA Propylaea Project may also generate information relevant to the older propylon. It may be consulted for such assistance at http://propylaea.org.
Other software used for materials archived includes MS Word (Word 97, Windows version - for this document), OpenOffice Writer (version 1.0 - for this document), FileMaker (versions 4.0, 5.0, and 6.0 for data tables), MS Access (Access 95 and 97 - for earlier versions of some data tables), and PhotoShop (release 5.0, Windows version - for scanning paper records). Other, unspecified software was used for the scanning of photographs (by an outside contractor, Luna Imaging of Culver City, CA). The standard ASCII character set (ISO 8859-1, Latin 1) and English have been used throughout.
II. CAD MODEL
The AutoCAD model, oproprxiv.dwg (file format for Release called AutoCAD 2000 and a second copy in R14 format written by AutoCAD 2000) began in a very early form of AutoCAD, probably version 10, but I no longer remember. Much of the model had been created previously in another CAD program (ARRIS) and portions in still other early CAD programs on Macintosh and PC machines, but the AutoCAD model was constructed entirely from the data; no transfer of information from the prior program was accepted.
The file has been migrated automatically from version to version of Autodesk's DWG file format via transparent AutoCAD processes. AutoCAD 2000 is not the most current version of the program, but it is the most current version in use at CSA. (CSA, the Center for the Study of Architecture, has been the organization directed by Eiteljorg since its founding in 1985; see http://csanet.org.)
Database linkage has been tried, and various rendering possibilities have been tried. There may be artifacts of those two processes in the CAD file, but there are no database links currently maintained in the AutoCAD file except as noted below (no implicit links at all), and there are no intentional data items for rendering.
Line colors and types have no meaning by intent. Only layer names have been used to convey information about the blocks that have been put into the CAD model.
The 0,0,0 location is roughly at the juncture between the south wall of the upper courtyard of the older propylon and the parastas wall that rests on the steps leading from lower to upper courtyard. "Drawing north" is aligned to the parastas wall. There is a second coordinate system (in AutoCAD parlance, a user coordinate system) called north that aligns the y-axis of the model with the north arrow (with the same 0,0,0 point as the base coordinate system), making the model align itself properly with the compass points. (It should be noted, however, that the north arrow has not been carefully aligned via computations and may be off by a few degrees.) There are no other user coordinate systems saved with the model.
There are scripts for use with the AutoCAD model; they are supplied in the same directory as the model itself. Running the scripts will result in the display of certain portions of the model (layers) while others are suppressed to provide specific illustrative needs. Their names are meant to be helpful. Those beginning AR refer to one of the archaic periods; those beginning CL refer to one of the classical periods of the entrance. Following either AR or CL are Roman numerals to indicate the phase within the archaic or classical periods; the M or P following indicates whether the selected material is for a plan view only or a fully modeled view. There follows an I, 1, or 2. I indicating only the in situ material will be included; 1 indicates the Eiteljorg reconstructions will be included; 2 indicates the Dinsmoor reconstruction will be included. (Examples: ARIP1 will display the first archaic period -- the period following 560 B.C.E. -- as a plan only and will include Eiteljorg's reconstructions; ARIIM2 will display the second archaic period -- 490-480 B.C.E.-- as a 3D model with the Dinsmoor reconstructions.) Unfortunately, Windows may misidentify script files as screensaver files; they are simple text files that can be read by AutoCAD, and the misidentification should have no impact on their utility.
There are also prepared views. Some are named with letters of the alphabet to indicate the vantage point (see below); all use the 0,0,0 point of the model as the aim point of the view. Others cannot be easily defined; they are not important enough either to remove from the model or to define fully here.
There are two external data tables, one with data items that apply to individual blocks in the model and one with comments about the model more generally. The data items about blocks are in a data table called OPattDATA.FP5 (a FileMaker file -- also available as a tab-delimited ASCII file called OPattDATA.txt). The structure of the table is discussed below. Model layers named with a D in the first position contain icons that indicate the blocks for which data items have been supplied and the number of the individual data item.
Notes of a more general nature are contained in another table (OPcomments.FP5 or OPcomments.txt). Layers named N plus a numeral (e.g., N1, N2, etc.) have icons and numbers to indicate the portions of the model discussed and the comment number of the database entry.
Any user may refer to the appropriate CAD layer(s) to find the references to external data table entries. Similarly, a user of the data tables will find there the names of the layers needed to see which items are referenced. This system --icons in the CAD file with explicit links to data tables --is intended to make it possible for the files to be migrated in the future to new formats without fear of losing information.
