THL Toolbox > Places & Geography > THL Place Dictionary > THL Place Dictionary Editor’s Manual

THL Place Dictionary Editor’s Manual

Contributor(s): David Germano, Andres Montano, Steven Weinberger

I. OVERVIEW

The THL Place Dictionary allows users to easily find information on Tibetan places and geographic features through searching, browsing, or the use of a THL Feature ID. Once a feature is identified, you can then view the basic information on that feature, view it on an interactive map, consult possible extended descriptive information, and see associated THL resources relevant to the feature (images, audio-video, texts, and so forth). These feature descriptions are created using THL’s online editorial interface for the Place Dictionary, which is explained herein. The manual is organized from the perspective of a first-time user, and provides each step in a logical sequence to introduce editors to the Place Dictionary.

II. Steps in Creating a New Place Dictionary Entry

1. DECIDE WHETHER TO PREPARE ENTRIES OFFLINE OR ONLINE

If you are simply entering the occasional item, then the best solution is to use the Place Dictionary's online interface. Even if you, we strongly urge you to prepare descriptions off line, and then copy and paste them into the Place Dictionary when finished, in order to avoid any accidental loss of work.

However, if you have a large number of items, you may be better served by entering your data into a spreadsheet, and then having them imported into the Place Dictionary. No matter how user friendly an online interface it is, an offline spreadsheet is also far easier for large amounts of data. See THL Place Dictionary Submissions by Spreadsheet Instructions for details on using a spreadsheet to prepare material offline. The spreadsheet can be very complex if you are entering all the variant names, sources, and so forth; but our standard spreadsheet template can also be reduced as you like to just the fields you need to enter. We suggest you contact us for advice before getting started if you want to do a lot of work by spreadsheet.

For a general discussion of workflow processes in THL – whether and when to use spreadsheets, word processing documents, the online editorial interface, or some combination, you can also see Contributing to THL Offline.

2. GET A THL USER ID AND AUTHORIZATION TO EDIT THE PLACE DICTIONARY

Editing the Place Dictionary requires a THL User ID with authorization for the editing the Place Dictionary. If you don't have one, contact THL to request an account.

3. LOGGING IN

1. Open up a browser window and type in the address "places.thlib.org". Alternatively, you can go to www.thlib.org, click on the PLACES tab, and then click on the "places dictionary" link that appears to the right.

You can also bookmark the actual Place Dictionary editing page at http://places.thlib.org/admin/features, in which case it will take you directly to the login screen (step #3), and after logging in, you will find yourself delivered directly to the editing home page (skipping steps #4-5).

2. Click on the upper right hand corner “login” link. This gives you the login screen.

3. Type your THL user name and password in the corresponding fields to login.

4. Once logged in, the upper right hand corner will show your name, and the page overall will show the Place Dictionary home page for end users.

5. To begin editing, click on the “show menu” in the upper right hand corner, and choose “edit” in the top set of options. This will cause the editorial home page of the Place Dictionary to appear in the main pane of the browser window to the left.

4. DETERMINE IF YOUR FEATURE IS ALREADY IN THE PLACE DICTIONARY

The first step in creating a new feature entry is to determine whether or not there is already an entry for that feature in the THL Place Dictionary. If so, you should edit that entry; if not, you will need to create a new entry. The best way to determine this is by searching. To search for a feature entry, put your search criteria into the search box on the top on the editing home page. You can search in any script (Tibetan, Chinese, etc.) that is used in the corresponding feature entry for the feature you are seeking. In other words, for the city of Lhasa, you can use Tibetan script, Chinese characters, or roman letters if those languages and scripts have been used in detailing the various names of Lhasa. All non-roman scripts should be entered using Unicode.

You can also browse if you are unsure about the form or spelling of the name. The Place Dictionary home page provides browsing on the right-hand side through reliance upon contemporary nation state administrative units. The browsing initially shows the top level categories – which for the contemporary nation state perspective is nations. Categories that contain subcategories indicate this with a plus sign (+) to the left of the category name. If you click on that category, the + sign changes to a minus sign (-) to indicate it is expanded, and its contained subcategories are detailed in an indented list below it. At the same time, the corresponding feature entry for the expanded category is displayed in the main pane of the browser window to the left. This provides a convenient way to explore a given perspective from which places are grouped together. Please note that the end user presentation is far more friendly than the editing page for browse functions, so you may want to look for your feature using the end user view of the Place Dictionary if you are need to search more carefully.

If you find a feature entry for the feature, then it will give you two choices – “view” or “edit.” Click on edit to start editing the entry. If you don’t find an extant feature entry, then proceed to the next step to create a new entry. All steps for editing are the same as for creating a new entry after this point, so the remaining instructions apply to both.

Deciding Whether to Make One or Multiple Features

One of the complex issues in documenting geographical features over time is when and if to decide changes have created a new geographical feature - which then deserves its own entry in the Place Dictionary - or whether these changes can simply be documented within a single Place Dictionary entry. This is a very difficult issue for which there is no "right" or "wrong" solution. Obviously geographical features change all the time - in name, type, location, relationships, attributes, and so forth - and thus it is rather arbitrary to decide that certain changes determine that it has become a "new" feature. Hence an initiative like ours simply has to establish its own guidelines, document them, and adhere to them. These guidelines are at present in a state of evolution, and rely on examples, though we anticipate that in the future they will become formalized into more explicit principles. We welcome you to contact us with questions or suggestions.

Change in name: a simple change in name by itself will not typically be sufficient criteria for establishing a new feature.

• Examples: a village or monastery is given a completely new name

Change in location: a change in location by itself will not typically be sufficient criteria for establishing a new feature.

• Examples: a village or monastery is moved across the river, but retains the same name and is understood by locals as being the same village

Change in feature type: a change in feature type may or may not be sufficient criteria for establishing a new feature. When a monastery becomes a granary for 10 years during the Cultural Revolution, we simply indicate that change in feature type for the same entry that documents its previous existence as a monastery. If a "town" grows in size and becomes a "city" according to some national classification system, we will also simply document that change in one feature entry.

The most complex issue, however, is political changes over time. Examples are when a kingdom is dissolved, and the territory is incorporated into a new polity, whether a historical polity or a modern nation. In such cases, often the name of the kingdom is preserved as the name of an administrative region within the new polity, and the territory of the new region may be generally identical to the old kingdom. In such cases, our policy is generally that a wholesale shift of political arrangements will engender the new administrative region being treated as a new feature distinct from the old kingdom. To clarify the relationship between the two, we use the "is succeeded by/succeeds (is.succeeded.by/succeeds)" relationship type with the perspective set as "political Relationships (pol.rel)" perspective. Of course there is a spectrum of possibilities here, and such political changes do not always necessitate the creation of a new feature entry for what results from the change.

On the extreme side is when a historical polity of some type gets incorporated into a modern nation state - in those cases, the new national administrative unit is considered a new feature. In general, we consider major political shifts historically in the same way, such that a similarly named and similarly located region in two totally different empires or kingdoms, for example, would tend to result in two different features. However, in many cases the political status of a region is not entirely unambiguous, and political shifts may go back and forth. A kingdom may be more or less independent, then more or less become a type of region under the control of others, and then again become more or less independent. In these cases, our inclination is to stress the continuity and document these all as shifts for a single feature. We will add specific examples to the documentation over time.

Finally there is the issue of "cultural regions" and their relationship to political regions. A cultural region is marked by people's perception of a region having a distinctive character, often including language, clothing, architecture, means of self-identification, and the like. Often such cultural regions share a name, and even a rough territorial extent, with a political region - whether a self-sufficient independent polity or a subordinate administrative unit of a larger polity. Indeed, the boundaries between the two can be hard to ascertain, since a cultural region may stem from a long-term political arrangement, or the cultural region and political arrangement may historically evolve together. In addition, whether a specific region is something we want to classify as a cultural region, rather than seeing it as chiefly a political region that shares certain traits due to the long-standing political uniformity, is often very hard to address. In general, our principle is to treat cultural regions and political regions - even when the name and territory are identical or near-identical - as entirely distinct features. The reason is that firstly the attributes of each are quite distinct from each other, and they co-exist in time. Thus it can be very confusing if you are looking at the feature from the perspective of it being a cultural region, and seeing all these attributes and changes that relate to the political entity. In addition, the cultural region has a more persistent and stable identity, whereas the polities can change dramatically over time. Thus this would entail the cultural region being split into multiple regions simply because its linked political entity is changing.

Change in attribute(s): of course attributes of a place change all the time, and the vast majority of these changes will not cause us to reclassify the feature in question as a new feature. This includes some changes that might raise some questions, such as a change in sectarian affiliation of a temple, for example. We anticipate some changes in attributes - attributes other than feature types - might cause us to create a new feature, but have not yet come up with an example.

