[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: FW: Revisions to Acronym proposal
JoAnn T. Hackos, PhD President Comtech Services, Inc. 710 Kipling Street, Suite 400 Denver, CO 80215 303-232-7586 joann.hackos@comtech-serv.com joannhackos Skype www.comtech-serv.com -----Original Message----- From: Kara Warburton [mailto:KARA@CA.IBM.COM] Sent: Tuesday, January 22, 2008 7:22 AM To: Bruce Esrig Cc: bhertz@sdl.com; Bryan Schnabel; christian.lieske@sap.com; dpooley@sdl.com; Erik Hennum; fsasaki@w3.org; ishida@w3.org; JoAnn Hackos; jogden@ptc.com; Michael Priestley; rfletcher@sdl.com; rmraya@maxprograms.com; Robert D Anderson; tony.jewtushenko@productinnovator.com; ysavourel@translate.com Subject: Re: Revisions to Acronym proposal Hi Bruce, I think I can comment on the last 2 points. My understanding is that the glossary XML file is not a reflection of the published format. During transform processing, the headword ('title' -- <glossTerm>) of the glossary entry would be given a published glossary entry with the definition. Other items such as part of speech and usage note, etc., would normally be omitted from the published glossary. All the <glossAlt> terms would also have an entry in the published glossary, but instead of a definition, there would be a "see" reference to the full entry represented by <glossTerm>. This style is according to industry practice and should be handled by transform processing. If one of the <glossAlt> terms has a "preferred" status value, it would assume the role of the so-called main term and would therefore get the full entry and in this case <glossTerm> would have a "see" reference pointing to it. Therefore, in the DITA glossary source, one should not have to "duplicate" entries in order to get a headword for each term in a DITA glossentry. For the people on the list, I would like to just clarify the role of the "glossStatus" element. It provides two useful functions: 1. Enabling content authors to indicate, among synonyms and variants, the term which is preferred and terms which should be avoided. This data can be pulled into both controlled authoring software and terminology databases to enable controlled authoring, which improves content and benefits translation. 2. It enables authors to indicate when an abbreviation or an acronym is actually the preferred term for use. In general domains this is not very frequent but it does occur. Take for example an abbreviation such as "ping" in the computing field. How many people would recognize this concept if I used the full form "packet internet groper"? Similarly, in a banking glossary, if I put the definition of ATM under the headword "Automated Teller Machine", with a "see" reference from ATM to "Automated Teller Machine", I would be doing my readership a disservice because I can almost guarantee that they will be searching in the glossary for ATM (if there is anyone left on this earth who doesn't know this concept) and not the full form. They would reach the definition they are looking for in 2 steps instead of 1 step. Also in some vertical industries such as telecommunications and aerospace, etc., abbreviations are very common and there may be a need to establish precedence of these terms over their corresponding full forms in content authoring and translation. In the case of ping and ATM, simply marking the <glossAcronym> with a "preferred" status value solves these issues. Kara Warburton IBM Terminology 905-413-2170 IBM Intranet links: Terminology WIKI: https://w3.webahead.ibm.com/w3ki/display/IBMterm/Home IBM terminology: http://w3.ibm.com/standards/terminology Terminology blog: http://blogs.tap.ibm.com/weblogs/page/kara@ca.ibm.com Bruce Esrig <esrig-ia@esrig.c om> To Robert D Anderson 22/01/2008 04:01 <robander@us.ibm.com>, "JoAnn AM Hackos" <joann.hackos@comtech-serv.com> cc <bhertz@sdl.com>, "Bryan Schnabel" <bryan.s.schnabel@tek.com>, <christian.lieske@sap.com>, <dpooley@sdl.com>, <fsasaki@w3.org>, <rfletcher@sdl.com>, <ishida@w3.org>, <rmraya@maxprograms.com>, <tony.jewtushenko@productinnovator. com>, Kara Warburton/Toronto/IBM@IBMCA, <ysavourel@translate.com>, Erik Hennum <ehennum@us.ibm.com>, Michael Priestley/Toronto/IBM@IBMCA, jogden@ptc.com Subject Re: Revisions to Acronym proposal Thanks again for the copy ... I have four remarks. 