[Including to the devel list again.] On 08/07/11 14:30, Stephan Gambke wrote: > Hi James, > >> Semantic Glossary (Version 0.1 alpha) (r91693) + Lingo >> >> Stephan somehow the Glossary stopped working after updating to all new >> SMW, SF ,SRF SVN versions, maybe if you have time you might verify >> that everything is still working, I don't hope that the last SMW >> changes caused to influence the SG functionality. >> >> In our case the properties for the SG are defined and can be displayed >> in Special:Browse but highlighting and display on the page itself >> stopped working and suddenly for some of the terms that worked before >> now Special:Browse shows "Has improper value for Glossary-Term" (Those >> terms did not change from the last testing, those were just fine and >> working.) > > well, r91449 implemented some changes that I do not really like. Basically > semantic objects not connected to a real wikipage are not supported anymore. > <snip>
I do not know about the detailed implementation of this extension, but some clarification is needed here. The situation is actually the opposite of what is written above: the recent changes are in fact *introducing* the creation of non-page subobjects as a *new feature*. Objects that are no pages until now have never been supported by SMW. Some code that is specific to the SQL-based store implementation for Type:Record allowed such objects to some extent, but this has never been documented as an intentional feature and there have been problems from the start (e.g. refreshing pages used to delete such objects). The new feature for the first time provides a solid basis for implementing extensions that require internal objects. If the current system is still not enough to support all requirements of extensions, then these requirements should be spelled out so that they can be taken into account in future development. On the other hand, I do not think that it is viable to have internal objects that belong to *no page at all*. This would imply that the data about these objects is not stored on any wiki page; so the data would not be versioned/archived/assigned to users/protected against vandalism/... SMW is built on the assumption that it is only a secondary storage for information written in the wiki (this is also why outdated data entries can be dropped without a warning). To allow page-independent data, one would have to implement versioning, user logging, etc. on semantic data. A practical way to overcome this would be to store the data on wiki pages (maybe hidden, maybe in a special namespace) from which it is reliably pushed into SMW on updates. This seems to be the only way to get the necessary history-keeping, and at the same time it would allow using SMW's new subobject management. Regards, Markus ------------------------------------------------------------------------------ All of the data generated in your IT infrastructure is seriously valuable. Why? It contains a definitive record of application performance, security threats, fraudulent activity, and more. Splunk takes this data and makes sense of it. IT sense. And common sense. http://p.sf.net/sfu/splunk-d2d-c2 _______________________________________________ Semediawiki-devel mailing list Semediawiki-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/semediawiki-devel
