[SCL] Re: Report on Common Logic

pat hayes phayes at ihmc.us
Mon Nov 3 14:00:16 CST 2003


>pat hayes wrote:
>>John Sowa wrote:
>>>Murray Altheim wrote:
>>
>>>>After that's done, an XML syntax still isn't completely
>>>>straightforward. Like any language, you need to decide exactly
>>>>what you're trying to do with it. Just *one* of the questions
>>>>I brought up early on was "web-enabled" or not? Use of URIs or
>>>>not? Any extensibility, so it can be the basis of other languages?
>>>
>>>Those answers were decided long ago:  web enabled, yes;
>>>URLs, yes; extensibility, yes; and basis for most common
>>>declarative languages, yes.
>>
>>Wait: when was that decided? To create a web-enabled syntax is 
>>indeed one goal, but not the immediate one (ie aiming at Dec 17); 
>>the aim there should be a basic SCL-XML exporting syntax allowing 
>>SCL to be ported through an XML pipe, with the minimum of bells and 
>>whistles. URLs are permitted but not obligatory (as in SCL 
>>generally) and I do not feel that the XML syntax should be 
>>particularly extendable or web-enabled (whatever that means, it 
>>requires a lot more discussion): SCL is itself extendable, but 
>>there is no need to provide for extensions in (this) XML syntax: 
>>any extensions would be changes to the syntax and hence to the XML. 
>>So my answers (in the short term) would be:
>>web enabled, no; URLs are used but not mandatory; extensibility, 
>>no. I realize this makes it into a very boring exercise for an 
>>XML/web expert, but that is what we need done by the 17th (actually 
>>by about the 12th).
>
>I'm not concerned any more than you with this being "boring". I am
>concerned with its usability and whether or not it meets a set of
>requirements. Those two things are of course inextricably tied. I
>would still prefer a set of specific requirements for this work; I'm
>guessing this isn't going to happen, so perhaps I can just field
>questions and see if nobody screams when I answer them as best I can
>based on what I see on the list, rather than in a requirements document.

OK, since time is short let me suggest answers to your questions. 
Anyone else disagree, chime in ASAP.

>   Web Enabled:
>      1. Do we want XCL documents to be able to refer to other XCL
>         documents over the web? This would be how modules could be used
>         to construct larger statements/formulae/ontologies/etc.

If this can be managed easily, yes. I think there is only  a need to 
do this at the 'top level', ie for one ontology to consist of several 
XCL documents. One way to do this would be to include a kind of 
'importing' mechanism, meaning: include all the SCL axioms from that 
document in here.

I think it would be good if XCL could be mixed with other content in 
an XML document without that affecting the SCL content.

>      2. are XCL documents meant to be exchanged over the Web?

By being given a URI and retrieved as XML, sure.

>      3. are there meant to be internal link targets within an XCL document?

Not it the first instance, I think.

>      4. compatible with XLink or use its own linking structure?

Your decision. It would be good if it were *possible* to use DTDs with XCL.

>
>   URIs or not:
>      1. If URIs are permitted in SCL, do they actually *mean* URIs, or
>         are they just string identifiers?

Delicate question. As far as the SCL syntax is concerned, it would be 
kosher to distinguish a URTi from an isomorphic character string as 
slong as there was some way to distinguish them syntactically, but it 
would also be kosher to treat them as indistinguishable.

>Is there any sense of them
>         being dereferenced?

Apart from the links in the 'imports' meachanism, no.

>Is there any requirement as to what is at
>         the other end of the reference?

I would be happy to say that if it is SCL then we use it, and if it 
isn't then we ignore it.

>      1. are URIs in XCL merely meant to be contained as strings, or
>         actually treated (in syntax) as URIs?

The abstract syntax is agnostic on this. I think we should take the 
simplest possible path consistent with being able to reconstitute the 
SCL itself from the XCL markup, which I think would be the answer 
uris=strings, at this stage. We can return to designing a more 
web-savvy XML version later.

>Is there to be any
>         differentiation between their various uses (as strings or as
>         links)?

No. Their use as links is irrelevant to SCL.

>In the case of links, do you want a mechanism to include
>         link semantics (e.g., role attributes), so that one *kind* of link
>         is differentiated from another? Is this list of link semantics
>         fixed or extensible? E.g., In HTML, there are many dozens of kinds
>         of links and link semantics, some fixed, some stated by the user.

These are exactly the issues I want to avoid at present and return to 
later. They are not part of the central SCL proposal, more to do with 
designing a Web logic based on SCL.

>      2. Are there any restrictions on the use of URIs within XCL?

No, because the issue does not arise, ie if some part of SCL syntax 
is a link then this is irrelevant to its role in SCL syntax.

>A query
>         language beyond fragment IDs ("#foo")? Use of XPath suggested or
>         permitted? If only intra-document links are used, do you want to
>         use XML IDs (which are guaranteed unique), but have naming
>         constraints (XML Names only, no weird characters or spaces)? If
>         not, how do you want to design the link structure? (There are a
>         *lot* more questions then.)
>      3. If URIs are not used, how are references made between various
>         parts of the grammar?

I don't understand the question. What kind of references?

