- From: Stella Mitchell <cleo@us.ibm.com>
- Date: Mon, 25 Aug 2008 09:42:23 -0400
- To: "Adrian Paschke" <adrian.paschke@biotec.tu-dresden.de>
- Cc: Leora Morgenstern <leora@us.ibm.com>, public-rif-wg@w3.org
- Message-ID: <OFBC07F2FE.E8378EBE-ON852574B0.0049D8C3-852574B0.004B4AC2@us.ibm.com>
Ok. I wasn't arguing for any particular syntax. Just that for the test cases we use the defined BLD syntax, including any changes to it that the WG agrees to make. Stella "Adrian Paschke" <adrian.paschke@biotec.tu-dresden.de> 08/25/2008 08:50 AM To Leora Morgenstern/Watson/IBM@IBMUS, Stella Mitchell/Watson/IBM@IBMUS cc <public-rif-wg@w3.org> Subject AW: [RIF Test Cases] Abridged Presentation syntax True, no problem to fix the examples and use explicit Forall, And etc. But, one more argument is: As Leora already pointed out these conventions and short cuts are frequently used in logic / logic programming. There is a long history here which we can and should not neglect. Language standards, e.g. ISO Prolog syntax, as well as many tools for engineering, editing, translating etc. exist. If our abridged presentation syntax is close to what is common practice we will support reusability of existing tools, will make it easy to learn and understand RIF, and will increase adoption of it. Just my two cents, Adrian Von: public-rif-wg-request@w3.org [mailto:public-rif-wg-request@w3.org] Im Auftrag von Leora Morgenstern Gesendet: Montag, 25. August 2008 14:30 An: Stella Mitchell Cc: public-rif-wg@w3.org Betreff: Re: [RIF Test Cases] Abridged Presentation syntax You're right, Stella! I had missed that. Thanks for pointing it out. In any case, those are easy to fix, and less of an issue than the abbreviations listed in the table. Leora Stella Mitchell/Watson/IBM 08/25/2008 08:24 AM To Leora Morgenstern/Watson/IBM cc public-rif-wg@w3.org Subject Re: [RIF Test Cases] Abridged Presentation syntaxLink They're listed as part of the abridged presentation syntax, immediately below the table. Stella Leora Morgenstern/Watson/IBM 08/25/2008 08:16 AM To Stella Mitchell/Watson/IBM@IBMUS cc "Adrian Paschke" <Adrian.Paschke@gmx.de>, "Hassan Aït-Kaci" <hak@ilog.com>, public-rif-wg@w3.org Subject Re: [RIF Test Cases] Abridged Presentation syntaxLink Stella, >The syntax left in UCR that is not official PS syntax is: > - shortcuts for external functions & predicates > - can leave out "And" and conjuction is assumed >- can leave out Forall for rules with variables and assume variables are quantified > - denoting the end of a rule with a period The last three are actually not even part of the Abridged Presentation Syntax, and do not figure in Adrian's table mapping Abridged Presentation Syntax to Presentation Syntax. They are just conventions that are frequently used in logical formalizations/ logic programming, and have crept into some of the examples. It is easy to fix the examples so that these aren't used, and I agree that we should do so. Regarding the first: I agree also that it would be best to have one unified convention for shortcuts. While we are revisiting the BLD syntax, as Adrian suggested, in order to formalize various open issues. it would be worthwhile discussing whether we should integrate some of the shortcuts of the Abbreviated Presentation Syntax into the shortcuts already defined in the BLD document. Leora Stella Mitchell/Watson/IBM@IBMUS Sent by: public-rif-wg-request@w3.org 08/23/2008 07:20 AM To "Adrian Paschke" <Adrian.Paschke@gmx.de> cc "Hassan Aït-Kaci" <hak@ilog.com>, public-rif-wg@w3.org, public-rif-wg-request@w3.org Subject Re: [RIF Test Cases] Abridged Presentation syntax Adrian, As I mentioned several times during our internal discussions, it would be good to reconcile with the shortcuts defined for BLD PS. In your UCR abridged presentation syntax table [2], rows 1, 2, 3, 5, 6 and 7 are redundant with the shortcuts in DTB [1]. Row 4 is about text with a language tag. Axel currently has a proposal for a shortcut for this in [3]. The syntax left in UCR that is not official PS syntax is: - shortcuts for external functions & predicates - can leave out "And" and conjuction is assumed - can leave out Forall for rules with variables and assume variables are quantified - denoting the end of a rule with a period For the (BLD) test cases, I think we should stay with normative BLD syntax, and only use shortcuts that the group agrees to add to that syntax. Stella [1] http://www.w3.org/2005/rules/wiki/DTB#Shortcuts_for_Constants_in_RIF.27s_Presentation_Syntax [2] http://www.w3.org/2005/rules/wiki/UCR#Use_Cases [3] http://lists.w3.org/Archives/Public/public-rif-wg/2008Aug/0074.html "Adrian Paschke" <Adrian.Paschke@gmx.de> Sent by: public-rif-wg-request@w3.org 08/23/2008 05:24 AM To "Hassan Aït-Kaci" <hak@ilog.com> cc public-rif-wg@w3.org Subject Re: [RIF Test Cases] Abridged Presentation syntax H Hassan, > > Hi Adrian, > > I appreciate the motivation you invoke for introducing an Abridged > Presentation Syntax (APS). This makes an APS expression an abbreviation > of a PS expression, itself an abbreviation for the normative XML syntax. > In DTB, Axel Polleres already defined abbreviations (although formally: > http://www.w3.org/2005/rules/wiki/DTB#Shortcuts_for_Constants_in_RIF.27s_Presentation_Syntax ) > and he used