This paper presents the results of a research project in Digital Humanities. The project
explores archival documents within the online system Engineering Historical Memory
(EHM). EHM enables users to read primary historical sources according to different
knowledge and expertise levels interactively. Here, as a showcase, the authors present
the interdisciplinary research (Diplomatics, History, 3D Visualisation) that empowered
the online publication of the original privilege issued by pope Gregory X to the
monastic community of Mount Sinai in 1274. The application embodies the following
research tools: 1) Digital recomposition and restoration of the photographic survey of
the original parchment split into two pieces and held in two different locations in
Egypt (Holy Monastery of St Catherine at Mt Sinai, and the Archive of the same monastery
in Cairo); 2) Force-directed graph to navigate the diplomatics structure of the papal
privilege; 3) Interactive geospatial visualisations of the place names (satellite views
and political maps); 4) Search engines for scholarly texts, images, and videos related
to the individual diplomatics, historical, and geographical entities itemised in the
application; 5) 3D modelling of the lost lead seal of the privilege; 6) Customary
web-based visualisation tool for the 3D model of the seal.
Questo articolo presenta i risultati di un progetto di ricerca in Umanistica Digitale
che esplora documenti d'archivio all'interno del sistema online Engineering
Historical Memory (EHM) consentendo agli utenti di leggere fonti storiche
primarie secondo diversi livelli di conoscenza e competenza in modo interattivo. Qui,
come caso di studio, gli autori presentano la ricerca interdisciplinare (diplomatica,
storia, visualizzazione 3D) che ha consentito la pubblicazione online dell’originale del
privilegio concesso da papa Gregorio X alla comunità monastica del Monte Sinai nel 1274.
L'applicativo comprende i seguenti strumenti di ricerca: 1) ricomposizione digitale e
restauro della campagna fotografico sulla pergamena originale che è conservata in due
parti in due diverse località egiziane (Sacro Monastero di Santa Caterina al Monte Sinai
e Archivio dello stesso monastero al Cairo); 2) force-directed graph per
navigare la struttura diplomatica del privilegio papale; 3) visualizzazione geospaziale
interattiva dei toponimi (viste satellitari e mappe politiche); 4) motori di ricerca
accademici per testi, immagini e video relativi alle singole entità diplomatiche,
storiche e geografiche identificate nell’applicativo; 5) modellazione 3D del perduto
sigillo pendente in piombo del privilegio; 6) sistema sviluppato ad hoc per la
visualizzazione sul web del modello 3D del sigillo.
Introduction
In recent years, the influence of visual imagery has advanced, and new tools and
techniques have been increasingly exploited to the benefit of interdisciplinary research
(). Information visualisation has become an essential
and meaningful approach for digital history to explore complex data sets (: 5). It can be construed as a powerful guide for exploring
primary historical information and generating new research opportunities to search in
online secondary literature repositories (). Nowadays,
several visualisation solutions can be tested and successfully used by historians in
their perennial chase for truth (). Therefore, good
information visualisation can help enlighten the relations among different historical
information () to promote the sharing and participatory
interpretation of historical memory.
Since 2007, Engineering Historical Memory (EHM) is an ongoing research project that
welcomes international and multidisciplinary collaboration to design and test
interactive applications for the virtual (re)organisation and delivery of historical
knowledge in the digital age. The EHM research team studies and practises "by what
means" traditional historical scholarship can supply machine-readable information. The
aim is to empower historical sciences with artificial intelligence and machine learning,
thus enabling all users to read primary historical sources according to different
knowledge and expertise levels interactively. In the history domain, EHM makes a
cross-disciplinary use of established research processes, such as mapping as understood
in mathematics and linguistics (i.e., an operation that associates each element of a
given set, the domain, with one or more items of a second set, the range) and parsing as
understood in computing (i.e., analyse narratives into logical syntactic components) to
kick-off the exploration of primary historical sources. Using these operations of
mapping and parsing for individual primary historical sources, EHM associates each
element of given sets of information provided by the domain of the traditional
disciplines (e.g., history, art history, philology, palaeography, diplomatics,
codicology, archaeology, epigraphy, sigillography) with one or more elements of the
range of machine-readable content management systems (e.g., spreadsheets, computational
notebooks).
The level of accuracy of this preliminary human activity is directly proportional to
that of the aggregations generated and visualised by the EHM algorithms from different
sets of similar written or depicted elements in the EHM database (e.g., geographical
names, people’s names, goods, ships, governments, events, architectures, drawings) and
potentially relevant publications, images, and videos retrieved in online repositories.
To understand and emphasise the unity of EHM as an interactive system in the approach to
historical information, the rhetorical and conceptual linking of mapping and
visualisation as well as search and visualisation should be hendiadys.
Thus, in its primary meaning, the term mapping is understood as the localisation
and description of elements, facts or phenomena that relate to a circumscribed area,
historically understood at the intersection of precise space and time coordinates. The
aim is to provide and test an example of an innovative epistemological process to distil
historical data visually. The visualisation process is seen not as a reductive
representation to epitomise and/or illustrate written narratives but as an investigative
tool that the historian (especially one who intends to enter fully into the digital era)
can use for discovering and organising new relationships between objects, in a new
historical landscape where past, present, and future can merge in a democratised whole.
Refer to and .
On EHM, five main types of information visualisation have been experienced as
participatory opportunities for digital history. The application approached some best
practices related to Radial Tree, Treemapping, Force-Directed Graph, OpenStreetMap and
Satellite View in digital history to initiate a discussion. This exercise generated an
information visualisation application that enables scholars to submit website reviews
using an online template ().