3a. Is there an enumeration value in order to instruct that the first occurrence be of the form "expansion (acronym)"? If not, authors who want this format will have to know which occurrence is the first. If they must explicitly write "<abbreviated-form> (<abbreviated-form>)", then they must do so at the first occurrence. 3b. Translators should not fill in the expanded form where the acronym is requested if my suggestion in 3a is implemented. Otherwise, "expansion (expansion)" will result at the first occurrence in the translation. There seems to be a process issue, which is that translators need a way to indicate that they have completed translating a glossary entry, including the acronym. An alternate way of supporting that need would be to provide an explicit attribute that the translator sets to indicate that the acronym has been left empty deliberately. Then an empty attribute is consistently a sign that the expanded form should be used. 1. Are there controls to permit alternate titles (aside from the expanded form) for the glossary topics in the generated glossary? These would indicate that in the generated glossary, the topic should appear under the expanded form, under the abbreviated form, or both. If the abbreviated form is not unique, the control could be on that element so that multiple instances could be generated. This mechanism assumes that there is a capability to sort the generated glossary topics by title. 6. Is there a see/see-also/compare mechanism for glossary entries as was discussed for index entries? Best wishes, Bruce At 07:14 PM 1/21/2008, Robert D Anderson wrote: >Hi, > >As we agreed at today's call, I had a follow-up meeting with Erik Hennum to >discuss the translation group's suggestions. As a result, there are now >several concrete suggestions as changes to the original proposal ( >http://lists.oasis-open.org/archives/dita-translation/200801/msg00003.h tml >). I'll list the specifics below, and then send a more general note to the >main TC list. > >1. The group suggested that expanded forms should come first. So, the >proposal can remove glossAcronym and glossAbbreviation as options for the >main 'title' of the glossary topic. Instead, the expanded term will be >located first, using the glossterm markup that is used for every glossary >entry. The glossAcronym or glossAbbreviation are listed after that as >alternate versions of the expanded term. > >2. The glossFullForm (also called glossExpandedForm) element can be >removed, because the expanded form of the term is already encoded as the >term. > >3. When authors reference an acronym inline with the <abbreviated-form> >element, the rules will follow those defined by the translation >subcommittee: >- The first instance in one context will pull text from the >glossSurfaceForm >- Future instances will pull from the first glossAcronym (similar to >today's discussion, in that we cannot prevent many but users should only >have one) >- The consensus on today's call was that if an acronym is not available in >a target language, the expanded value should be specified. However, if the >acronym is empty because a translator removed it, the primary term (the >expanded form) should be used. > >4. The existing <glossStatus> element has a new value of "preferred" in the >enumeration. This can be used to indicate that an alternate form (such as >an acronym) is the preferred representation of a term. This is still not a >required element for acronym processing. If a <term> element is used to >reference a glossary entry, the following rules will be used when >retrieving the term: >- As a first choice, the form of the term marked "preferred" will be >retrieved >- If no term is marked preferred, or if the preferred term is empty >(possibly because it does not exist in a target language), the primary term >will be retrieved >- NOTE: using <abbreviated-form> will always use the processing rules >defined in #3, regardless of what is marked "preferred" > >5. As mentioned at the call, the language in the proposal about acronyms >needs to be cleaned up to include all of the translation issues described >in the original proposal. > >Please send any comments back to the list. Unfortunately Erik hasn't had >time to review this, so if I mis-characterized something he may be speaking >up soon. > >Robert D Anderson >IBM Authoring Tools Development >Chief Architect, DITA Open Toolkit >(507) 253-8787, T/L 553-8787 (Good Monday & Thursday)
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]