>Are there one or two different mechanisms?
>         If two, how do they interact? (I'd recommend one: use URIs, but
>         if you don't want external linking to other documents, simply
>         prohibit it, either in the spec prose or in the DTD/schema, and
>         use the XML IDs.)
>
>   Extensible:
>      1. Are documents conforming to the XCL spec required to include
>         only XCL content, or can it be mixed with other markup?

Mxing is better as long as we can cleanly separate the SCL from the 
rest. We have a proposal for mixing arbitrary other markup into SCL 
(as 'wrappers') but if this can be done more neatly in XML then by 
all means make a proposal. I would like to be able to mix XCL into 
XHTML 'invisibly' to both, but I doubt if we can get that done in the 
next week or so.

>I.e.,
>         the XML Topic Maps spec forbids any mixing within an XTM document.
>      2. Do we want the element and attribute names to be fixed?

In the first instance lets do that. We can always make a more 
sophisticated proposal later.

>  A reason
>         why we might is that XCL may be used for different purposes. For
>         authoring and reading the markup (which will often or even
>         usually be done by humans, sad to say), we'll want the element names
>         to be human-readable (e.g., "<formula>", "equiv"), whereas for
>         really large documents, people may want "<fm>" "eq", etc. to save
>         space. This kind of thing reduces interoperability, but XSLT can
>         easily be used for transforms. It'll also depend on the markup-
>         to-content ratio.
>      3. Is there one XML Namespace prefix permitted, or anything goes?
>         I.e., is the default (either "" or "xcl:") fixed?

Well, SCL itself is an abstract syntax with many (possibly 
open-ended) concrete instances, so ideally we would want any 
namespace to be used, but the namespace defines the SCL syntactic 
categories for a particular SCL language. Something like cfopc:scland 
versus kif:scland versus cg:scland (?) Or is it possible to declare 
that cfopc and kif and so on are SCL-kosher and then just use the 
namespace with that assumption?

>      4. Are there any DTD- or schema-based mechanisms provided for
>         extending XCL, or is this left to others? (If not, the ways that
>         people might alter the DTD may create incompatibilities, whereas
>         if everyone uses a provided mechanism, this will make DTDs/schemas
>         interoperable and easier to use).

I would prefer a single mechanism to be at least provided, but at 
this point my XML skills run out.

>      5. If it's know that XCL will be extended beforehand, do we want to
>         provide the hooks so that those DTDs/schemas can be built upon
>         the core XCL DTD/schema, or would those projects have to start
>         from scratch?
>
>Some of these things are implemented in the DTD/schema, others are
>constrained in the prose of the spec.
>
>There are perhaps other issues, I'm just pointing out that this is not
>straightforward.

I understand it is not, which is why I don't think there will be time 
to get it adequately done by the 17th.

Pat

>
>Murray
>
>......................................................................
>Murray Altheim                    http://kmi.open.ac.uk/people/murray/
>Knowledge Media Institute
>The Open University, Milton Keynes, Bucks, MK7 6AA, UK               .
>
>  White House defiant despite deadliest strike on forces since end of war
>    http://www.guardian.co.uk/Iraq/Story/0,2763,1076578,00.html
>  US defiant after Baghdad attack
>    http://news.bbc.co.uk/1/hi/world/middle_east/3186074.stm
>  US defiant on X-Ray prisoners
>    http://old.smh.com.au/news/0202/10/world/world7.html
>  US defiant on war crimes court
>    http://www.cnn.com/2002/WORLD/europe/07/01/bosnia.peacekeeping/?related
>  US stands defiant despite isolation in climate debate
>    http://www.guardian.co.uk/bush/story/0,7369,526607,00.html
>  US defiant and unrepentant over largest chemical warfare campaign in history
>    http://www.guardian.co.uk/weekend/story/0,3605,923715,00.html
>  Americans defiant over growing divisions with Europe
>    http://www.telegraph.co.uk/news/main.jhtml?xml=/news/2002/02/06/wter106.xml
>  US defiant over missile defense
> 
>http://www.globalsecurity.org/space/library/news/2001/space-010717-wwwh1j17.htm
>  Bush defiant over 'Bin Laden tape'
>    http://news.bbc.co.uk/1/hi/world/middle_east/3204612.stm
>  Bush defiant ahead of visit to Britain
>    http://fpeng.peopledaily.com.cn/200107/18/eng20010718_75285.html
>  'Bring them on,' defiant Bush warns Iraqi militants
>    http://seattletimes.nwsource.com/html/nationworld/135155984_iraq03.html
>  Greenhouse gas emissions soar in defiant US
>    http://observer.guardian.co.uk/international/story/0,6903,515198,00.html
>  Europe seethes as defiant US goes its own way
>    http://www.guardian.co.uk/bush/story/0,7369,747743,00.html
>  Defiant US threatens UNs peacekeeping role
>    http://www.abc.net.au/7.30/s596992.htm
>  US isolated but defiant on Cuba sanctions
>    http://www.sunsonline.org/trade/process/followup/1996/04160196.htm


-- 
---------------------------------------------------------------------
IHMC	(850)434 8903 or (650)494 3973   home
40 South Alcaniz St.	(850)202 4416   office
Pensacola			(850)202 4440   fax
FL 32501			(850)291 0667    cell
phayes at ihmc.us       http://www.ihmc.us/users/phayes



More information about the SCL mailing list