In this research context, the archival document has its peculiarity and needs to be
investigated using the methods of three major academic disciplines: palaeography,
diplomatics, and history. Usually, in academic literature, these disciplines explore and
study the documents in research silos. This application can be construed as an answer to
the critical question addressed and discussed in the webinar organised by the Laboratorio di Cultura Digitale and EHM on 22 and 23 June
2020. This consideration is because this EHM application establishes a landmark in
unifying the three multidisciplinary approaches to archival documents and providing a
digital humanities interdisciplinary experience to the user. Refer to Figure
1. Identifying an alternative method for the digital processing of information embedded
in archival documents relies upon the demonstration that archival documents are highly
structured, and traditional disciplines can teach digital technologies how to decode the
content and transfer it into a machine-readable and machine-understandable format. This
process also addresses the second key question of the webinar, which was about creating
digital editions of historical sources that do not respond to a single category of
users, usually coinciding with that from which the edition itself originates. This EHM
application shows how making information machine-understandable allows digital systems
to satisfy different audiences simultaneously and the various queries that come from
individual audiences under different circumstances.
The paper is structured in six sections (excluding this introduction). Section 1
presents the digital recomposition and restoration of the photographic survey of the
original parchment, which is split into two pieces and held in two different locations
in Egypt (Holy Monastery of St Catherine at Mt Sinai and the Archive of the same
monastery in Cairo). Section 2 introduces the force-directed graph to navigate the
diplomatics structure of the papal privilege. Section 3 illustrates the functions and
operability of the interactive geospatial visualisations used for place names (satellite
views and political maps). Section 4 describes the search engines for scholarly texts,
images, and videos related to the individual diplomatics, historical, and geographical
entities itemised in the application. Section 5 showcases the 3D modelling of the lost
lead seal of the privilege. Section 6 explains the development and implementation of the
customary web-based visualisation tool for the 3D model of the seal. The paper ends with
the acknowledgement of research collaborations and bibliographical references.
Screenshot from EHM. URL
https://engineeringhistoricalmemory.com/MountSinai.php?vis=sat
1. Digital recomposition and restoration of the original parchment’s photographic
documentation
The original artefact is presently divided into two pieces: ink on parchment (ca
57x68cm), Ιερά Μονή Θεοβαδίστου Όρους Σινά, Αγίας Αικατερίνης/Holy Monastery of St
Catherine (Egypt), right part (ca 28.5x68cm) in the cabinet of the Archive of the
monastery at Mount Sinai, left part (ca 28.5x68cm) in the strongbox of the Archive of
the monastery in the Metochion (dependency) at Cairo. The overall state of
preservation can be described as fair since the readability has not been affected, even
if the supporting script has suffered minor mutilations that correspond to the
intersections of the folds, three horizontal and three vertical ones, and is now bereft
of the lead seal suspended from a silken thread. On both the front and the back, there
are chancellery notes that document its archival life.
The digital reconstruction is based on the image information provided by the
digitalisation of 27 analogue photographs capturing various sections of the 1274
privilege issued by Pope Gregory X for the monastic community of Mount Sinai.
When described as a different version, it means that the photograph could
either be as follows.
The same shot but colour corrected, rotated, or digitally manipulated in any other way
to increase the visibility of certain features of the papal bull.
A different shot of the same section of the bull, with the same approximate angle and
magnification.
Meanwhile, a detail or zoom shot means a closer shot or a zoomed-in
crop of the same section of the bull, focused on a specific feature or point of
interest. Regarding units with multiple versions, it is necessary to decide on one for
the reconstruction process. Additionally, while the incorporation of detail/zoom shots
is prioritised as much as possible, they may still be omitted when their image quality
is equal to or lower than that of the photograph of the section. Deciding on which photo
is most suitable for the reconstruction process is an exercise in trial-and-error and
inevitably introduced some subjective artistic judgement. However, colour and image
clarity are the two most deciding factors in the elimination process. It is often the
case that unedited versions are preferable with colours as it retains more information
potentially lost due to colour correction. It is also easier to achieve a coherent
overall look with unedited versions than edited ones, which may have exaggerated
unnatural colours or lighting quirks (). How clear or sharp
some images are, is often directly determined by the image size and resolution (). To construct a composite image, the emphasis is put more
on choosing sections that come together seamlessly as possible. Some areas heavily
edited to increase the legibility of faded texts may not be incorporated into the
composite but should still be used as a reference when trying to transcribe or decipher
the writings in question. The necessary steps are then repeated until the final
reconstruction is completed. In the following sections, each step is described in a more
detailed manner.
Photographed by Andrea Nanetti during a palaeographic mission of the Centre
of History and Palaeography of the National Bank of Greece Cultural Foundation
(Director, Agamemnon Tselikas , 19-24 January 1994). Digital reconstruction of the
artefact by Andree Hansen Wibowo (May 2020). Copyright of the Holy Monastery of St.
Catherine at Mount Sinai. The images of the papal bull have been first published with
permission (Rev. Father Symeon , Librarian of the Monastery of St Catherine, and
Agamemnon Tselikas, January 2001) as Tables 3-4 and 7-8 (right part) and Tables 1-2
and 5-6 (left part) in Nanetti , 2008. The images on the left show four of the
original exaggerated shadows. The edited version's intense blue tinges on the right
present a challenge in creating a coherent image composite.
This is an example of decisions that had to be taken in the digital
reconstruction . Although the detail below has a higher resolution and pixel size, it
is apparent that the image above has an inherent greater clarity and would therefore
be a more fitt fitting photograph to be used for the image composite. These images
were used by Antonella Ghersetti and Frédéric Bauden for the transcription,
transliteration, and translation of the archival information in Arabic (refer to [27]
).