5. CREATING A NEW PLACE DICTIONARY ENTRY

1. To create a new place dictionary entry, click on the “New” link to the upper right. This offers a new THL Feature ID (F###), which it shows on the next screen. The only thing inherent to the feature per se is an ID. All other information is stored in separate objects – feature name, kind of feature it is, etc. Thus you first create the feature to make an empty skeleton with a unique identifier. Each ID begins with a “f,” which stands for “feature,” and distinguishes it from other THL IDs. When a new feature is created, the program automatically assigns it the next available number. You can then add additional components to this skeleton: “names,” “feature types,” "geocodes", “locations", “feature relations,” and "descriptions". Each instance of these models is intrinsically connected to a single feature object, and cannot be linked to another feature object.

2. This first screen allows you to specify whether you want the item to be public or not. By default, the public option is checked, which means that the new entry will be seen by the public. Unchecking it will means the entry will not be public until such time you recheck it. Click on the “create” button to create the new entry.

3. The next screen provides the home editing screen for a place dictionary entry. It provides a representation and link to editing for each of the seven components of a place dictionary entry in the following order:

1. General Information: just the THL ID and a checkbox to make the entry public or not.
2. Name: in all variant forms.
3. Type: as specified from the THL feature thesaurus.
4. Geocode: any codes or identifiers by which this place is referred.
5. Location: the latitude/longitude and altitude (elevation), as well as any narrative notes about the feature’s location and sources of the data.
6. Caption: general-level information on the feature in up to 140 characters (spaces included).
7. Summary: a presentation of the important aspects of the feature in up to 750 characters (spaces included).
8. Description: one or more essays of any length (no minimum or maximum number of characters).
9. Illustration: one or more images associated with the feature.
10. Feature Relations: specification of other places/features to which this one is related in some way.

For a new entry, each of these components will just have “new” links to create a new item for that component. Once items are created in a component, the items are listed with links to “edit” them on this home editing screen. We recommend you right click on “new” or “edit” from this screen and open them up in a new tab. Then when you have completed creating or editing the item, you just close the tab to return back to the home editing screen.

All these components are repeatable, such that you can add an infinite number of items for each component for a given place.

Each item should be given a date or date range for which that item is valid. For example, when was this name applicable to the feature, for features that changed names over time. For, when was this feature a “village”, and when did it become a “town” in terms of feature types?

Finally, many components can have a source specified, such as the text that was the source of the information in question, or a narrative note about a conversation that was the source of the information.

All associated data that is connected to this place – bibliographies, audio-video, images, immersive objects, related texts, and so forth – is not entered into the Place Dictionary entry, but rather is automatically collated by virtue of those resources having been cataloged in their respective repositories as being related to this place by means of its THL Feature ID.

Also see THL Participatory Tibetan Township Documentation Project for a subset of instructions relating to our efforts to find volunteers to help document Tibetan townships in more detail, which essentially is the smallest administrative unit in contemporary China and hence is a project at documenting local villages, mountains, rivers, monasteries, schools, and the like.

Also see the THL Tourism Site Assessment Tool for a system to assess a site – whether an entire county or one mountain – in terms of its possibilities and needs in relationship to developing community-beneficial forms of tourism.

III. Universal Features

Many items in the Place Dictionary – names, locations, feature relationships, and so forth – can have a date, source, and note attached to it. This provides for a tremendous range of complexity to be expressed at all levels of information about a given place.

1. Time

Every item in the Place Dictionary – each name, each location, each feature relationship, and so forth – can have one or more dates specified for it. Thus you can say that this name pertains to the feature for this time period, or that this location is relevant for this range of twelve years. Obviously this adds tremendously to the overhead in documentation, but it is the only way to create a clear record of the temporal provenance of each item of information. Otherwise, we have no idea how to contextualize this information in terms of history: that was its name, but when? that was its location, but then? Even in the contemporary realm, people will do things like say "is current", but this is meaningless information – all it means is that as of the date that this information was recorded, it applied, but that becomes part of the record of the past the moment we specify it. If a contemporary source gives a date and cites a historical Tibetan text or other historical source, then the date for the name is the date of the historical source. However, if the only date you have for a historical name is the date of publication of a contemporary source (such as a place name index published in 2006) then do NOT enter any date for that name.

We have a sophisticated date module. The first step in adding a new date is to specify the calendar used – Gregorian for international dates, or Tibetan for Tibetan calendar-based dates. Editors then will get a template suited for that calendar. Since there are also numerous specific calendrical systems in Tibet, if you choose a Tibetan date, then you also will be asked to further specify which Tibetan calendar is relevant to the date you are entering.

Secondly, you are asked whether you are specifying a point or a range. For example, if you are giving a founding date of a monastery, it is most likely just a point in time: 1345. But if you are specifying the validity of a name, it is likely going to be a range, such as from 1345 to 1678. One complexity here is that often you only have points but what you really want to specify is a range. An example is you have a source that gives a name for a village in 1654, but you don't know when that name first stated, or when/if it ended. In those cases, all you can do is specify the point, and attach an explanatory note if relevant.

The actual fields should be self-explanatory – year, month, date, and so forth; or with Tibetan, Rapjung, animal, element, etc. The most important thing to note is the way we are handling certainty. Historical dates are marked by uncertainty – we often just don't know exactly when a given thing happened. Rather we only now a range of times in which we are sure, or suspect, it happened. Thus every element of the date module – year, month, hour, and so forth – has a "certainty" attribute. Next to the item is a drop down list that says "certain" by default, but which can be changed to either "probable" (=reasonably sure but not completely) or "estimated" (=a rough but informed guess). If you choose "probable" or "estimated', then the program asks you for a range. Thus if you are specifying the year, and you know it was from 1450-1460 that the monastery was founded, you would choose "probable" and specify 1450-1460. If you have a date that is "circa 1750", or says "17th century", then you have to decide how to specify that and why. For example, you might decide to interpret 17th century as "1601-1699"; or you might decide, depending on the context, that it is 1640-1660. That depends on context and your scholarly judgment. Best to rely heavily on attached notes to clarify your thinking so people are not led astray and instead can follow your reasoning.

Generations: if a date is given in the format "three generations": use 20 years for the length of a generation and calculate the number of years; set the certainty to 3; add a note that says "Three generations; estimate based on 20 years per generation."

2. Sources

Many items in the Place Dictionary – names, locations, feature relationships, etc. – can have a source specified for it. In this way, the source – text and page number, a URL for a website or other online resource, or a person – can be specified for each individual item of information. Thus you can indicate that a particular name comes from page 4 of a particular book, and so forth. Obviously this adds tremendously to the overhead in documentation, but it is the only way to create a clear record of the source, significance, and integrity of the item of information in question.

Sources that are bibliographic resources are cataloged in the Media Management System (MMS), a bibliographical application within THL that keeps track of images, audio-video, web sites, articles, books and more. The Place Dictionary source only records the MMS ID for such sources, along with specific information about the passage – such as page number. The Place Dictionary then dynamically communicates with MMS to retrieve the title, author, publisher, and so forth for display to end users.

For an oral source, pull down the INFO SOURCE TYPE menu and select ORAL SOURCE. Then pull down the INFO SOURCE menu and select the name of the person. If the name is not in the pulldown, you must add it by clicking on the four horizontal lines at the top right corner of the menu bar, pressing DATA MANAGEMENT and then ORAL SOURCE, and the pressing NEW ORAL SOURCE at the top right. Enter the name and any descriptive text in the Description field. Be sure also to enter data in the CODE field, using the first three letters of the first and last names, capitalizing the first letter of the first name and the first letter of the last name. Example: the name of the person you are adding as an oral source is Ngawang Dorjee, so the code is NgaDor.

At present, editors are only offered a field to type in the MMS ID. Thus you have to manually go to mms.thlib.org to either find the ID for the source in question, or create a new MMS item if necessary. We are currently developing a more powerful editorial interface that will let you search, browse, and create new entries in MMS from within the Place Dictionary. Stay tuned…

Creating a record in the MMS for a text resource:

1. Go to mms.thlib.org/media_objects and log in
2. change URL to mms.thlib.org/media_objects
3. click on “new text” link at bottom of page
4. assign author from pulldown menu or add it if not already there
5. after the new text record is created, click on “edit catalog record”
6. click “add title” and then add the title

3. Notes

The following fields in the Place Dictionary can have a note attached to them

• names
• locations
• feature relationships
• feature types
• subjects characteristics
• geocodes

Every note can have a title and can have one or more authors attributed to it. The online interface provides a rich text editor to use to compose the note. For details on this, and using the online text editor, see below for the section "Descriptions/Essays Component."

