← Prev in month
← Prev in thread
Next in thread →
Next in month →
Minutes: XDI TC Telecon Friday 2014-11-21
XDI TC Minutes Following are the minutes of the unofficial telecon of the XDI TC at: Date: Friday, 21 November 2014 USATime: 09:00AM - 10:30AM Pacific Time (16:00-17:30 UTC) ATTENDING Amanda NavarroPeter DavisHubert Le Van GongLes ChasenDrummond ReedMarkus SabadelloPhil WindleyWIlliam DysonJoseph Boyle GUESTSREGRETSNEWS & UPDATESPRESENTATIONS/DISCUSSIONSReport from XDI editors subcommitteehttps://wiki.oasis-open.org/xdi/XdiOneSpecshttps://github.com/OASIS-XDI-Technical-Committee/xdi-spec-docbook The XDI editors call was canceled this week. ABNF progress Joseph has created a more complete version of the XDI ABNF that validates successfully in the APG parser: https://lists.oasis-open.org/archives/xdi/201411/msg00020.html https://wiki.oasis-open.org/xdi/XdiAbnf http://www.coasttocoastresearch.com/interactiveapg Markus has added an experimental ABNF-based parser to the XDI2 code: https://server.xdi2.org/no-value-node/XDIParser https://github.com/projectdanube/xdi2/tree/no-value-node Markus proposed a few small changes in the message thread linked above. Drummond reviewed these changes and agreed with them. We discussed the use of rules to represent delimiters, which is the approach that Joseph has taken. Markus explained that the organization of the ABNF rules is not just for expressing the rules, but also for giving programmatic access to the parser tree. In that light, Markus prefers breaking out the delimiters into their own rules. For high-performance implementations, a code base will most likely not use a formal ABNF parser anyway. So having a deeply parsed tree is not an overall implementation issue. In that case, Drummond is fine with proceeding to represent each delimiter as its own rule. #ACTION ITEM: Joseph to review Markus’ proposal revisions and then post a proposed final ABNF to the TC wiki page. Glossary progress Drummond has worked on terms for the XDI Core spec: https://wiki.oasis-open.org/xdi/Glossary https://lists.oasis-open.org/archives/xdi/201411/msg00019.html Markus has worked a bit more on glossary terms for the XDI Messaging spec, incorporating feedback from Hubert and others to use the terms “synchronous response” and “asynchronous response”. Peer roots and inner roots While going over Drummond’s new glossary terms for XDI Core, we had a discussion about peer roots and inner roots. Drummond is working on a paper about root nodes. Hubert asked what the following statement meant: (=robertson)(=charlie/=alice)=alice<#age>/&/25 Drummond explained that the contents of different peer roots must be logically consistent, whereas the contents of inner roots are specific only to the inner root and may be different in other inner roots. A peer root is a common root in its own context. Markus asked how this logical consistency of peer roots can be enforced across different XDI endpoints in a network. Drummond answered this can’t really be enforced technically, but it serves as a model which a community may choose to implement. Markus added that Srini of Neustar has contributed a chapter to the XDI Tutorials that explains peer roots and inner roots: http://tutorial.xdi.org/tutorial.html?3|8 Markus suggested that TC members take a close look at this tutorial to see if it makes sense and to find potential mistakes. XDI Messaging and XDI Bindings contributions Markus has worked on new content for the specs: XDI Messaging: Definitions of $get, $set, $del, $add, $mod operations XDI Binding: XDI Browser Binding See latest versions on “markussabadello” branch: http://xdi.org/xdi-spec-docbook-markussabadello/xdi/xdi-bindings-1.0/xdi-bindings-1.0-wd02.xml http://xdi.org/xdi-spec-docbook-markussabadello/xdi/xdi-messaging-1.0/xdi-messaging-1.0-wd02.xml See pull request: https://github.com/OASIS-XDI-Technical-Committee/xdi-spec-docbook/pull/10 #ACTION ITEM: Everyone to review Markus’ proposed content and pull request. XDI Policy and Connections We did not further discuss this topic. Transactional integrity We did not further discuss this topic. NEXT CALLThe next call is next week at the regular time, however we expect this to be a "very light" call without major decisions, because of U.S. Thanksgiving holiday on the previous day
← Prev in month
← Prev in thread
Next in thread →
Next in month →