1.1. Insert Photograph as Smart Object
This reconstruction leverages the Non-Destructive Editing (NDE) using Smart Object
workflow available using Photoshop (PS from now on). This workflow allows retaining
memory of an image's original state regardless of the number of edits applied to
it.
1.1.1. Match Lighting and Colour
shows that the lighting of the papal bull sections
is varied and uneven, so the best solution is to first adjust the segments'
brightness to a commonly accepted standard. The photographs' non-standardised
nature required multiple adjustment layers to achieve a unified look across all
the different sections. In this adjustment` process, Levels are an advanced
version of Brightness/Contrast. It factors in the image’s brightest highlights and
darkest shadows instead of applying a sweeping brightening or darkening image.
Curves is an even more fine-tuned version of Levels and has the added feature of
allowing users to sample an image to copy its colour tone, which significantly
aids the matching process. A clipping mask was set onto each adjustment layer to
ensure that each adjustment layer only affects its targeted photographic
segment.
1.1.2. Align Photographic Segments
Evaluating the photographed sections shows that adjacent segments overlap and
present the same image information on the papal bull's specific areas.
Theoretically, this makes it possible to position adjacent pieces and adjust them
with each other until the overlapping areas align perfectly. However, perfect
alignment is impossible due to the variance in the original photographs' lighting
condition and lens magnification. Furthermore, the papal bull's physical condition
did not provide consistent shapes across the different photographic sections. This
characteristic makes the blending mode a helpful tool to indicate the degree of
alignment between two photographic segments. If aligned well, we should see a
strip of black where the two pieces overlap. Otherwise, we will see false colours
where the two image layers misalign. This is also why the previous step
standardises the lighting. A difference in the image brightness would be
registered as a difference in colour under the Difference layer blending mode.
Thus, with this Difference blend mode on, the next step was manipulating the
scale, position and rotation of the photographic sections as needed, using each
other as a marching guide. Refer to for an
example.
An example of a well-aligned section (Top) compared to a misaligned
one ( Below ) under a ' Difference ' blending mode.
The photographic segment’s position, scale, and rotation were manipulated. Once
the overlapping areas are satisfactorily aligned, the changes can be confirmed,
and the layer blending mode can be turned back to Normal. This process restores
the photographic segments’ true colours in overlapping areas.
1.1.3. Create a Seamless Transition
After the photographic segments are adequately aligned, there may still be traces
of misalignment or change in image brightness, manifesting as a break between the
two elements. The subsequent creation of a Layer Mask allowed a smooth transition
between the two segments. Layer Mask and masking allows to effectively ‘erase’
certain parts of a photographic segment’s edge while still retaining the image
information if it should be needed back in line with the Non-Destructive Editing
principles. Finally, masking the image with a soft-edged brush integrates the two
image segments with a gradual transition between slightly different brightness or
lighting conditions.
1.2. Global Adjustments
Applying the previously outlined procedures to reconstruct both the recto () and the verso () of the
papal bull, a relatively complete image composite of the provided photographs was
obtained.
Recto (front side) of the 1274 Papal Bull issued by Pope Gregory X,
work-in-progress.
Verso (backside) of the 1274 Papal Bull issued by Pope Gregory X,
work-in-progress.
As shown in and , although
the primary form of the papal bull could be restored, still some imperfections are
visible and compromise the appreciation of the content. Therefore, global adjustments
would be required to tackle these imperfections, meaning that the edits performed in
this stage are intended to affect the photographic segments. Instead of aligning two
or three pieces together, the adjustment aims to improve cohesiveness and coherence
across all the photographic elements. The two main issues tackled in this stage are
the following.
Correct Alignment of Textual Information.
Extrapolation of Missing Paper Edges.
1.2.1. Correct Alignment of Textual Information
On the recto of the papal bull, a closer look at the image composite reveals that
the textual information is not lining up correctly in certain areas due to the
letter's creasing and tearing. and provide two significant examples.
Example of how creasing causes some text to be misaligned
.
Example of how tears cause some text to be misaligned .
The correction of this imperfection requires selectively distorting parts of the
image composite, which could be done in a few steps.
The first step was to determine how much of the image composite should be
distorted. This was done by considering the number of lines that are incorrectly
aligned and the nature of the tear.
A selection will then be made of this part of the image composite. This selection
will create both a layer mask and a duplicate of that area of the letter. See for an explanatory diagram.
The duplicated section is then made into a Smart Object to retain its original
image quality before being distorted and warped to align with the letter's
remaining areas.
Creating a seamless transition between the duplicated section and the remaining
sections after aligning them is done using the same method described in Step 4 of
the Image Reconstruction Process.
This process is repeated for all the misaligned textual information in the recto
image composite until they are corrected.
(A) Area of the letter where selection is made is shown in red. (B)
Layer Mask created using the section in (A). (C) Selection in (A) duplicated as
a new image layer .
1.2.2. Extrapolation of Missing Paper Edges
Due to the photographs' framing, some document edges were not captured in the
photographic sections. This lack of image information results in the image
composite as abrupt hard edges ().
Highlighted in cyan are examples of the straight edges on the recto (
left ) and the verso ( right ) because of missing photographic information in
the photographic documentation .
When mirrored, the verso can add image information to the recto's edges and vice
versa. Rather fortuitously, where the edge is missing on one side, it is present
in the other, and as such, this extra information is enough to extrapolate how the
missing edges might have looked like. The following steps would describe the
extrapolation process using the verso as an example. The first step was to bring
in the recto as a Smart Object layer and reflect it horizontally. The recto would
then be transformed and distorted to fit the shape of the verso best, as shown in
.
The mirrored recto, distorted to fit the form of the verso best and
placed as a layer behind it.
