- From: <noah_mendelsohn@us.ibm.com>
- Date: Wed, 14 Apr 2004 17:43:14 -0400
- To: jcowan@reutershealth.com
- Cc: www-xml-schema-comments@w3.org, MFC@uk.ibm.com
Well, it's worth noting that the proposed type is in fact intended to support the emerging IEEE standard (IEEE 854 ... I don't have official links but unofficial text is at [1]). This is also essentially the same type supported by java.math.BigDecimal [2]. While there's some discussion ongoing regarding the nuances, it is also either very close to or an exact match to .Net system.Decimal [3]. All of these distinguish "4" from "4.0" and in turn from "4.00". (Specifically these are {value=4, scale=0}, {value=40, scale=1} and {value=400, scale=2} respectively.) IBM is supporting IEEE decimal in a range of software products (not just Java) as well as in some hardware implementations, and we anticipate that other vendors will do the same. In short, we expect that, like IEEE 754, this type will have great significance as an industry standard. Indeed, we believe that XML Schema made a mistake in adopting the original xsd:decimal and that there will be significant benefits to supporting the IEEE standard type in XML Schema. FWIW: I suggest that any futher discussion be moved to the Schema IG list, as we normally use the comments list just to gather and frame issues. Noah [1] http://www.cs.berkeley.edu/~ejr/projects/754/private/drafts/854-1987/854-1987.pdf [2] http://java.sun.com/j2se/1.4.2/docs/api/index.html [3] http://msdn.microsoft.com/library/default.asp?url=/library/en-us/cpref/html/frlrfsystemdecimalclasstopic.asp -------------------------------------- Noah Mendelsohn IBM Corporation One Rogers Street Cambridge, MA 02142 1-617-693-4036 -------------------------------------- jcowan@reutershealth.com Sent by: www-xml-schema-comments-request@w3.org 04/14/2004 05:13 PM To: www-xml-schema-comments@w3.org cc: (bcc: Noah Mendelsohn/Cambridge/IBM) Subject: [ht@inf.ed.ac.uk: Re: [xml-dev] RE: Schema vs Schema-free] On xml-dev, Henry S. Thompson wrote: > The W3C XML Schema Working Group is considering a proposal to add a > datatype to the next version of W3C XML Schema which would distinguish > e.g. 3.0 from 3.00 _as values_. Reasoned statements of support or > disagreement would be very welcome on www-xml-schema-comments@w3.org I oppose this. Mathematically, the numerals "3.0" and "3.00" are the same thing, representations of the number 3. People who make distinctions between them are really trying to use using a single numeral to express an *interval* value, most typically [2.95, 3.05) versus [2.995, 3.005). (The [ marks an inclusive interval bound, the ) marks an exclusive one.) What is worse, this convention works only for interval sizes less than 1. There is no telling, given the numeral "1000", whether the intended interval size is [500, 1450), [950, 1050), [995, 1005), or [999.5, 100.5). Rather than supporting this hack, it would be more appropriate to add proper support for intervals. If people can convert their date data to ISO 8601, they can convert their "significant figures" numbers to proper interval values with a lower bound (inclusive or exclusive) and an upper bound (inclusive or exclusive). -- John Cowan www.ccil.org/~cowan jcowan@reutershealth.com www.reutershealth.com Monday we watch-a Firefly's house, but he no come out. He wasn't home. Tuesday we go to the ball game, but he fool us. He no show up. Wednesday he go to the ball game, and we fool him. We no show up. Thursday was a double-header. Nobody show up. Friday it rained all day. There was no ball game, so we stayed home and we listened to it on-a the radio. --Chicolini
Received on Wednesday, 14 April 2004 17:44:52 UTC