IS:additionalName ocp

From wiki.railML.org
Jump to: navigation, search

additionalName
 


Scheme description / Schemenbeschreibung / Description du schéma

Position of additionalName in the XML-Tree / Position von additionalName im XML-Baum / position de additionalName dans l’aborescence XML

  • Children: None

Multiplicity / Anzahl / Multiplicité

[0..∞]

Semantics / Bedeutung / Sémantique

The element <additionalName> provides additional names with descriptions in other languages, dialects, encodings...

Attributes of additionalName / Attribute von additionalName / Attributs de additionalName


  • type: This is the meaning of the name. Possible values are:
    • operationalName the ocps name under operational aspects
    • trafficName the ocps name under traffic aspects
    • localName an name in the local language
    • other:anything Any value that does not fit any value from the previous enumeration list, fulfilling the constraint: at minimum two characters, whitespace is not allowed.

    (introduced with version 2.2)

Syntactic Constraints / Syntaktische Beschränkungen / Contraintes syntactiques

One <ocp> can have an unlimited number of <additionalName> elements.
  • type: union of (restriction of xs:string, tOtherEnumerationValue); tOtherEnumerationValue is an arbitrary string starting with 'other:' followed by at minimum two characters, white space not allowed for extending railML enumeration lists; optional

Best practice & Examples / Empfohlene Anwendung & Beispiele / Bonnes pratiques & exemples

Stations with public names in different languages

(deprecated with version 2.1)

<ocp id="ocp_DBZ" name="Bautzen">
  <propOther>
    <additionalName value="Bautzen" type="trafficName" xml:lang="de"/>
    <additionalName value="Budyšin" type="trafficName" xml:lang="hsb"/>
  </propOther>
</ocp>

(introduced with version 2.1)

<ocp id="ocp_DBZ" name="Bautzen">
  <additionalName name="Bautzen" xml:lang="de"/>
  <additionalName name="Budyšin" xml:lang="hsb"/>
</ocp>

Since both operational and public names are identical in the default language of the file (German, de), it is not necessary to repeat the German public name. But it is recommended to do so for clarification especially from railML 2.1 when type became deprecated.

Stations with simply a difference between operational and public name

<ocp id="ocp_LF" name="Bft. Falkenberg (E) ob. Bf.">
  <additionalName name="Falkenberg (Elster)"/>
</ocp>

No need to use the xml:lang attribute at all since there is no matter of different languages.

Sometimes, at stations of DB Netz AG, there is a kind of semi-difference between operational and public names because of some strange shortening of operational names of Deutsche Bahn due to the usage of out-dated software… There is no direct need to use the shortened names in railML but sometimes it is necessary simple for exact data transfer.

<ocp id="ocp_BL" name="Berlin Hbf-Le Bf">
  <additionalName name="Berlin Hauptbahnhof - Lehrter Bahnhof"/>
</ocp>

Stations with different operational and public names

(deprecated with version 2.1)

<ocp id="ocp_DG" name="Görlitz">
  <propOther>
    <additionalName value="Bft. Görlitz Pbf." type="operationalName"/>
    <additionalName value="Zhorjelc" type="trafficName" xml:lang="pl"/>
  </propOther>
</ocp>

(introduced with version 2.1)

<ocp id="ocp_DG" name="Bft. Görlitz Pbf.">
  <additionalName name="Görlitz" xml:lang="de"/>
  <additionalName name="Zhorjelc" xml:lang="pl"/>
</ocp>

Please note that there is never a defined nor guaranteed order of sequenced elements of the same type in railML. So a reading software should not assume the first additional name to be the default one! In some cases, there even would not be an agreed default language for public station names.

<ocp id="ocp_CRU" name="Crianlarich upper">
  <additionalName name="A’ Chrìon Làraich" xml:lang="gd"/>
  <additionalName name="Crianlarich" xml:lang="en"/>
</ocp>

”Foreign” stations outside the validity of the default language of the railML file

The following example shall be assumed to be part of a railML file of mainly German stations with German as the default language.

<ocp id="ocp_XPKZG" name="Krzewina Zgorzelecka" xml:lang="pl">
  <additionalName name="Krzewina Zgorzelecka" xml:lang="pl"/>
  <additionalName name="Ostritz" xml:lang="de"/>
</ocp>

Here, the xml:lang attribute of the parent element is used to declare that the language of the operational name is not the default language of the file. Again, this name is optionally repeated to clarify that there is no difference between operational and public name in Polish. At the German name, the usage of xml:lang="de" also is optional - since it is the default language - but recommended for clarification.

This example is very typical for border regions anywhere in the world of railways. Nowadays, it is common for interactive public information systems (such as online timetable search machines) to accept all spellings of a station. (You can type “Prague” or “Prag” but do not need to know that the official spelling is “Praha”.) The example also shows that there is no objective ‘first’ additional language.