A new layer was then created. The layer was masked to the shape of the mirrored
recto and then used to recreate the verso's colour and texture using a combination
of Stamp Tool and Spot Healing Brush Tool. While the Stamp Tool copies the sampled
part of the image onto wherever the user chooses to apply it, the Spot Healing
Brush Tool uses a process known as Content-Aware Sampling that automatically
generates image information to blend away perceived anomalies based on the
software’s analysis of the image.
1.3. Result of the digital reconstruction
and show the digital
recomposition and restoration of the original parchment’s photographic
documentation.
Recto of the 1274 Papal Bull issued by Pope Gregory X, result
.
Verso of the Papal Bull issued by Pope Gregory X, result .
1.4. Result of the digital reconstruction
The web experience optimisation has been achieved through three meaningfully
consequential actions: split, highlight, store.
1.4.1. Split
To provide easy and fast browsing of high-resolution images on webspace, EHM uses
Zoomify, which is both a format and a
tool for deep zooming into high-resolution images. The tool resizes high-quality
image into different lower resolution images, then splits them into smaller pieces
with a fixed size of 256×256, which can be loaded partially corresponding to how
users interact with the map. For example, the digitally reconstructed image of
pope Gregory X’s parchment has 3876×4335 pixels and is split as follows.
This format allows a smooth, easy loading of the zooming level selected by the
user. At the same time, it safeguards the copyright of the high definition
original image, which is not available.
1.4.2. Highlight
This action aims to collect the coordinates of all items that the user can
select on the original document and activate specific EHM tools and searches. The
coordinates are a list of numbers that describe the points set on the
digital document to identify the perimeter of individual items. The value of the
numbers depends on the resolution of the image. A pair of two adjacent numbers
constitute the abscissa and ordinate of the point. After that, the coordinates are
transferred into a dedicated column of the Google Spreadsheet. The other
information about the item is also saved (e.g., ID, title, commentary, English
translation, etc.).
1.4.3. Store
The image pieces split by Zoomify are then stored on Microsoft Azure virtual
machine (Linux).
2. Force-directed graph to navigate the diplomatics structure of the papal
privilege
The tenor formularis (i.e., the diplomatic organisation) constitutes the
juridical structure of the archival document. It is essential for the historical
reading of the document. This methodology is rooted in the diplomatics teaching
imparted by Paolo Selmi at the School of Palaeography, Archival Studies, and
Diplomatics of the Archivio di Stato di Venezia. However, most historians
seem not to consider its importance because they lack diplomatics training. As a
result, the historical information extracted from archival records does not clearly
distinguish between standard formulas, common to all similar documents, and the
formulas, which are peculiar to the specific iuris actus of the
document.
On the other hand, the palaeographer and the diplomatist don’t find it easy to
retrieve the necessary historical contextualising information to read the iuris actus
with other relevant documentation critically. Therefore, the EHM research team
explored different visualisation methods to provide the historian with a clear and
straightforward explanation of the tenor formularis and the palaeographer
and diplomatist the relevant historical contextualisation. This double and
complementary understanding of the primary source is now on the same page, and
different disciplinary approaches can interact and discuss together.
The choice was to use a force-directed graph drawing algorithm to provide a clear and
straightforward understanding of the tenor formularis. The force-directed
graph is a class of algorithms for drawing diagrams in an aesthetically friendly way
as well. The purpose of these algorithms is to position the nodes of a graph in
two-dimensional or three-dimensional space so that all the edges are of equal length.
There are as few crossing edges as possible by assigning forces among the set of
edges and the set of nodes, based on their relative positions, and then using these
forces to simulate the motion of the edges and nodes or minimize their energy.
Screenshot of the interactive force- directed graph to navigate the papal
privilege's tenor formularis ( diplomatic structure ).
The force-direct graph has been used here to visualise the methodology for the
analysis of the tenor formularis (diplomatic structure) of a papal document
that was taught to Andrea Nanetti by Paolo Selmi in the School of Palaeography,
Archival Studies, and Diplomatics at the Archivio di Stato di Venezia in
1991-1992 (refer to and ).
In the analysis of the tenor formularis, what arises continually is the
similitudes with the privilege of Celestine III in favour of the Canonica
Portuense of Ravenna dated 20 May 1196. This privilege renewed word for word
the preceding ones of Innocent II, 1130-1143, and of Adrian IV, 1154-1159 (refer to
Ravenna, Archivio di Stato, Tabulario Portuensi,Capsa C.
Num. 671, published in : 167-170, no. LXXXVII). This
fact proves that the privilege for Mount Sinai uses the standard formulary put in
place by the Holy See for privileges granted to monastic communities. A brief
description of the tenor formularis is provided in the next subchapters. For
a full appreciation of the document, refer to , Italian,
and in English.
2.1. Protocollum
The protocol (protocollum, r. 1) opens with the intitulatio. The pope
nominates himself without the ordinal number with the title of ‘episcopus’ and
the formula ‘servus servorum Dei’, as usual in documents containing
privileges. There follows in the dative case the inscriptio ‘venerabili fratri
episcopo et dilectis filiis fratribus Montis Synay …regularem vitam professis’. “The
papal letter regularem vitam eligentibus is not found in the old book of
the papal chancellery, namely in the Liber diurnus, but in that of the first
half of the thirteenth century as a typical expression of the grant granted to the
Cistercians. From 25 points of the Cistercian privilege, ten identical formulae are
found in Mount Sinai's privilege. At the same time, a further two are taken from the
kind of papal privileges beginning with the words “In eminenti Apostolicae Sedis
specula”. Refer to : 286. The words
“regularem vitam professis” also constitute the closure of the
inscriptio in the above-mentioned privilege of Celestine III in favour of
the Canonica Portuense dated 1196. Refer to :
167-170). The usual honorary appellations used for bishops are observable
(venerabilis frater) and for all the persons in general (dilectus
filius), where the emperor and king had the address charissimus in
Christo filius and ladies had dilecta in Christo filia (refer to : 40). The protocol closes with the formula of
perpetuity ‘in perpetuum’, which during this period substituted the salutation when
the grant was accomplished, as is the case here, without limitations of time and
place.