An important issue is that in any given section of the Place Dictionary, you can attach a note to a very specific item of information, such as a single name, or you can attach a global note to the entire section – that is, to all names. Thus you can make a note about the erroneous character of a given name, or write a note that addresses a synthetic history of the naming of this place.

This ability to make granular annotations for all information in the Place Dictionary provides a powerful means to combine narrative descriptions and analysis with discrete, granular facts. This ideal combination allows for the nuancing of factual information, as well as the expression of details and and analysis best described in narrative form. For example, you may attach a note to a location to express exactly how you determined this location (I stood in the middle of the village square with my GPS; or I was told that it was 5 miles south of this river, and so I...."; or you may attach a note about the sectarian conversion of a monastery; you may write a note about history and diversity of a place's names over time.

IV. The Components of a Place Dictionary Entry

1. THE GENERAL INFORMATION COMPONENT

This component allows you to specify whether you want the item to be public or not. By default, the public option is selected, which means that the new entry will be seen by the public. Unselecting it means the entry will not be public until such time as you or someone else reselects it.

2. NAMES COMPONENT

This presents the various names for the feature over time and language. This is by far the most complex of all components. You have to create each type of name – various languages, various forms of transcription (sound) or transliteration (spelling), alternative names, and so forth.

For each name, it indicates the language and writing system in use, the time period for which the name applies, and the type (official, popular, etc.). It also indicates the relationship between names when a given toponym is the original name, and other names are derivative of it. An example, is the Tibetan language name for Lhasa is primary, while “Lhasa,” “Lasa,” and the Chinese character name for Lhasa are all alternative names derived from it.

The first step to adding a new name is to provide the general information. Click on “new name” in the NAMES section of the home page for editing a feature. That gives you a template for describing the name:

• Name: here is where you type the actual name in.
• Type: “official” or “popular,” or leave blank. “Official” is of course the government name for administrative units or settlements, whether the government endorsing and promoting the name in question is local, regional or national. “Popular” means it is a name only used in popular speech, or in non-official written documents. Thus “New York” is official, but “the City” and “Big Apple” are popular. Each place name can also be marked as “primary” or not. However, for Tibetan monasteries, “official” names can also signify the formal and usually longer name that the monks use for the monastery rather than more informal short hand names people often use in conversation.
• Language: specify what language the term is, regardless of how it is being represented linguistically, from the drop down list of languages. Language has nothing per se to do with the writing system being used. In other words, if a Tibetan language place name is rendered in another writing system through either transliteration or phonetic rendering, the “language” of the place name remains Tibetan. Thus “lhasa,” “lha sa,” “lasa,” or the Chinese characters for it are all Tibetan-language names. Only if the place name is translated into the other language, or partially translated, do we then change the language specification to the other language – thus Shannan is a Chinese-language place name, since it translates the Tibetan “lho kha” into the Chinese characters for “Southern Mountains” rather than representing the sound “lhokha”; in contrast, “lhokha” in latin script is still a Tibetan-language place name. The former is a semantic translation into Chinese, while the latter is a phonemic rendering into Latin script. If a place name is a hybrid – part phonemic rendering and part translation of another toponym – then the language of the toponym should be switched to the new language. In other words, New York is English – but Nueva York would be language=Spanish. If the only part translated in a place name is the generic name, such as “hotel,” “school,” “monastery,” and so forth, then the language of the name does not change.
• Writing system: specify the writing system representing the term from the drop down list (Chinese traditional characters, Tibetan, Latin, etc.). Tibetan script includes the use of it in Bhutan, though there are certain conventions about syllable formation that are different. Chinese then comes in “simplified” and “traditional” characters, which we treat as different writing systems. Latin script is the standard script used for English, French, Spanish, German, and so forth, and includes its use with special diacritic marks.
• Etymology: For etymologies, use this form. Always start with “Literally, meaning A-meaning B-meaning C,” where meaning A, etc. are the literal meanings of the syllables or words that are meaning-units in the toponym. The meanings should be given in the order in which they occur in the name itself. Following this, the editor should write whatever other content she wants to express about the name’s etymology. An example (lha sa):
  Literally, “god-place.” The name derives from the city’s….
• Etymology Language: select from the dropdown menu the language the etymology is written in.
• Primary for Popular Romanization: end users can select the linguistic form in which they want to view place names, such as “popular romanization” or “Tibetan script.” Popular romanization is an easy-to-pronounce version of names in roman script, and is intended to be the most common choice for people around the world. For Tibetan features, this is usually THL Phonetics; for Chinese, this is pinyin. However, in some cases, we want a well-known international name – for example, “Mount Everest” – to display as the primary name in romanization rather than the phonetic transcription of the Tibetan name (Jomo Langma); another instance is a full translation of a name, or when general types are in the place name, such that we want the partial translation “Puhreng Hotel” appearing instead of the full transcription “Puhreng Drönkhang.” In these cases, change this field to “yes”.

A question that often comes up is whether a generic place type is included in a place name – such as “Monastery” (dgon pa), “Hotel” (mgron khang), “School” (slob grwa), “County” (rdzong), and so forth. Use the name as it appears in your source, whether it is a literary source or an oral source. For oral sources, that means observed conversation, or a transcript of speech, but not posing artificial questions to people about how the name is said – which will tend to produce artificial and hence unreliable answers. If a place name has a generic place type in it – such as “monastery,” “hotel,” “autonomous county,” “school,” etc. – then you should also make an English translation of the term that combines the phonetic rendering of the other part of the name with an English translation of the place type. Thus “Sera Monastery,” “Purang Hotel,” and so forth. Such names have two relationships to the original name: translation and phonetic transcription. These English translations should then be marked as “primary for popular romanization.”

Features that Have No Known Name if you do have not identified a name for a feature, then give it a name that is the same as its feature type. Examples: for a salt lake that has no known name, give it this name: Salt Lake. For a building for which you have not identified a name, give it this name: Building.

LAKES:

1. In general, "Lake" goes at the beginning, like Lake Nam for gnam mtsho
2. For names that begin with mtsho, such as mtsho sdod sbal, "Lake" goes at the beginning: Lake Döbel
3. When the generic component of the name includes གཡུ་མཚོ (g.yu mtsho) at the end, then "Turquoise Lake" goes at the end. Examples: ma pham g.yu mtsho=Mapam Turquoise Lake; g.yar 'brog g.yu mtsho=Yamdrok Turquoise Lake
4. When the generic component of the name includes བུལ་མཚོ (bul mtsho) at the end, then "Alkaline Lake" goes at the end. Example: de'u rangs bul mtsho=Deurang Alkaline Lake
5. When the generic component of the name includes tshwa kha at the end, then "Salt Lake" goes at the end. Examples: mtsho rdo tshwa kha=Tsodo Salt Lake; mtshor tshwa kha=Tsor Salt Lake; mdzo dgu yul tshwa kha= Dzoguyül Salt Lake
6. For a feature that is a Mongolian word but for which we don't have the Mongolian spelling, and we have names that are transcriptions of the Mongolian into Tibetan script with mtsho at the end (such as ཨུ་ལན་ཨུ་ལ་མཚོ།) and transcription into simplified Chinese characters with hu (lake) at the end (such as 乌兰乌拉湖), then make the Tibetan script name the root name, and make the Chinese name a child of the Tibetan name. Note: for both names, language=Mongolian
7. Specific lake names: རིང་མཚོ་གོང་མ།=Upper Lake Ring; རིང་མཚོ་འོག་མ། =Lower Lake Ring; rdo skas mtsho ring= Doké Long Lake

RIVERS and STREAMS:

1. In general, put "River" or "Stream" at the end. Example: ཕར་ལུང་གཙང་པོ།= Parlung River
2. For names that begin with chu, such as chu sngon, "River" (or "Stream") goes at the end: Ngön River
3. གཙང་པོ (gtsang po) is translated as "River." ཆུ་ (chu) is either translated as "River" or as "Stream," depending on what the feature type is. If the feature type is any type of stream, then translate it as "Stream"; if the feature type is any type of river, then translate it as "River."
4. lung, lung pa, and lung chu: translate these as "River" or "Stream," depending on the feature type.
5. When "upper" or "lower" is included in the name, translate that. Example: sngo chu stod="Upper Ngo River"
6. When "east," "west," or other directions are included in the name, translate it. Example: lho brag nub chu=West Lhodrak River

You can specify the time span for which the place name is operative. If you have precise start dates and end dates, enter them. If you only have a single date, even this year, then put that. If you have imprecise information, then enter it in a note field.

Then click on “create” to create the new name with the specified information. You will then be shown the information you just entered, along with two further sections – “Feature Name Relations” and “Citations.” These allow you to specify how this name is related to other names for the same place and to specify the source of your information on this place name respectively.