Station names in different scripts

<ocp id="ocp_ΑΘΗΝ" name="ΑΘΗΝΑ">
  <additionalName name="ΑΘΗΝΑ" xml:lang="el-Grek"/>
  <additionalName name="ATHINA" xml:lang="el-Latn"/>
</ocp>

Since ‘-Grek’ is the suppress-script sub-tag of Greek (‘el’), it would not be necessary to specify this sub-tag in the first additional name (Greek with Greek letters). Since (modern) Greek is assumed to be the default language of the file, the whole first xml:lang attribute could be omitted.

In Greece, it is even common up to now to distinguish between ancient and modern Greek. Combined with the spelling of some western ‘tourist’ languages, all the names of Athens main station look like this:

<ocp id="ocp_ΑΘΗΝ" name="ΑΘΗΝΑ">
  <additionalName name="Ἀθῆναι" xml:lang="grc"/>
  <additionalName name="ΑΘΗΝΑ" xml:lang="el-Grek"/>
  <additionalName name="ATHINA" xml:lang="el-Latn"/>
  <additionalName name="Athens" xml:lang="en"/>
  <additionalName name="Athen" xml:lang="de"/>
  <additionalName name="Athènes" xml:lang="fr"/>
</ocp>


Some cases where railML currently does not help

(not to be taken too seriously)

There is currently no possibility to distinguish between an old and a new spelling of a renamed station in the same language:

<ocp id="ocp_PRMN" name="Praha Masarykovo nádraží">
  <additionalName name="Praha Masarykovo nádraží" xml:lang="cs"/>	<!-- 1919-1940, 1990 ff. -->
  <additionalName name="Praha střed" xml:lang="cs"/>		<!-- 1953-1990 -->
  <additionalName name="Praha Hybernské nádraží" xml:lang="cs"/>	<!-- 1940-1945 -->
  <additionalName name="Praha státní nádraží" xml:lang="cs"/>	<!-- 1862-1919 -->
</ocp>

There is also no possibility to specify ‘vernacular’ names:

<ocp id="ocp_DBW" name="Schiebock" xml:lang="de-DD???" >
  <additionalName name="Biskopicy" xml:lang="hsb"/>
  <additionalName name="Bischofswerda" xml:lang="de"/>
</ocp>

And what to do with a (more or less) short and a (very!) long name version in the same language?

<ocp id="ocp_LLAN" name="Llanfair PG" xml:lang="cy">
  <additionalName name="Llanfair Pwllgwyngyll" xml:lang="cy"/>
  <additionalName name="Llanfairpwllgwyngyllgogerychwyrndrobwllllantysiliogogogoch" xml:lang="cy"/>
</ocp>

Notes / Anmerkungen / Notes


Concerning ocps in the meaning of stations for railway access, there are often station names in more than one language due to local habit. Besides this, even in single-language regions, there may be a public name of a station different from the operational name.

  • It is recommended to use the direct name attribute of an ocp for the operational name and to list any public name in additionalName elements.
  • In cases where there is no difference between the operational and the only public name, or if public names are not relevant or not known, it's ok to use the direct name attribute only.
  • If necessary, the direct xml:lang attribute holds the language and/or script of the operational name if it differs from the semantical ‘natural’ or expected value. It is not common to specify the language of the operational names in railML because this would mean a massive usage of xml:lang attributes throughout the files. Rather, if deemed necessary, a header value can be used (see <dc:language>).
  • As long as there is one additionalName element only (at one ocp), its xml:lang attribute can be omitted if the language and script are the same as the global operational.
  • As soon as there is more than one additionalName, their xml:lang attribute shall be used to distinguish them whenever possible.
  • As recommended in IETF BCP 47, it is not necessary to specify the script nor region sub-tags or any other sub-tags if they are the default values for the actual language (defined at IANA Internet Assigned Numbers Authority (external link)). So for most cases in railML, only the primary language tag is needed.
  • Especially it is not needed to specify the region sub-tag for station names in railML except it is really needed to distinguish between two regional dialects of the same language (which, concerning station names, is hard to imagine) but not to repeat the country where the station is situated.
  • Since railML files are normally coded in Unicode (UTF-8), there is never a real need to specify the script sub-tag - the script (character sub-set) follows implicitly from the characters used in the name. So, the only case when to use the script sub-tag is to distinguish between two <additionalName> elements of the same language. This is common in regions where the native station names are not written with Latin script for better international understanding (see examples below).
  • Note that it is not possible (by definition) to use the script sub-tag to vary the character set of the railML file (defined in <?xml encoding=…?> at the head line of the file). The script sub-tag is only a kind of hint for the reading software to assign the station name to a typical usage or give it a “right to exist”. But it does not change the behaviour of a railML parser when decoding the file. So, if the file is not coded in Unicode for some reason, it is not possible to use other characters in names than defined by the character set of the encoding attribute.

Open issues / Offene Punkte/Pedenzen / Questions ouvertes

Not yet described. / Noch nicht beschrieben. / Pas encore décrit.