2.2. Mesocollum
In the text (or mesocollum), after the arenga (r. 2), which
proceeds from the ‘religiosam vitam’ to ‘infringat’, there is a
long multiple iuris actus, from ‘Eapropter’ (r. 2) to ‘salva
Sedis Apostoliceauctoritate’ (r. 29), in which the juridical act is expressed in the
following arranged clauses.
2.3. Escatocollum
The closing section of the document(known as escatocollum,
refer to : 33-81) commences with the pope’s
subscription, articulated in three separate elements arranged on the same line (r.
33, refer to : 21-22).
3. Interactive geospatial visualisations of the place names (satellite views
and political maps)
The online application for pope Gregory X’s privilege for the community of Mount
Sinai (1274) used satellite view and OpenStreetMap. Satellite view provides image
data obtained by photographing or scanning earth or other planets by various
artificial earth satellites. It allows the contextualisation of
historical-geographical data into the physical world. See an example in . OpenStreetMap is an online mapping collaborative
effort that provides user-generated street maps that can be free editable (). It allows the contextualisation of
historical-geographical data into the present political situation. See an example in
. The combination of these two tools provided a
wealth of geographic data and spatial analysis.
Screenshots of selected properties of the monastery in Cairo and
Jerusalem show the connection between the document and the geospatial
visualisation tools (satellite view for Cairo and OpenStreetMap for
Jerusalem).
4. Search engines for scholarly texts, images, and videos related to the individual
diplomatics, historical, and geographical entities itemised in the
application
EHM has developed search algorithms to aggregate knowledge (). Each item is identified using a relevant Wikipedia page whenever available.
Based on this identification, EHM picks keywords and filters to aggregate knowledge from
leading online repositories about any item selected by the user in the online
application in real-time. Some of the resources are searched through their API
(application programming interface), which is either public (e.g., Europeana, Gallica,
YouTube, Vimeo, Google Images, Bing Images) or shared upon research collaboration
agreement (e.g., Taylor & Francis, Scopus-Elsevier). Other online repositories do
not have the policy to share their APIs for federated searches publicly. In these cases,
the user is prompted to link to the official websites one by one in separate tabs; an
automated query consistent with the selected item is automatically filled to facilitate
the search. Online repositories without API are not included in the EHM search (e.g.,
SmartHistory). In EHM, the search results are visualised under four categories as
follows.
Scholarly publications from Europeana, Gallica, Taylor & Francis, Scopus-Elsevier,
via API.
Images from Google Images, Bing Images, via API.
Videos from YouTube, Vimeo, I Media Cities, via API.
Guided search in other relevant resources that do not share their API publicly (e.g.,
GettyImages, Google Scholar, JSTOR, China National Knowledge Infrastructure, Wanfang
Data, Pelagios, Blackwell)
With most of the resources, search APIs work by looking for input keywords in the
digital documents accessible online (title, author, abstract, full text). Taylor &
Francis also provides a cutting-edge service, which allows searching by concept.
This means that the algorithm also searches the Taylor and Francis databases for
concepts related to the set keywords.
Screenshots of EHM Search Engines for scholarly texts (b, d, f), images (a),
and videos (c) related to the individual diplomatics , historical , and geographical
entities itemised in the online application for pope Gregory X’s privilege for the
monastic community of Mount Sinai (1274). The identifier of the item is the
corresponding Wikipedia page (e).
Within EHM, the user can explore the digital images of primary historical sources and
their critical editions and stay tuned with relevant secondary literature interactively
and in real-time. These are two EHM essential innovations to approach the online
publication of primary historical sources. In EHM, information is organised into two
main streams: texts and images. In the case of Gregory X’s document, the process started
with the text of Andrea Nanetti’s critical edition and commentaries (history and
diplomatics; refer to ), whose information was transferred
into a machine-readable format using Google Spreadsheet as a content management system
and data storage solution because of several concurrent advantages: powerful, popular,
secure, robust, customizable, easy backup, and, finally, the similarity with MySQL
database in table-like construction. A diagram in
illustrates the EHM workflow.
Workflow of the EHM search drawn by Khoi Vu.
5. Digital process for the 3D modelling of the lost lead seal of the
privilege
The parchment used as writing support, quite often, and above all whitened only on the
part on which the text is placed, as is customary in bulls, allows one, together with
the other usual intrinsic and extrinsic characters, to recognise the document as a
solemn privilege and to suppose the presence of a lost leaden seal cum filo
serico (with a silken cord, ), or rather cum
serico, meaning by sericum a silken cord, consisting of numerous single
threads partly red and partly yellow (for the litterae cum serico refer to : 57-71, with footnote 9 at 57-58; for pontifical
sigillography, refer to ; ; ). The bullatores, often illiterate, assigned a
cord of silk or hemp to a document based on the initial letter of the pope’s name: if
cum spatiis et floribus erat (it was with spaces and flowers), they used to
assign a silken cord, and if impleta incausto (full of ink) then one of hemp
(cum filo canapis; refer to ). Furthermore, the
parchment shows the usual folds before the plica (i.e., surrounding cover) of
the lower margin, where the two holes through which the sericum sealed with the
lead must pass are still discernible.