Source: you can add a source for a name. If the source gives only the name in Tibetan font, the Wylie and THL Phonetic versions of the name will have the same source (even though the source does not include the Wylie and/or phonetic.) If a Chinese name is given, the pinyin version of the name will have the same source as the name in Chinese characters even if the source does not actually give the pinyin.

For feature name relations, click on “select” and it will show you a list of other names for this feature with select buttons for each one. Choose the name to which you want to relate your present name, and it will take you to a form in which you can specify how these two names are related. The form asks you to answer the following four questions, and then allows you to specify a timespan as well. These four questions all relate to whether the name in question is derived from the other name in a standard linguistic transformation – a translation, a phonemic transcription, and so forth:

• Is name #1 a translation of name #2?: this is answered yes or no. If the name is a partial translation, like Nueva York, then enter yes.
• Is name #1 a phonetic transcription of name #2? this is answered yes or no. There is also a drop down box to specify which phonetic system is involved if the answer is yes. You can also just choose one of the systems and it will automatically change the answer to "yes." If the name is a partial translation and partial transcription, like Nueva York or Drepung Monastery, then select the type of phonetic transcription (in this case, THL Simplified Phonetics).
• Is name #1 an orthographic transliteration of name #2? this is answered yes or no. There is also a drop down box to specify which orthographic system is involved if the answer is yes. You can also just choose one of the systems and it will automatically change the answer to "yes."
• Is name #1 an alt. spelling of name #2? this is answered yes or no. There is also a drop down box to specify what type of alternative spelling is involved if the answer is yes. You can also just choose one of the systems and it will automatically change the answer to "yes." Please note “acronym” is considered an alternative spelling, such as TAR for the Tibet Autonomous Region.

You then click create to save your answers, and in the next screen can add a citation as the source for this information if relevant. In most cases there is no citation, since the relationship between the two names is clear. But in the case of relationships that are based on oral testimony, or a piece of scholarship, you would specify that source in the citation module.

Please note that for Tibetan places we require you give the name at least in Tibetan script (using a Unicode font), THL Extended Wylie Transliteration, and THL Simplified Phonetic Transcription. We offer online converters to make this easy:

• http://thlib.org/reference/transliteration/wyconverter.php
• http://thlib.org/reference/transliteration/phconverter.php

Notes on Transliteration and Transcription systems

The types of orthographical transliteration systems are:

• ACIP Tibetan: used by the Asian Classics Input Project for their monastic input project of Tibetan Buddhist scriptures in India.
• Chinese Wylie: This is a system used only in China for transliterating Tibetan in roman script. It is very close to Wylie, but, for example, uses a v instead of an ' for the a chung.
• General Wylie: this is only used in a few cases where the Wylie scheme produces a result other than THL Extended Wylie. Otherwise all Wylie transcription should be marked as THL Extended Wylie.
• Indological Standard Transliteration: for representing the spelling of Nepali, Hindi and Sanskrit words from the Devangari script in Latin script through the addition of special diacritic marks.
• Library of Congress Cyrillic Mongolian to Latin Transliteration: used by the US Library of Congress for transliterating Tibetan.
• Library of Congress Mongolian Vertical Script Transliteration: used by the US Library of Congress for transliterating Tibetan.
• Library of Congress Tibetan: used by the US Library of Congress for transliterating Tibetan.
• THL Extended Wylie: for representing the spelling of Tibetan words in Latin script.
• THL Mongolian-Cyrillic Transliteration:
• THL Mongolian Script Simplified Transliteration:
• Traditional to Simplified Chinese: This is the system used to create simplified Chinese characters renderings of Chinese terms in traditional Chinese characters.
• Unidentified System of Chinese Transliteration: used when you are unsure of what system has been used to create an attested rendering for this language, or suspect that it may be a popular rendering of that specific term with no system behind it.
• Unidentified System of Dzongkha Transliteration: used when you are unsure of what system has been used to create an attested rendering for this language, or suspect that it may be a popular rendering of that specific term with no system behind it.
• Unidentified System of Mongolian Transliteration: used when you are unsure of what system has been used to create an attested rendering for this language, or suspect that it may be a popular rendering of that specific term with no system behind it.
• Unidentified System of Nepali Transliteration: used when you are unsure of what system has been used to create an attested rendering for this language, or suspect that it may be a popular rendering of that specific term with no system behind it.
• Unidentified System of Tibetan Transliteration: used when you are unsure of what system has been used to create an attested rendering for this language, or suspect that it may be a popular rendering of that specific term with no system behind it.
• Vladimirtsov-Mostaert Mongolian Vertical Script Transliteration: used when you are unsure of what system has been used to create an attested rendering for this language, or suspect that it may be a popular rendering of that specific term with no system behind it.

The types of phonetic transcription systems are:

• Amdo Transcription:
• Chinese to Tibetan Transcription: for representing the sound of Chinese words in Tibetans script.
• Dzongkha Transcription:
• English to Chinese Characters Transcription: for representing the sound of English or other European languages words in simplified or traditional Chinese characters.
• English to Tibetan Transcription: for representing the sound of English words in Tibetans script.
• Ethnic Pinyin-Tibetan Transcription: a romanization scheme only used in China to apply the principles of the Pinyin system to directly transcribe Tibetan words in roman script.
• Hopkins System of Tibetan Transcription:
• Indological Standard Transcription: for representing the sound of Nepali, Hindi and Sanskrit words in Latin script without special diacritic marks.
• International Phonetic Alphabet Transcription:
• Kapstein System of Tibetan Transcription:
• THL Simplified Tibetan Phonetics Transcription: for representing the sound of the Tibetan words in Latin Script in very simplified if imprecise fashion.
• Pinyin Transcription: the contemporary standard for representing the sound of Chinese characters in Latin script. See these guidelines for capitalization and combining syllables. For a converter from Chinese characters (either traditional or simplified) to pinyin, see https://www.branah.com/chinese-to-pinyin or http://www.pin1yin1.com/ (this also provides a rough English translation). NOTE: the tone numbers or marks need to be removed and the syllables need to be combined.
• Sanskrit-to-Tibetan Transcription: for representing the sound of Sanskrit words in Tibetan script. This is used for the transcription of Sanskrit titles into Tibetan at the beginning of Tibetan translations of Sanskrit texts, as well as other places (such as the Tibetan philosophical/ritual term ཙིཏྟ་).
• Tibetan-to-Chinese Transcription: for representing the sound of Tibetan words in traditional or simplified Chinese characters. If you have both traditional and simplified, the simplified is made a derivative of the traditional.
• Unidentified System of Chinese Transcription: used when you are unsure of what system has been used to create an attested rendering of this language, or suspect that it may be a popular rendering of that specific term with no system behind it.
• Unidentified System of Dzongkha Transcription: used when you are unsure of what system has been used to create an attested rendering of this language, or suspect that it may be a popular rendering of that specific term with no system behind it.
• Unidentified System of Mongolian Transcription: used when you are unsure of what system has been used to create an attested rendering of this language, or suspect that it may be a popular rendering of that specific term with no system behind it.
• Unidentified System of Nepali Transcription: used when you are unsure of what system has been used to create an attested rendering of this language, or suspect that it may be a popular rendering of that specific term with no system behind it.
• Unidentified System of Tibetan Transcription: used when you are unsure of what system has been used to create an attested rendering of this language, or suspect that it may be a popular rendering of that specific term with no system behind it.
• Wade-Giles: the earlier standard for romanization of Chinese.

In China, a further difficult issue in contemporary feature names is that of “ethnic pinyin.” There is poorly formalized practice of rendering Tibetan names in roman script according to the Chinese ear with what is called “ethnic pinyin.” Modeled upon the standard roman script transcription system of “pinyin” for Chinese language, it renders the sound of Tibetan place names in roman script in accordance with Chinese phonological proclivities. One challenge is that in any given case it may be unclear just which dialect of Tibetan is being transcribed, and from what dialect of Chinese the transcriber is working. Thus the “ethnic pinyin” rendering may seem quite bizarre, but when you understand the dialect of the Tibetan being spoken, the dialect of Chinese the transcriber knew best, and the conventions of pinyin itself (that is, how specific roman script letters are pronounced in pinyin), it may make a lot more sense. Unfortunately, it appears to be poorly formalized and inconsistent. We urgently need to document what we can of this practice, including ample examples. Since Chinese character renderings of Tibetan toponyms may themselves be simply attempts to render the sound of the Tibetan in Chinese, it also may be that the “pinyin” of the Chinese and the “ethnic pinyin” of the Tibetan toponym are identical.

Changing the Prioritization of Names

At the top of the Names component, there is a link "Changing the Prioritization of Names". Click on it and it gives you a list of all feature names. You can click on a given name, hold the clicker, drag the name to a different location and let go. The sequence of names will determine priority. Thus if you have two names at a given level, you can use this to indicate which has priority.

