This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
XML Schema 1.0 requires all implementations to support at least 18 digits in the decimal type. This has two problems: * If an implementation strictly implements a 18 digit decimal facility, it cannot accomodate long which requires 20 digits. * 18 decimal digits cannot be accommodated in 64 bits. In April 2004 the WG decided to resolve this thus: * Note that the existence of types with infinite value spaces commits us already to allowing implementations which support some but not all values; the minimum implementation limits we specify are intended to provide some guidance in the matter. So it's legal to have only a partial implementation of long. (Whether it's smart is a different question.) * In 1.1, the minimum implementation requirement is to be 16 digits. Wording to this effect was adopted in January 2005, but RQ-150b has remained open because some concern was expressed over whether the words adopted in January 2005 were correct.
In email to the IG on 23 August, Dave Peterson said that his concerns with RQ-150b have been satisfied and that in his view the issue has been resolved and can be closed. http://lists.w3.org/Archives/Member/w3c-xml-schema-ig/2005Aug/0062.html The WG should either agree with this proposal, or decline to do so and give further instructions to the editor.
I had thought that sometime during September we had finally agreed to close the issue, but there is no trace of it in the minutes from September or October, at least none that can be found by searching for the strings "1918", "150", and "http://lists.w3.org/Archives/Member/w3c-xml-schema-ig/2005Aug/0062.html" So I reluctantly conclude either that we didn't actually agree to close it, or that whoever was scribing the meeting failed to record the decision. Either way, this issue cannot be marked closed. So I'm marking it needsReview. If anyone can find a written record of our having disposed of this item, I'd be grateful for a pointer.
We completed the resolution at the Toronto f2f 2005-11. The resolution was actually decided in Edinburg but not so recorded.
Although no formal request for closure was made, since the reporter also noted the resolution of this bug over two years ago, I'm marking it closed.