Papal bullae acted as seals on official papal documents to authenticate their
provenance and authority. On the bulls with leaden seals of Pope Gregory X still extant,
we see two faces following the model fixed under Paschal II (1099-1118) and remaining
unchanged after that (refer to : 49). An example of lead
alloy Papal Bulla of Gregory X (date: 1271-76; diameter 37mm, thickness 5mm, weight
39.63g, excellent condition) is in London, British Museum. A .jpg image is
available under CC BY-SA 2.0 on the Wikipedia page of Pope Gregory X, with the note
thanks to The Portable Antiquities Scheme/The Trustees of the British Museum, File
2593. However, according to the British Museum Images Account Manager, any
images available on Wikipedia are not authorised copies (e-mail 24 June 2020).
Refer to .
To offer a complete reconstruction of the document, the research team decide to recreate
a 3D model of the lost lead seal and an online costumery web system to visualise the
model itself on the web, as shown in . The work is based on
different sets of 2D photographs of still extant lead seals held in the Archivio di
Stato di Venezia taken by Gianni Penzo Doria on 20 July 2020. Firstly, to ensure
that the scale is as accurate as possible, the dimensions (3.7cm by 0.5cm) of the seal
were measured. A cylindrical polygon was used to block out the seal's basic shape and
critical features in Autodesk® Maya®. Once the rough mock-up was mapped out, the mesh
and geometry were later imported into the ZBrush® software for further detailing.
Several problems arose from this process, mainly the alignment and the approximation of
the coin to the source images. For example, the reference images were not orthographic
in view.
Furthermore, a side-view image of the seal was not provided, causing the sculptor to
rely mainly on a singular view as reference material. To alleviate this problem, several
back and forth rotating or trial-and-error techniques were used to fit the
mesh onto the source image to sculpt the seal's critical landmarks accurately. Most of
the work was dedicated to carving the letters stamped on the seal. The sculpting of St
Peter and St Paul's facial features was a less intricate process while compared to the
letters. In sculpting, one must keep in mind the location of the ridges and the crease
because the abbreviations themselves were not flat and had a particular extrusion value
assigned individually. The geometry was exported as an ‘FBX’ into Substance Painter,
where textures could be applied. After several colours, the generator achieved its final
‘wear and tear rustic look’, and the mesh was finally exported back into Maya for a
final render, look and showcase.
Screenshots of the ZBrush® interface with the model (20a); the finished
model placed in a realistic environment to test realistic lighting render (20b); the
3D Model with the silk cord (20c); the texture mapping (UV Mapping) (20d). The work
has been done by the NTU-ADM student assistant Yap Wei Wen Marc with the materials
provided by Andrea Nanetti and under the supervision of Davide
Benvenuti.
6. Creation of the customary web-based visualisation tool for the 3D model of the lead
seal attached to the parchment with a silken cord
Initially, the 3D model of the lead seal was uploaded onto Sketchfab™, an online
platform used by content creators to post their 3D assets and make them accessible for
public viewing (refer to : 1-2; and : 157). In the context of 3D rendering, three lights would have to be placed
strategically to achieve this enhanced effect. Subsequently, vignettes, ambient
occlusion and depth of field effects were added to adjust the post-processing before
publishing the model. As far as Sketchfab™ is not open access, EHM opted to create a
dedicated JavaScript code of the model was later copied and prepared for embedment onto
the EHM website.
The costumery creation of a web display platform for 3D object represents a step forward
in making the EHM platform open access as using from the shelf software has limitation
regarding the amount of data that can be uploaded and the number of visualisations that
can be allowed. One major limitation hurdle in creating a customary interface is the
loading time of the object is heavily dependent on the size of the file. The research
team had to strike a compromise between the quality of the display and the time that the
user will have to wait for the object to appear. After several tests run by the EHM
research team, the loading time of the model and page took nine seconds. Hence, more
work needed to be done to optimise the window's position in the user interface of the 3D
model; for example, making the 3D model’s display as an option for the user to toggle
with or testing out a larger polycount and file size. Further troubleshooting was
necessary to test the loading time of the model. This was done by adjusting the scaling
and textures accordingly. The file was later converted into the Graphics Language
Transmission Format or ‘gITF’ as a recommended workflow by the developers at
‘three.js’.
Conclusions
In conclusion, this EHM application pioneers and tests a new set of methods to explore
the complexity of archival documents in a web environment interactively. EHM is free and
open access. Within EHM, the user can explore the digital images of the document along
with their critical editions and stay tuned with relevant secondary literature
interactively. This methodology is rooted in the diplomatics teaching imparted by Paolo
Selmi at the School of Palaeography, Archival Studies, and Diplomatics of the
Archivio di Stato di Venezia. Based on this knowledge, EHM has engaged a
multidisciplinary team for diplomatics, history, digital image reconstruction, web
development, animation, and customary web-based visualisation. This application is a
scalable model for teaching and research projects in the palaeography and diplomatics of
the papal document, history of the relationship among Venice, Rome, and Mount Sinai, and
digital humanities at large.
This interdisciplinary identification of an alternative method for the digital
processing of information embedded in archival documents achieved results so far
unexplored. The theory is that technology should serve traditional disciplines and not
vice-versa. In practice, as fewer coding skills we require to the historian, the better
it is. In the case of the papal bull of Gregory X published in EHM, the juridical
structure of the archival document is made available together with the historical
information in two distinct but interactively communicating datasets. The EHM research
collaborators have access to the semi-structured data from the source. Using the Excel
spreadsheet as a content management system is the sole tech skill required to the
diplomatist and the palaeographer to make their knowledge machine-readable and
machine-understandable.