3. FEATURE TYPES COMPONENT

This allows for specifying the type of feature (village, mountain, etc.) as outlined in THL’s Feature Thesaurus. If you are unsure what type of feature a given feature is, then you should contact THL to discuss it. In some cases, if its simply unknown, then you can use the “Unspecified” option from the Feature Thesaurus. The feature thesaurus needs an “Unspecified” option so that if we don’t know the feature type we can still place times on the feature.

When you create a new type, you can either type in the feature type in the "type" box, and as you type the program will suggest what term you need using "auto-complete" as drawn from the THL feature thesaurus. You can also click on "select from the tree" and view the complete feature thesaurus and select your type from there.

The time period for that type is specified, in case the feature changes type over time. Multiple feature types may thus be indicated.

When finished, click "create" or "update" to register your changes.

Once submitted, you can also access a citation module to specify source.

Changing the Prioritization of Feature Types

At the top of the Names component, there is a link "Changing the Prioritization of Feature Types". Click on it and it gives you a list of all feature types. You can click on a given type, hold the clicker, drag the type to a different location and let go. The sequence of types will determine priority. Thus if you more than one feature type, you can use this to indicate which has priority.

4. Knowledge Maps of Subjects Characteristics Component

This is the essential component allowing one to expand the structured description of a given place in ways specific to that type of place. This means instead of just giving that information within a narrative essay, you specify it in fields that are clearly labeled. Thus for a monastery, you may want to formally express – rather than in a note – its religious sect, predominant ritual system, tutelary deity, and so forth; for a country, you may want to express its population, economic output, etc.

To handle this, the Place Dictionary lets editors open up THL's Knowledge Map of Subjects and select any topic for affiliation with the place in question. The Knowledge Map of Subjects is an application THL has made to create complex, multilingual, annotated hierarchies ("ontologies") of topics that represent knowledge on various subjects – geographical features, literary genres, rituals, language families, and the like. The geographical feature thesaurus discussed directly above is just one example. The "Characteristics" section thus works much like the "Feature Types" section, but instead of being limited to that one Subject category, it allows editors to draw upon any topic within any Subject.

The editorial interface allows you to either type in the name of the topic in the "characteristic" box, and as you type the program will suggest what term you need using "auto-complete" as drawn from the THL Knowledge Map of Subjects. You can also click on "select from the tree" and view the complete array of Subjects and select your topic from there.

In addition, for each specified category, you can also add a "numerical label" or "text label" (see immediately below). The time period for that characteristic is specified, since characteristics can change over time. You can add as many characteristics as you like. When finished, click "create" or "update" to register your changes. Once submitted, you can also access a citation module to specify source.

Every specified characteristic can then also have a date, source, or note attached to that category. The combination of "ontologies" with dates, sources, and rich titled notes allows for an extraordinary range of data to be expressed in structured and narrative forms. Examples are you can use the "Events" knowledge map to specify something like "founded" for a monastery, and then use the "text label" to indicate the founder ("By Tashi Tsering"), as well as the date module to specify when the founding was. Another example is you can use the "Population" knowledge map to indicate how many monks lived in a monastery, with the "numerical label" being "500" (or however many monks there are). You then apply a "source" to detail the source of that information, and "date" to specify when that population number was relevant.

One particularly important Knowledge Map category is "Exists" (Geographical Features > Feature Condition > Constructed Entities > Exists). By selecting this knowledge map topic and adding a date and a source, you can indicate for example that a monastery existed at a particular time since it appears on a map from that time.

5. LOCATIONS COMPONENT

This details any information pertaining to the location of the feature, including latitude and longitude, altitude, address, and any narrative comments about the location.

For entering latitude and longitude, it is necessary to enter decimal degrees only (36.134). Never enter degrees, minutes, and seconds (36° 14' 35"). If your data is in minutes and seconds, you have to convert them to decimal degrees. This is easy to do: use http://www.fcc.gov/mb/audio/bickel/DDDMMSS-decimal.html. If you collect data from Google Earth, you can configure the settings to display latitude and longitude in decimal degrees: pull down the Tools menu and click Options. In the 3D View tab, in the Show Lat/Long section, click “Decimal Degrees.”

NOTE: do NOT enter N, S, E, or W as part of the latitude or longitude. Enter only a number. For latitudes north of the equator, enter a plus sign (+) before the number; for latitudes south of the equator, enter a minus sign (-) before the number. For a longitude east of the prime meridian, enter a plus sign (+) before the number; for a longitude west of the prime meridian, enter a minus sign (-) before the number.

For altitude, you MUST give a measurement in meters. Never use feet. If your data is in feet, use this online converter: http://www.calculateme.com/Length/Feet/ToMeters.htm. If you collect data from Google Earth, you can configure the settings to display altitude in meters: pull down the Tools menu and click Options. In the Show Elevation section, click “Meters, Kilometers.”

Changing the Prioritization of Locations

At the top of the Names component, there is a link "Changing the Prioritization of Locations". Click on it and it gives you a list of all locations for this feature. You can click on a given location, hold the clicker, drag the type to a different location and let go. The sequence of locations will determine priority. Thus if you more than one location, you can use this to indicate which has priority.

Specifying Location of Features Via Relationship to Other Features

For regions with vague boundaries - such as cultural regions or historical polities - it is impossible to precisely pinpoint the location in terms of latitude and longitude. In these cases, it is common to roughly correlate such regions to contemporary administrative regions - which tend to have precise and known boundaries - so that they can be mapped at least in a rough way. This can be done in the Place Dictionary through using the RELATIONSHIP function. The Mapping system then will use those relationships to map the region, if no geometry of its own is expressed. Please see below under the RELATIONSHIP section of the editorial manual for details.

6. GEOCODES COMPONENT

This details any identifiers or codes used for this feature in any source, whether national ids like a postal code, ids assigned to places in scholarly publications, or otherwise.

The most important is THL’s own feature id, which begins with the letter “f” and then consists of a numerical string.

Creating a new geocode involves selecting the type of ID from a drop down list, and then typing in the actual value for the ID. You can also add a note, specify a timespan, or add citations of sources.

If you cannot find the feature ID type you need, just write us and we will add it. This includes site IDs used by a scholar in his corpus.

7. FEATURE RELATIONS COMPONENT

Features can be related to each other in diverse ways. Such relationships can be spatial in character, or can be cultural relationships. This component allows for specification of relationships the feature has to other features, such as "is administered by" (like a county within a prefecture), is located in, is related to, etc.

Clicking on "new feature relation" displays a simple version of the Place Dictionary home page with the right-hand browsing tree and topmost search box. You can use this to find the feature to which you want to create a relation, and then click on “select” next to the desired feature. Once you select a new feature, the program asks you to specify relationship type and perspective, as well as dates:

• Relation Type: It provides a drop-down list of types of relationships between features, from which one must be chosen.
• Perspective: It provides a drop-down list of broad perspectives in which to understand the specified role, from which one must be chosen.

Feature Relation Types

The current set of feature relation types is as follows. In order to facilitate understanding, the list of types is organized conceptually rather than alphabetically. The first set of relationship types is purely based upon issues of location, and the individual types are here presented from closest to furthest:

• has the same location as (same.location): two features that are located in the same place.
• has entirely located within it (has.entirely.located.within.it)/ is entirely located in (is.contained.by): this is a purely geographical relationship where one feature is spatially entirely contained by another feature. (See the description of the “is part of” relation below to determine which relation to use.) Examples are a monastery, mountain, village, and pretty much anything that is not an administrative unit being located within a township or a county. This includes a monastery, mountain, school, park, hotel, and so forth that is located within a village. It also includes a historical polity that is entirely located within a contemporary administrative unit such as a county. Use this for the relationship between a village and the township it is located in. The relationship between a village and the township it is located in occupies a gray area in this scheme. It could be argued that the relationship between a village and the township it is in is the same as the relationship between a township and the county it is in (is.administered.by). However, while the townships in a county occupy the full geographic extent of the county, the villages in a township do not similarly exhaustively comprise the township. For this reason (and to prevent the display of the Browse tree from becoming overloaded and unwieldy by including villages) we have decided that the relationship between a village and township is is.contained.by.
• is the partial location of (is.the.partial.location.of)/ is partially located in (part.loc): for a feature that is located in two or more other features, and not wholly contained by either, nor wholly containing either. If one of the features is entirely contained by the other feature, use the "is entirely contained by" relationship type. Thus, for example, if you have a historical polity the geographical area of which is within parts of four counties, but does not cover the entire county for any of the four, the relationship is "partial location of". Also, if you know the geographical area of a historical polity was located in a contemporary administrative unit but you are not sure that it was entirely located in this contemporary administrative unit, the relationship is partially located in.
• intersects with (intersects.with): these are two features that overlap with each other. Examples are a natural area, river, or mountain which spans multiple administrative units, all of which it “intersects with.” This can also be used when you are unclear on the nature of the relationship between two regions, but know that the two things partially overlap in geographical territory. That is in fact similar to "is the partial location of," and we generally recommend its use in those cases. If you have a case where "intersects with" seems preferable - perhaps because it is a symmetric relationship that doesn't require you to specify which is the "location of," and which is the "location in" - please let us know so we can include that.
• is adjacent to (is.adjacent.to): this means two features which share a physical border, in contrast to “is near.”
• is near (is.near): this means two features are near to each other. We utilize this, for example, for our tourism documentation to point to sites which are nearby the tourism site being discussed and may offer day trips opportunities for the tourist.
• is centered in (is.centered.in)/ has centered in it (has.centered.in): used when, for example, we know that a historical kingdom was centrally located in a territory now correlated to a modern administrative unit, but its full extent may have spilled out into other areas that are now part of other modern administrative units. An example is Degé Kingdom and Degé County, with the county covering the core geographic area of the kingdom. This relationship can also obtain between a cultural region and a contemporary administrative unit. The perspective for these examples is geographic relationship (geo.rel). Note that this relationship should be specified in addition to separately using "entirely located within" or "partially located within" relationship types to indicate the precise geographical relationship between the historical polity and the modern administrative unit. Thus, for example, you might say that the historical polity Degé Kingdom is "centered in" the modern administrative unit Degé County, but it is still necessary to also say that Degé Kingdom is "partially located in" Degé County.

The second set of relationship types is based upon largely cultural considerations. The order of presentation is based upon the conceptual (rather than locative) closeness of relationship indicated by the type:

• has as an instantiation (has.as.an.instantiation)/ is an instantiation of (is.an.instantiation.of): refers to an instantiation of a metaphysical feature. Thus, for example, Padmasambhava’s Copper Colored mountain is a pure land located outside of our ordinary geography, but there are many temples across the plateau understood to exemplify that pure land. Thus they are “instantiations” of it. Likewise, many of the great Buddhist cult mountains are understood to be “instantiations” of the great mandala of the Buddhist deity Cakrasaṃvara. On the more mundane level, a McDonald's restaurant would be an instantiation of McDonald's as a corporate entity.
• has as a part (has.as.a.part)/ is part of (is.part.of): this signifies a relationship where one place is part of another place, but not in a hierarchical or administrative manner. Usually the “is part of” relationship is between places that have the same feature type or closely related feature types, although there may be exceptions to this. Examples include cultural regions, where you have a large cultural region and many smaller cultural regions that are "part of" it (for example, the Degé cultural region is part of the Kham cultural region); a mountain being "part of" a mountain range; and an individual field that is part of a farm or estate (whether the field is contiguous with the farm/estate or not). Do not use this relation to specify a village is part of a township, for example; instead, use "has entirely located within it/is entirely located in."
• administers (administers)/ is administered by (is.administered.by): this is particularly important because such relationships can be used to create a hierarchical tree for browsing within a given “perspective.” The administers (administers)/ is administered by (is.administered.by) relationship involves an actual administrative transactional relationship between the two features in question. The classic example is contemporary administrative units such as a township and the county it is in, where the relevant issue is not that a given unit is contained within another unit geographically, but rather that one unit is subordinate to another in terms of authority. Another example is a polity that is subordinate to a larger polity. Do NOT use this for the relationship between a village and the township it is located in; use "is entirely located in" (is.contained.by). Do NOT use this for a feature like a mountain that is located in a township, since the township does not actively administer the mountain in the same way that a county administers a township within it.
• has as an administrative seat (has.as.an.administrative.seat)/ is the administrative seat of (is.administrative.seat.of): this is used to articulate the relationship between capitals or seats and their respective nations, historical polities, or political administrative units.
• administrative headquarters of (is.administrative.headquarters.of)/ has as an administrative headquarters (has.as.an.administrative.headquarters): this is used to articulate the relationship between a headquarters and its corresponding corporation, school district, or other entity.
• is mother of (is.mother.of)/ is child of (is.child.of): examples are monasteries that are “children” or “branches” of “mother” monasteries. Further research is necessary on the diversity of such relationships.
• is succeeded by (is.succeeded.by)/ succeeds (succeeds): this applies when a place succeeds another place, such that there is both a fundamental difference and a fundamental continuity, and there is the perception that the one feature has supplanted the previous feature. An example is a kingdom which is reduced to the status of a county in a new political formation, such that the county "succeeds" the kingdom. This includes documenting that a modern "county" in some sense "succeeds" a previous kingdom, for example, which may have had the same name and even a similar geographical footprint. In such cases, use the "political Relationships (pol.rel)" perspective. In most cases both the features have the same name, and the feature that succeeds the other feature usually has a geographical area that is the core area of preceding feature. Additionally, one of the features usually immediately succeeds the other feature temporally. In terms of the geographical area covered by the two features, there are three possibilities: 1) the two features share roughly the same geographical area; 2) the preceding feature is somewhat larger than the succeeding feature; 3) the preceding feature is much larger than the succeeding feature and the succeeding feature covers the core area of the preceding feature (example: Degé Kingdom and Degé County). Please note, however, that the succeeds/is succeeded by relationship has NOTHING to do with location per se, and thus it is still necessary to use other relationships to indicate the nature of the geographical relationship between the historical polity and contemporary administrative unit.
• is owner of (owns)/is owned by (is.owned.by): this signifies a relationship of ownership or property, such as a monastery owning a city residence or an estate owning a field.
• is affiliated with (is.affiliated.with): this signifies a relationship that is not hierarchical in character. For two polities, this is used for “allies”; or two monasteries involved in a non-hierarchical formal relationship. It can also be used to describe the relationship of sacred sites that are interlinked but not in a hierarchical manner. Contrasted to "is related to," "is affiliated with" conveys a stronger relationship or association between the two places.
• is in conflict with (is.in.conflict.with): this signifies a relationship of conflict, such as two polities at war or in more low level conflict.
• is related to (is.related.to): this is a generic expression of relationship between two places when none of the other more specific relationship types apply.

Encyclopedia module extensions can also express more complex types of relationships.

Perspective Relationships

Perspectives express an overall framework of relationships between geographical features in which the specified role makes sense. Thus when you specify that two features are related to each other, you have to specify the “perspective” which contextualizes that relationship.

At present there are ten perspectives overall:

• Administrative Relationships (admin.rel): administrative relationships between two features that are not polity administrative units. This is very close to "Organizational Relationships", and we need to document specific cases of each.
• Cultural Regions (cult.reg): used to document the relationship between regions defined by the resident’s sense of their belonging to non-official cultural groups, such as the “Minyak” cultural region “is part of” the Kham cultural region; sometimes called “folk regions.” Always use "is part of" for the relationship type.
• Cultural Relationships (cult.rel): this is a general relationship type for any framework that is rooted in cultural factors, and not included in one of the other perspectives. DO NOT use for "Cultural Regions." For a monastery, lake, mountain, and so forth located within a cultural region, relationship type=has entirely located within it/is entirely located in OR is the partial location of/is partially located in and perspective=cultural relationships .
• Environmental Relationships (envir.rel): used to document the relationship between various elements of the natural landscape, such that this mountain is a “part of” this mountain range, or a watershed area containing other natural features, and so forth. This covers geological relationships as well as relationships that relate to flora and fauna. The relationship of a river or stream and a watershed is perspective=Natural Features and relationship type=is part of.
• Geographical Relationship (geo.rel): this is a general relationship type for any framework that is rooted in geographical location, without being part of one of the other perspectives. As such, it is closely related to the Site Relationships perspective; the difference is that Geographical Relationship deals with relationships on a larger scale than Site Relationships, which focuses more on smaller areas; and the former tends to include more natural features, while the latter tends to be about human sites like a large monastery and the individual buildings, courtyards, and so forth that comprise it, or an archaeological site and the individual features that comprise it.
• Historical Polity Administrative Units (hist.pol.admin.unit): used to document the relationship between a single historical polity's administrative units. A classic example is if you specify “administers” as the relationship between the Eighteen Gyelrong States and Drakteng, which is one of its component 18 polities; the perspective is “Historical Polity Administrative Units" perspective. For a feature such as a mountain, village, monastery, or lake that is located within a given historical administrative unit, the relationship type=has entirely located within it/is entirely located in OR is the partial location of/is partially located in, and perspective=Historical Polity Administrative Units.
• National Administrative Units (pol.admin.hier): used to document the relationship between a single modern nation state's administrative units. A classic example is if you specify “is administered by” as the relationship between a “township” and the “county” that is its location and administrator, then that would be according to the “National Administrative Unit" perspective. For a feature such as a mountain, village, monastery, or lake that is located within a given modern administrative unit, the relationship type=has entirely located within it/is entirely located in OR is the partial location of/is partially located, and perspective=National Administrative Units.
• Organizational Relationships (org.rel): used to document the relationship between organizations and/or organizational units. USE FOR: two monasteries that are engaged in a "mother" and "child" relationship.
• Political Relationships (pol.rel): for specifying relationships between two separate polities – alliances, conflicts, and so forth. This includes documenting that a modern "county" in some sense "succeeds" a previous kingdom, for example, which may have had the same name and even a similar geographical footprint.
• Religious Relationships (rel.rel): this covers monasteries, pilgrimage routes, sacred sites, sacred mountains, and all other networks of religious relationships between features.
• Site Relationships (site.rel): to address the issue of having a building that is part of a larger building complex or overall site.