Crucial to the use of the CAD model is an understanding of the layer-naming system used. Each layer is named with a string of characters --18 in all --and each character or group of characters indicates an attribute of the material on the layer. (This system both informs and is governed by the CSA Layer Naming Convention which may be found on the CSA Web site, http://csanet.org/inftech/csalnc.html. The CSA Layer Naming Convention is also included here as a document in the archived materials, though the version at the Web address may be more recent.)
The names may be parsed with this list of characters and attribute meanings.
Data links and annotation links are icons with numbers and lines to indicate the point or area of interest.
Cracks and holes may apply to plan and/or model layers depending on the surfaces affected.
Labels are for printing out plans.
Character Two: "in-situness"
Character Three: the general type of the finding area, such as a public area, a domestic area, a military area, and so on
Character Four: the specific part of the area
Examples: Using MF as the third and fourth characters of a layer name indicates a military fortification; PR a road in a public area.
Characters Five and Six: usage (general and specific respectively). The first of the two letters signifies a broad category -- e.g., wall, stair, or pavement for a structure; the following character narrows the range of the first -- e.g., retaining, foundation, or surrounding for wall.
Examples: Using MFWS as the third through sixth characters of a layer name indicates a surrounding wall of a military fortification; RESS a step of a stair in a religious entry.
The following 5 characters (nine through thirteen) indicate a starting date for the material in question, with a beginning negative (in place of a leading zero) to indicate a date B.C.E. Dates are often conventional, not intended to be taken as literal accurate-to-the-year indicators. Consulting the Eiteljorg volume (1994) is recommended for full discussions of the dates of construction and destruction in this area.
The following 5 characters (fourteen through eighteen) indicate an ending date for the material in question, with a beginning negative (in place of a leading zero) to indicate a date B.C.E. A date of 09000 here indicates something still in place and now taken to be part of the modern environment. Dates are often conventional, not intended to be taken as literal accurate-to-the-year indicators. Consulting the Eiteljorg volume (1994) is recommended for full discussions of the dates of construction and destruction in this area.
Examples: PIREWVBO-0489-0478 = Plan version, in situ material in a religious entrance, wall veneer of cut-stone poros masonry, built in 489 B.C.E. and replaced in 478 B.C.E. MMMFWSRL-120009000 = Modeled version, in situ but disturbed by natural events subsequent to abandonment, military fortification surrounding wall, made of roughly-shaped limestone, erected in 1200 B.C.E. and still standing.
III. DOCUMENTATION OF THE LAYER NAMES
A data table called OPlayers (as both FP5 and TXT files -- the latter tab-delimited) contains information about actual layers in the model. It was constructed by Eiteljorg in the mid-1990s (in MS Access 95 and Access 97 and then in FileMaker 5.0 and 6.0 with tab-delimited ASCII files used as export/import format for moving data from Access to FileMaker) and finished in December of 2003. Work on this was done entirely by Eiteljorg. The table presents all layer names beginning with M, C, H, D, L, E, or N so that a user of the model may determine the nature of the material shown on the layer in question. Layer names beginning with P (plan-only layers showing the material modeled on layers beginning with M) are not included in the data table. The names may be parsed by comparing them to names beginning with M or by using the description of the layer-naming system provided above.
The system for naming layers is very rigorous and is explained above, but this table shows the actual layers made and used.
As implied above, some names do not fit the system. Layer E holds only elevation contour lines. Layer EC contains only the labels for those contour lines. (No other layers begin with E.) Layers beginning with L are labels and are named in self-explanatory ways, though LVIEWS may be unclear. That layer contains a circle showing the locations of vantage points for prepared views (with the 0,0,0 point always being the aim point). Thus, invoking the AutoCAD VIEW command to call up stored views requires consulting this layer (and at least a few plan layes for orientation) to understand the orientation of the stored views. (C or CL will be appended to view names to indicate close-up views.)
Layers beginning with N (plus a digit) contain annotation icons (see below). There are several such layers, but the names are not meaningful; the digit has no importance. There are simply several layers to prevent overlap of icons.
The layer Z is an empty layer that can confidently be used as the current layer in a program like AutoCAD. It will always be empty and therefore has no affect on any view when it is made the current layer. (Those who understand the way AutoCAD treats layers will understand the value of such a layer.)
Layer 0 is AutoCAD's default layer and cannot be removed. The copyright statement is on that layer and should show on opening the model.
Some other layers have been created by AutoCAD and cannot be removed. They are ACADASE, AME_FRZ, and ASHADE. Those layers contain no model entities.
Documentation of OPlayers Table (OPlayers.fp5 and OPlayers.txt)
This table was created by Eiteljorg in Access 95, migrated to Access 97 format and then transferred to Filemaker (4.0) via tab-delimited export files. It was then migrated to the 5.0 format (which is used by version 6.0 as well). It was first created in the mid-1990s and completed in December, 2003.