On this basis, the EHM algorithms follow up and aggregate the information visually. On
EHM, five main types of information visualisation have been experienced as participatory
opportunities for digital history. The application approached some best practices
related to Radial Tree, Treemapping, Force-Directed Graph, OpenStreetMap and Satellite
View in digital history to initiate the discussion. Thus, the historian can rely on a
digital system to be aware of the juridical structure of the archival document and
clearly distinguish between repetitive formulas common to all similar documents and the
clauses, which are peculiar to the specific document under examination. At the same
time, more skilled historians have at their disposal all the tools to go back to the
original document and make an individual (re)reading of all potentially controversial
aspects. The results can be shared on the same platform and generate a discussion
without involving traditional journals that fragment and delay the advancement of
learning.
Therefore, this process also addresses a second critical question about publishing
historical sources without responding only to a single category of users, usually
coinciding with that from which the edition itself originates. EHM demonstrates how
making information machine-understandable allows digital systems to satisfy different
audiences simultaneously because of the different layers of information. Furthermore,
the system can be used in different settings (e.g., portable devices, desktop computers,
large interactive screens, projection mapping) and answer various queries from
individual audiences under different circumstances. Finally, on EHM, the historian
cannot adopt a manipulative selection of information. Individual historians can build
their narrative, write their stories in a system that highlights what has been selected,
neglected, and ultimately how historical information becomes historical knowledge with
given authorship.
Altman, R. 2008. A theory of narrative. New York: Columbia University
Press.
Baaken, G. 1970. Die Verhandlungen Kaiser Heinrichs VI. Und Papst Coelestins
III. Deutscher Archiv 27: 479-513.
Bascapè, G. C. 1978. Sigillografia. Il sigillo nella Diplomatica, nel
Diritto, nella Storia e nell’Arte. Vol. 2. Sigillografia
ecclesiastica. Milano: Giuffrè.
Bertin, J. 2010. Semiology of Graphics: Diagrams, Networks, Maps.
Redland: ESRI PRESS.
Bock, F. 1953 Problemi di datazione nei documenti di Gregorio X.
Rivista di Storia della Chiesa in Italia 7: 302-336.
Birn, J. 2013. Digital Lighting and Rendering. Berkeley, California: New
Riders.
Bolick, C. M. 2006. Digital Archives: Democratizing the Doing of History.
International Journal of Social Education 21 (1): 122-134.
Cheng, X., He, H., Ren, Y., & Ba, S. 2020. Teaching Discussion on
Information Visualisation Design. International Conference on Human-Computer Interaction, 393-404.
Springer: Cham.
Claverie, P. V. 1998. Un aspect méconnu du pontificat de Grégoire X: les débuts
de sa politique orientale (1271-1273). Byzantion 68 (2):
281-310.
Cohen, D. J. 2008. Interchange: The Promise of Digital History. Journal
of American History 95 (2): 452-491. https://doi.org/10.2307/25095630.
Evangelisti, P. 2001. Un progetto di riconquista e governo della Terrasanta:
strategia economica e militare e proposta di un codice etico-politico
attraverso il lessico regolativo-sociale minoritico. In Alle frontiere
della cristianità: I frati mendicanti e l'evangelizzazione tra '200 e '300.
Atti del XXVIII Convegno internazionale della Società internazionale di studi
francescani (Assisi, Centro interuniversitario di studi francescani, 12-14
October 2000), 135-199. Spoleto: Fondazione Centro italiano di studi
sull'alto Medioevo.
Duba, W. O., & Schabel, C. D. 2015. Bullarium Hellenicum. Pope
Honorius III’s Letters to Frankish Greece and Constantinople (1216-1227).
Turnhout, Belgium: Brepols Publishers.
Fantuzzi, M. 1802. Monumenta ravennati de secoli di mezzo per la magior parte
inediti. Vol. 2. Venezia: Francesco Andreola.
Fan, L., Bo, H., Li, F., & Sui, J. 2020. Information Visualisation Design
for Museum Exhibition Panels.Journal of Dongnan Culture 4: __-__.
Frenz, T. 1998 I documenti pontifici nel medioevo e nell'età moderna
[Translation by Pagano, S., & Santoro, V. of Frenz, T. 1986. Papsturkunden
des Mittelalters und der Neuzeit. Wiesbaden-Stuttgart: Franz Steiner
Verlag]. Città del Vaticano: Scuola Vaticana di Paleografia, Diplomatica e
Archivistica.
Herde, P. 1967. Beiträge zum päpstlichen Kanzlei- und Urkundenwesen im
dreizehnten Jahrhundert. Zweite, verbesserte und erweiterte
Auflage. Kallmünz/Opf.: Verlag Michael Lassleben.
Hofmann, G. S. I. 1951 Lettere Pontificie edite ed inedite intorno ai monasteri
del Monte Sinai.Orientalia Christiana 17: 302-303.
Il sigillo nella storia della civiltà attraverso i documenti dell'Archivio
segreto vaticano. Mostra documentaria, 19 febbraio-18 marzo 1985:
testo in italiano con sommario in francese, inglese, tedesco, spagnolo.
Città del Vaticano: Tipografia poliglotta vaticana, 1985.
Laurent, V. 1940. Rome et Byzance sous le pontificat de Célestin III (1191-1198).
In Échos D'orient 39: 26-58.
Pfaff, V. 1974. Das Papsttum in der Weltpolitik des endenden 12.
Jahrhunderts. Mitteilungen des Instituts für Österreichische
Geschichtsforschung 82: 338-376.
Nanetti, A. 2008. L’originale del privilegio Religiosam vitam di papa Gregorio
X per il Monte Sinai (1274). Edizione e commento. In Canetti, L., Caroli,
M., Morini, E. & Savigni, R. (Eds.). Studi di storia del Cristianesimo.