them for the examples of the DTB document. My question is, > are the abridged forms for DTB consistent with those your introduce > in http://www.w3.org/2005/rules/wiki/UCR#Use_Cases? If not, can it be > made so? If so, could the two be specified formally along the rest of > the EBNF rules for BLD Rules and BLD Conditions? Thanks for the hint - didn't know about the new abridged syntax in DTB. At the time when we specified the abridged presentation syntax (actually it was you and Harold who started it) for UCR, there was nothing in DTB. Will check if both are consistent or something is missing in DTB. Moreover, I will then simply link to DTB for the syntax of examples in UCR. >Currently Axel's rules > are hyperlinked to where they are defined in other W3C documents all > over the place as an extension of the original BLD EBNF. It would be > good to put together somewhere the complete set of lexical and syntactic > EBNF rules for the real PS as it is actually used in *all* RIF documents > (whether UCR, Core, BLD, DTB, PRD, ...). Yes, agree. I think it is now time to review again the full and abriged presentation syntax, complete it, and solve open issues. For instance, I remember the ":-" or "=>" question for rules presentation. We need final versions to update the examples in UCR, PRD, Core, ... and Test Cases. >That would surely help *me* in > my task to provide a correct working APS parser to automate generating > the serialized XML for of all the examples in these documents > (http://www.w3.org/2008/08/19-rif-minutes.html#action08). > > At any rate, I will be on vacation from tonight through August 31 and I > will resume work on this after I am back. Enjoy your vacations. -Adrian > > Adrian Paschke wrote: > > Hi Jos, > > > > You asked: > > > >> In > >> addition, it is unclear to me which syntax they use. it is certainly > not > >> valid presentation syntax. > > > > It is the abridged presentation syntax from UCR > > (http://www.w3.org/2005/rules/wiki/UCR#Use_Cases) which was introduced > there > > to have a very compact and easy readable human-oriented format. > > > > It might be also usable for the test cases to get a quick picture what > the > > rules of the test case are. The full presentation syntax can become very > > complex, take e.g. the simple example of "?X>= (?Y+2) " which would be > very > > long-winded in the full presentation syntax and hard to read for a > human. > > > > But you are right; we need full presentation syntax to automatically > > translate them into the concrete XML syntax. > > > > Hence, I would propose to describe the test cases in full presentation > > syntax in the premises and conclusion field (or alternatively already in > > concrete XML syntax) and optionally represent them in abridged > presentation > > syntax together with the narrative description of the test case in the > > "Description" field. > > An alternative would be to have several (optional) premise / conclusion > > fields which represent the test case in different syntaxes (abridge, > full, > > XML, PRD, BLD, ...). > > > > - Adrian > > > > > > -----Ursprüngliche Nachricht----- > > Von: public-rif-wg-request@w3.org [mailto:public-rif-wg-request@w3.org ] > Im > > Auftrag von Jos de Bruijn > > Gesendet: Donnerstag, 21. August 2008 15:46 > > An: Chris Welty > > Cc: Public-Rif-Wg (E-mail) > > Betreff: Re: Call for test cases > > > > > > > >> Below are instructions to create new test cases on the WIKI. The test > >> cases will be automatically classified into the category of the used > >> template and the specified dialect. We probably might need more > >> templates (categories) later, as described here > >> http://www.w3.org/2005/rules/wiki/Test#Categories_of_RIF_Test_Cases > >> > >> But let's start simple first and collect positive entailment tests > which > >> demonstrate BLD and DTB. > >> > >> The properties of the templates for test cases are described here > >> http://www.w3.org/2005/rules/wiki/Test_Case_Format > > > > The page lacks descriptions of the properties Text and Format and > > guidelines about how to format the title. > > Then, it is not very clear to me what the difference is between the > > properties Purpose and Description. > > > >> Some example test cases for BLD (positive entailment test cases) can be > >> found here > >> > >> http://www.w3.org/2005/rules/wiki/Test_Case_Ordered_Relations > >> http://www.w3.org/2005/rules/wiki/Test_Case_Unordered_Relations > >> http://www.w3.org/2005/rules/wiki/Test_Case_Frames > >> http://www.w3.org/2005/rules/wiki/Test_Case_Equality > > > > All the examples like the required properties title and purpose. In > > addition, it is unclear to me which syntax they use. it is certainly not > > valid presentation syntax. > > > > > > I tried to write a test case (a negative entailment test), but I was not > > sure whether it is in the correct format. Please check: > > http://www.w3.org/2005/rules/wiki/Test_Case_Local_Constant > > > > > > Best, Jos > > > > > > > > > > > > > -- > Hassan Aït-Kaci * ILOG, Inc. - Product Division R&D > http://koala.ilog.fr/wiki/bin/view/Main/HassanAitKaci > -- Ist Ihr Browser Vista-kompatibel? Jetzt die neuesten Browser-Versionen downloaden: http://www.gmx.net/de/go/browser
Received on Monday, 25 August 2008 13:43:34 UTC