Practical Guidelines to Determining Place Relationships

There are many legitimate questions one may ask about how to best express the relationship between places. We are trying to provide guidance by making a list of specific cases here. If you don't find your case covered, please write us and spell out your issue. We will write you back and add the resolution to the Place Dictionary Editor's Manual.

For short-hand use, we use the term "location term" to mean that you use one of the relationship types that specify location relationships between two features. For details on each, please see the documentation above

• has the same location as (same.location)
• has entirely located within it (has.entirely.located.within.it)/ is entirely located in (is.contained.by)
• is the partial location of (is.the.partial.location.of)/ is partially located in (part.loc)
• intersects with (intersects.with)
• is adjacent to (is.adjacent.to)
• is near (is.near)
• is centered in (is.centered.in)/ has centered in it (has.centered.in)

Relations between modern administrative units: specify the feature having a relationship to the next largest administrative unit that subsumes it; relationship type = "is administered by" and perspective = "National Administrative Units" (pol.admin.hier). For example, in the PRC the relationship is between a township and the county it is in, and for Bhutan, this would be between the Chiwog and the Gewog it is in. In the PRC for a village or town related to the township it is in, the relationship type = "is entirely located in" (is.contained.by) and the perspective = "National Administrative Units" (pol.admin.hier).

Relations between administrative units belonging to a pre-nation historical polity like a kingdom: specify it having a relationship to the next largest administrative unit that subsumes it; relationship type = "is administered by" and perspective = "Historical Polity Administrative Units" (hist.pol.admin.unit).

A specific human place (village, hotel, school, monastery, hotel) located within a contemporary or historical administrative unit: relationship type = "has entirely located within it/is entirely located in" OR "is the partial location of/is partially located in," and perspective=Historical Polity Administrative Units (hist.pol.admin.unit) or National Administrative Units (pol.admin.hier).

A specific human place within a cultural region: relationship type=location term (usually "is entirely located in") and perspective=cultural relationships (cult.rel).

A specific human place that owns or is owned by another human place: relationship type=owned/is owned by and perspective=organizational relationships (org.rel). Examples are a residence in the Barkor in Lhasa owned by Sera Monastery, or a field owned by an estate.

A natural feature (mountain, lake, etc.) within a contemporary administrative unit: If that administrative unit is the smallest type of unit, this can be very helpful since often such units don't have available boundaries in terms of latitude and longitude (hereafter referred to as lat/long). Thus you can't use the lat/long to see what mountains are in that unit. However, there is also a problem, in that the boundaries of these smallest administrative units typically change much more frequently than larger administrative units. Thus such an association in 1994 may not hold in 2004. Thus saying this mountain is in township XXX in 1994 may not be true in 2004. Therefore it is very important to make sure you register the date for which you know this association is valid based upon your sources.

• Guideline: relationship type= location term; perspective = National Administrative Units.

A natural feature (mountain, lake, etc.) located within a historical polity's administrative unit: you have a mountain, lake, or other natural feature, and you have an association of it with an administrative region of a historical polity. Even if the natural feature is clearly identifiable in contemporary times, and its location is clear, this is still a very helpful association because it can help people understand the range of the historical polity, or its administrative region, the boundaries of which are usually very unclear.

• Guideline: relationship type= location term; perspective = Historical Polity Administrative Units.

Feature related to both a cultural region and a contemporary administrative unit: you are confused about what to do when you have a Cultural Region and a contemporary administrative unit of the same name, and you want to express the relationship of a discrete feature like a temple to the broader feature in which it is located. Because most resources specify location in terms of administrative units, we give that primacy. We do care about the association of geographical features to Cultural Regions, but in many cases we will try to do that by associating contemporary administrative units to them, such that all the features contained within the latter will be automatically associated with them.

• Guideline: These are two entirely different things – you need to locate it within the latter regardless of the former.

A cultural region X & related modern administrative unit Y: There is no intrinsic relationship between a cultural region and a modern administrative unit other than location. Thus we use the same relationship type and perspective for specifying a temple or village is in a given modern administrative unit. That said, admittedly a modern administrative unit may share the same name with a cultural region, and the cultural affiliation naturally will shape how the administrative unit is represented and perceived. However, we have chosen to focus primarily on the locative relationship.

• Guideline: X is "location term" Y (type) in "geographical relation" perspective.

A historical polity Z & related modern administrative unit Y: when there is significant overlap in name, geographical extent, local perception, and/or other relevant factors between the two -

• Guideline: Z "is succeeded by" (is.succeeded.by) Y with the perspective Political Relationships (pol.rel) perspective.
• Separately from this, in order to simply use the latter to specific a geographical extent for the former, use Z "is located in" Y (type) in "geographical relation" perspective. Can be "is entirely located in", "is partially located in", or "is centered in".

A historical polity that is partially located where one or more modern administrative units are now located. Use "is partially located in" (part.loc) with the perspective Geographic Relationship (geo.rel), and create a feature relationship to each modern administrative unit in which the historical polity is partially located.

A historical polity is partially located where one or more modern administrative units are located and also covers the entirety of one or more modern administrative units. For example, a historical polity that covers a geographical area that is currently divided into five counties: one county is entirely located within the area covered by the historical polity, and the geographical area of the historical polity is within *parts* of the other four counties but does not cover the entire county for any of the four. For the relationship of the historical polity to the county entirely located within the area of the historical polity, relationship=“has entirely located within it” (has.entirely.located.within.it); perspective=geo.rel. For the relationship of the historical polity to *each* of the other four counties, relationship=“is partially located in” (part.loc); perspective=geo.rel.

A historical polity is located in a single contemporary administrative unit BUT you are not sure that it did not extend into other contemporary administrative units: relationship=“is partially located in” (part.loc); perspective=geo.rel

A related historical polity Z & cultural region X: Generally, and unlike the relationship of cultural region with modern administrative units, we do not use this relationship for specifying geographic location. In addition, historical polities on the whole tended to have closer relationship with the cultural groups within their territory, with the polity at times being co-extensive with a cultural region. Despite this, the relationship between the two tends to go hand in hand with the geographical character of their relationship - closer if one is centered in the other, farther if one is only partially located in the other. Thus we express this relationship in terms of location - you should use Z is located in Y (type) in "geographical relation" perspective. "Located in" should be either "is entirely located in", "is partially located in", or "is centered in". The relationship could be equivalent - a kingdom which is more or less equal in extent to a cultural region of the same name, or much more peripheral - a kingdom with an extent that ranges over 4 or five different cultural regions. In addition, the relationship can go either way - namely the historical polity might be "entirely located in" the cultural region, or the cultural region might be "entirely located in" the historical polity; the relationship could go in either direction.

• Guideline: historical polity Y is centered in (or other located in relationship) cultural region Y in geographical relationships perspective, etc.
• Example: Degé Kingdom is centered in Degé cultural region.

Two cultural regions that are not hierarchically related to each other: This can be useful given how little we often know about locations of cultural regions - namely we can specify their relationship to other cultural regions.

• Cultural region X is adjacent to cultural region Y from the cultural regions (cult.reg) perspective.

Specifying Location of Features through Specifying Relationship to Other Features

For regions with vague boundaries - such as cultural regions or historical polities - it is impossible to precisely pinpoint the location in terms of latitude and longitude. In these cases, it is common to roughly correlate such regions to contemporary administrative regions - which tend to have precise and known boundaries - so that they can be mapped at least in a rough way. This can be done in the Place Dictionary through using the RELATIONSHIP function. The mapping system then will use those relationships to map the region, if no geometry of its own is expressed.