Per Alba Maria Orselli, 235-268, with 13 plates at 209-217. Ravenna: A.
Longo Editore.
Nanetti, A., Liu, D, & Vu, N. K. eds. 2010. EHM – Information
Visualisation for Digital History. Retrieved from:
https://engineeringhistoricalmemory.com/Visualisation.php (Accessed on
2021-2-4).
Nanetti, A., & Cheong, S. A. 2018. Computational History: From big data to
big simulations. In Computational Social Science and Humanities,
337-363. Springer: Cham.
Pharr, M., Jakob, W., and Humphreys., G. 2016. Physically Based Rendering:
from Theory to Implementation. Amsterdam; Boston; Heidelberg:
Elsevier.
Rabikauskas, P. S. I. 1998. Diplomatica pontificia (Praelectionum lineamenta):
ad usum auditorum. VI ed. emendata et aucta (first edition 1964). Roma:
Ed. Pontificia Università Gregoriana, Facultas Historiae Ecclesiasticae.
Rodolico, N. 1990 Note paleografiche e diplomatiche sul privilegio pontificio
(da Adriano I a Innocenzo III). Bologna: N. Zanichelli.
Serafini, C. 1910-1928. Le monete e le bolle plumbee pontificie del Medagliere
Vaticano: Storia della collezione di St. Le Grelle e tavole. Milano:
Hoepli.
Wibowo, A. H., Benvenuti, D., & Nanetti, A. 2020. Entry 'Digital
Recomposition of Pope Gregory X’s Privilege for the Community of Mount Sinai'.
In Nanetti, A., Vu, N. K. (Eds.). EHM – Pope Gregory X’s Privilege for
the Holy Monastery of St Catherine of Sinai (24 September 1274). Retrieved
from: https://engineeringhistoricalmemory.com/MountSinai.php?vis=sat&video=1
(Accessed on 2021-2-4).
Haklay, M., Weber, P. OpenStreetMap: User-Generated Street Maps. In
IEEE Pervasive Computing, 7(4), 12-18. Oct.-Dec. 2008. https://doi.org/10.1109/MPRV.2008.80.
Ricci, S. 1985. Il Sigillo nella storia e nella cultura [Catalog of the
exhibition held in Venice at the Correr Museum between 6 July and 31 August
1985]. Sesto San Giovanni, Milano: Jouvence.
In this section, subsections 1.1, 1.2, and 1.3 are based on the report submitted by
Andree Hanse Wibowo, who was tasked by Andrea Nanetti to digitally reconstruct a
composite image of the document with the aid of an image editing software under the
supervision of Davide Benvenuti. The steps detailed in this report can be construed
as the most accurate and convincing image composite of the 1274 document based on the
available photographs. While not perfect, this reconstruction has done justice at
creating a reliable digital documentation of the original document for any future
analysis and study of this Gregory X’s privilege, until a new digital photographic
campaign is done. Subsection 1.4 is based on the report submitted by Khoi Vu for the
research assistantship supervised by Andrea Nanetti at Engineering Historical
Memory in November 2020.
The first critical edition was provided in 2008 (refer to ; and now, the online edition with commentary in English ). The document has not yet been included in Monasterium.net.
The software used is Adobe Photoshop® (PS). As the document’s writing has a
transverse course (i.e., parallel to the parchment sheet's minor side), the image is
displayed vertically.
This section is based on Andrea Nanetti’s work (;). The force-directed graph has been developed by Khoi
Vu under Andrea Nanetti’s supervision and is now accessible online interactively ().
This section is based on the report submitted by Liu DanYun for Liu, DY., Nanetti,
A., Vu, N. K. (Eds.). EHM – Information Visualisation for Digital History.
Retrieved from https://engineeringhistoricalmemory.com/Visualisation.php (accessed on
2021-2-11).
This section is based on the report submitted by Khoi Vu for the research
assistantship supervised by Andrea Nanetti at Engineering Historical Memory
in November 2020.
This section is based on , , and on the the report submitted by Yap Wei Wen Marc for the student work
supervised by Davide Benvenuti.
This section is based on the report submitted by Xu Peisen for the student work at
Engineering Historical Memory.
![This is an example of decisions that had to be taken in the digital reconstruction . Although the detail below has a higher resolution and pixel size, it is apparent that the image above has an inherent greater clarity and would therefore be a more fitt fitting photograph to be used for the image composite. These images were used by Antonella Ghersetti and Frédéric Bauden for the transcription, transliteration, and translation of the archival information in Arabic (refer to [27] ).](https://umanisticadigitale.unibo.it/article/download/12567/version/10877/13099/49604/image4.jpg)
![Lead seal of Gregory X ( cum filo canapis , with hemp cord ). – Obverse ( left image) . On the left , St. Paul is depicted with full facial features, long straight hair , and a beard . On the right , St. Peter is also shown with complete facial features and pellets from his curly hair and beard. Above the heads, the saints’ abbreviations are still partially visible, with ‘SPA’ for St Paul to the left and ‘SPE’ for St Peter to the right. Both heads were separated by a tear-drop formation of linear beads, which surrounds the head motif with a cross pattée supported on a staff between them. – Reverse . GRE/GORIVS/PP X’. The description is based on [35]: 156, crediting Venice , Archivio di Stato, SS. Felice e Fortunato di Ammiana , in Procuratori di S. Marco ‘de supra Canale’, n. 136) © Archivio di Stato di Venezia, Procuratori di San Marco ‘de supra Canale’, b. 136 ( Gregorius X, bulla plumbea cum filo canapis ); digital images: courtesy of Dott. Gianni Penzo Doria.](https://umanisticadigitale.unibo.it/article/download/12567/version/10877/13099/49620/image19.png)