The table uses the standard ASCII character set.
IV. DATA TABLES LINKED TO CAD MODEL
There are also two data tables for information explicitly linked to the CAD model. These tables were originally constructed by Eiteljorg as experiments in the use of attached data tables for CAD models. The tables were originally created with MS Access 95 and modified in Access 97. As it became apparent that the attachment system in AutoCAD was not robust and would not withstand migration well, the tables were moved to FileMaker (originally 4.0 then 5.0, the format for which is also used by 6.0) via an export/import process using tab-delimited ASCII files. After the transfer to FileMaker, the linkage made explicit with drawing icons and numbers (shown as text in the CAD model) rather than implicit with direct linkage from model elements to data table entries. The tables were last changed in December of 2003. All work on them was done by Eiteljorg.
1. OPattDATA.fp5 or OpattDATA.txt This table contains information about specific blocks. As a result, the information in any entry will apply to material on one layer. Therefore, the layer used for the icon linking the note to the block is the same as the layer for the block except for the substitution of the M as the first character with a D. (Correspondence to plan layers would not always be exact, since all icons are intended to show the 3D objects referenced in the data table.)
2. Opcomments.fp5 or Opcomments.txt This table contains information of a more general nature, information that may explain conditions or otherwise apply to an area rather than a particular block.
V. PHOTOGRAPHS
Selected slides and black-and-white negatives from the project were scanned by Luna Imaging (Culver City, CA). Experimentation by CSA had shown that company to be reliable and the quality to be excellent, but I can supply no information about the scanning procedures, hardware, or software. Some of the color slides were scanned in the latter part of 2000. The others were scanned in the summer of 2003. All black-and-white negatives were scanned in the summer of 2003. All the slides were taken by Eiteljorg, the slides with a Nikon F series camera and various lenses, including an 8 mm. fish-eye and a perspective-control 35 mm. lens. The black-and-white negatives were taken with a Mamiya RB67 camera using 120 or 220 film to produce 6 x 7 cm. images.
As stated above, scholarly use of these images is permitted. Commercial use requires prior written approval.
There is another data table to provide information about the scanned photographs. The slides and black-and-white negative have been described in that table.
OPphotos.fp5 or OPphotos.txt
A data base generated by FileMaker Pro 5.0 v 1 to contain data on the scanned photographic images taken of the Older Propylon during Harrison Eiteljorg, II's research and creation of a CAD model. All scanned images are in uncompressed TIFF format. The TIFF file names can be generated from fields f1+f2 (examples "op00001.tif", "oppg6401.tif"). Not all photographs generated during the project were scanned or archived, only those deemed to have sufficient uniqueness to qualify for archival preservation. The data table was first created in October, 2000, and last changed in December, 2003. At the time of the last alteration, Eiteljorg used FileMaker 6.0, but the file format was not changed as a result.
This table was created and maintained by Susan C. Jones and last modified by Eiteljorg
There are a total of 151 records for 105 scans of slides and 46 scans of medium-format black-and-white negatives.
VI. SCANNED DOCUMENTS
All images were scanned with PhotoShop 5.0 and an HP OfficeJet R80, set to gray-scale at 300 dpi, except where noted. The resulting scans were cropped and saved as uncompressed TIFF-formatted files (in PC order, a PhotoShop option). This work was done in the fall of 2003.
All file names are in the format op99xx.tif, where "op" is older propylon, "99" is a rough chronological sequence of 01 to 99, and "xx" is the type of document (co= correspondence; fn= notes, both taken in the field and appended during the modeling process; fd=field drawings; rd=redrawings; and pe=post-excavation material). There are a few files with an "a" as the 7th character in their name; these images are made from xerox copies of the op99xx.tif images that have additional notations or measurements on the copy.
Most files are from 1975; files from later occasions include the year in their name after the document type -- op99xxYY.tif, where YY is year (1987 or 1990).
Notes: 1. None of the drawings is to scale.
2. Some scans were originally misclassified as redrawings when they actually contained modeling notes. Those files were renamed and reclassifed with a file name of op1##fn.tif; their original name would have been op##rd.tif.
Documentation was carried out by Susan C. Jones, beginning July 1, 2003, and augmented by Eiteljorg in December of 2003. There is no data table for these scans; the following list serves as the index.
List of scans and content:
Field drawings
Redrawings
Notes
Correspondence
VII. DIRECTORY STRUCTURES
Although the archival repository will structure these materials in ways appropriate for their purposes, we have supplied them as follows:
On one CD:
On one CD:
The photographs could not be put on a single CD because they were too numerous. Three CDs with color images only were supplied along with one CD with both color and black-and-white images and two with only black-and-white images.