You have a temple or monastery, and you want to express its location in terms of contemporary administrative units. Guideline: specify it having a relationship to the smallest administrative unit you can specify; relationship type= "is entirely located in (is.contained.by)" , and perspective = "National Administrative Units". If you see multiple places with the relevant name, because you have a "prefecture" called "XXX" and then also a "city" called "XXX", make sure the city is inside the prefecture, and then choose it – if you know the temple is inside the city. If it is on the boundaries and you aren't sure, then choose the larger unit and add a note. If you want to instead say it is "near", "intersects with", etc. then use the appropriate relationship type that best expresses the relationship of the two geographical features.

For cultural regions or historical polities, you may want to specify their geographical location through relating them to contemporary administrative units. This can be done through specifying it being the same as a given administrative unit (use the "is entirely located in (is.contained.by)" relationship), or as being equivalent in spatial extent to multiple contemporary administrative units (for each, use, "is partially located in ( part.loc)" relationship). In addition, is centered in (is.centered.in)/has centered in it (has.centered.in) can be used when, for example, we know that a historical kingdom was centrally located in a territory now correlated to a modern administrative unit, but its full extent may have spilled out into other areas that are now part of other modern administrative units.

Please note that if you are unsure of the location relationship, and thus just want to say such a region is related in some unknown way to the location of a given administrative unit or a set of administrative units, you can use the relationship type intersects with (intersects.with); of you can be even less committal by just using the generic expression "is related to (is.related.to)".

In all of these, the perspective is "Geographical Relationship (geo.rel)" since that is the relationship that deals with locations on a large scale.

Special Issues on Place Names in China

The issue of the relationship between the names of administrative units and their seats is difficult at the prefectural, county and township levels in China. Such difficulties do not typically apply at the provincial level, since the seats tend to be large and prominent cities with very distinct and well known names. The basic problem is that people will often refer to the seat itself as the “prefecture,” “county,” or “township” itself, even though in fact of course it is just one part of the unit in question. Thus, for example, someone might say “I am going to the county,” when in fact they are already in the county and are going to the seat. Likewise, the seat may have its own distinctive name, but locals often are in the habit of referring to it by the name of the county in question, and then even deny the seat has a name other than the name of the county in which it is contained. Thus it can take research and forceful persistence to figure out this issue in any given case.

That said, in some cases, the name of the administrative unit and the seat may be identical. If that is the case, do NOT try to differentiate them by including the name of the type of unit in the name. In other words, the city of Dartsedo and the county of Dartsedo should both be called Dartsedo, and not Dartsedo City and Dartsedo County respectively. The one exception to this is that in China some administrative units have a predominance of a given ethnic minority. In those cases, the name of the unit includes “autonomous” and often also the specification of the dominant ethnic minority by name, for example, “Tibetan Autonomous Prefecture.” In these cases, the type of administrative unit IS included in the actual name according to THL guidelines. Thus “Sichuan” is the name for the province of Sichuan, but we have Ganzi Tibetan Autonomous Prefecture (Ganzi Zangzu zizhizhou / Kandzé Börik Rangkyong Khül) instead of just Ganzi.

8. CAPTION, SUMMARY, and DESCRIPTION COMPONENTS

There are three discrete fields for describing the feature in prose of various length.

Caption

The Caption component contains a condensed version of the Summary and is limited to between 1 and 140 characters. This includes spaces but does not include HTML markup such as <em> </em> to make text italic and so forth. If a caption exceeds 140 characters the page will deliver a general error when updated. There can be only one caption in a given language, but there can be for example a caption in English, a caption in Tibetan, and a caption in Chinese. Captions should be written in complete sentence(s), with a period at the end of each sentence. To create a new caption, one clicks on “New Caption” on the feature’s main editing page. In addition to providing a space for writing the caption itself, you must select the language of the caption from the dropdown menu. You must also select the author of the caption from the dropdown menu. All appropriate tags (event, person, place, etc.) should be identified in the caption. Captions should provide general-level information on the feature and be clear enough to allow a user to identify the feature when viewing it in a list of search results. The caption will not be displayed directly on the feature page in end-user view, but rather will appear in mouse-overs and search results.

Summary

The Summary component consists of a prose summary of the feature limited to between 141 and 750 characters. This includes spaces but does not include HTML markup such as <em> </em> to make text italic and so forth. If a summary is longer than 750 characters, the page will deliver a general error when updated. There can be only one summary in a given language, but there can be for example a summary in English, a summary in Tibetan, and a summary in Chinese. Summaries should be written in complete sentence, with a period at the end of each sentence. To add a summary one clicks on “New Summary” on the feature’s main editing page. On the feature page itself, summaries appear directly below the feature’s name and breadcrumbs. As such, this component plays an important role in a user’s experience of the Place Dictionary and should therefore be accorded special attention. In addition to providing a space for writing the summary itself, you must select the language of the summary from the dropdown menu. You must also select the author of the caption from the dropdown menu. All appropriate tags (event, person, place, etc.) should be identified in the summary. Under certain circumstances, summaries may be easily created by condensing the information found in the feature’s Description(s) component (that is, its essays).

Description/Essay

This allows for an infinite number of descriptions or essays to be attached to a given place entry. Each essay can be of any length; there are no minimum or maximum number of characters required. Each essay has a title, language, and one or more authors specified. While a single essay can be quite lengthy, to facilitate reading over the Web, we suggest authors consider instead submitting lengthy pieces in separate, shorter essays. There can be an unlimited number of essays, including multiple essays in a given language (for example, three essays in Tibetan).

Each essay has a top "title box", where you insert its title. Please put the Title in title case ("An Introduction to Lhasa"). Use simple titles for more descriptive essays ("An Introduction to Lhasa", "The Economy of Lhasa", and so forth), and to repeat the place name in the title. For short, paragraph-length overviews, use the title "An Overview of X". For a longer essay, use the title "An Introduction to X". If your description is more interpretative, then you are encouraged to give it a more unique title (see examples below). Please use a simplified romanization for the place name specified in the title if you are writing in a European language. For standard descriptions, we suggest using the following titles. If your introduction covers all these topics in summary form, then we suggest using the parenthetical title as a section title within the introduction:

• An Introduction to PLACE NAME
  • The Economic Dimensions of PLACE NAME (Economy)
  • The Environmental Dimensions of PLACE NAME (Environment)
  • The Religious Dimensions of PLACE NAME (Religion)
• The Cultural Traditions in PLACE NAME
• Tourism in PLACE NAME
• A History of PLACE NAME

To specify the author, click on the "Add new Author" link at the bottom and chose an author. You can add as many authors as you like. If the author's name is not on the list, please contact us to add him/her and we will do it promptly.

At the bottom is a checkbox that is used for indicating if the description in question should be considered the primary description. If you check it, then this description will become the primary description, and will be located first in the list of descriptions for this place, if there are more than one descriptions. The primary description should be a general introduction to the place, not a focus on one aspect of the place, or a highly interpretative essay.

The Caption/Summary/Essay Editor

The WYSIWYG (“What You See Is What You Get”) rich text editor for descriptions and essays is a THL editor that is used across THL in many applications. For that reason, we keep its documentation separate so that a number of editorial manuals can refer to it. For anyone writing captions, summaries, and/or essays/descriptions, it is essential that you refer to the THL Online Essay Editor Manual for details on how to format your text and use the editor.

Illustration Component

The Illustration component consists of one or more images attached to the feature; the main image associated with a given feature will appear to the right of the feature’s map element. To add a new illustration one clicks on “New Illustration” on the feature’s main editing page. The editor must first select whether an illustration has an MMS ID (that is, it is in THL's image collection in the MMS) or is “external” (that is, it is found elsewhere on the web). An external illustration must be loaded from a known URL. Ideally all images should be located from a THL collection – thus forgoing the need for external URLs – but in the event that one does need to use an external image, its associated terms of use should be consulted prior to entering data for it in the edit interface. For images internal to THL, one can find the MMS ID by looking up the image in the corresponding THL collection (http://mms.thlib.org/media_objects?type=Picture). One difference between the two types of images is found in the user’s ability to edit image captions; externally linked images can be freely given a new caption by the editor, while images linked through an MMS ID come pre-associated with a caption that cannot be edited except in the MMS record for the image itself. A final note: currently there is a known bug affecting the justification of illustrations vis-à-vis the map element on a given feature’s front page. This is caused by the image caption exceeding a certain length and for now can be corrected by shortening the image caption at fault, except in the case of MMS images, whose captions come pre-loaded from THL.

NOTE: in the future this will include images from Shared Shelf and will involve the Drupal Gallery image management module.

Project Component

The project component identifies the project or projects that a feature is part of, such as the Kham monasteries project, or the Center for Contemplative Sciences (or one of its sub-projects) and so forth. A feature can be associated with more than one project. If the project you want to enter is not in the list of projects when you pull down the menu in the main edit interface page and select project, then contact us to add the project to the list.

Provided for unrestricted use by the Tibetan and Himalayan Library