Thanks Javier for the clarifications. On Mon, Jun 8, 2015 at 5:55 PM, jfernand <[email protected]> wrote:
> Hi Junyue, > > first of all, thanks a lot for selecting and developing this project. Any > new integration with HDT is welcome! :) > > Let me briefly introduce the status of the project. So far, RDF/HDT > practical development has been mainly driven by the original co-authors, > that is, Miguel A. Martínez, Javier D. Fernández and Mario Arias. All the > HDT modifications for other use-cases, such as Streaming HDT, Waterfowl, > Linked Data Fragments, etc., have used or modified our tools, disregarding > the particular specifications of the serialization format. > > The W3C submission was accepted back in 2011, so it was designed in 2010. > Unfortunately, W3C member submissions are "static" (in contrast to other > W3C documents), i.e. the document cannot be modified. As can be imagined, > from 2010 on, we have slightly improved the format, e.g. adding CRCs, and > these changes couldn't be reflected in the proposal. We manage an internal > document with these changes. I will review the document, clarify some parts > for the public audience and release it in rdfhdt.org. I cannot quantify > the number of changes now, but it certainly shouldn't affect the complexity > of implementing the main interface, it is just a matter of being consistent > with the current version. > > In my opinion, one should be more careful in the serialization of the > Dictionary as, by default, we use a Plain Front Coding compression > technique. I will also try to sketch the format in the document. I will > contact you as soon as this is available (1-2 days), and you could help us > providing feedback if some parts are not clear. This would be useful for us > too :) > > All the best! > > > > > > --- > Javier D. Fernández > WU Vienna > Institute for Information Business > Welthandelsplatz 1, Building D2, Entrance C, 2nd Floor > 1020 Vienna, Austria > > El 2015-06-08 15:45, Junyue Wang escribió: > >> Hello, >> >> I'm working on a project [5] of creating a Sesame [1] parser for RDF >> HDT. There's a Java implementation [2], with the licence of LGPL. Does >> the implementation strictly conform to the W3C specification [3]? Are >> there any differences between [3] and [4]? Is [4] a subset of [3]? >> >> If I need to develop a parser from scratch for the licence reason, >> which specification should I follow? It's been years since the >> invention of RDF HDT. Are these specifications and implementations out >> of date? >> >> Thanks you! >> >> yours, >> junyue >> >> [1] http://rdf4j.org/ [1] >> >> [2] http://www.rdfhdt.org/manual-of-the-java-hdt-library/ [2] >> [3] >> http://www.w3.org/Submission/2011/SUBM-HDT-Implementation-20110330/ >> [3] >> [4] http://www.rdfhdt.org/hdt-internals/ [4] >> [5] https://issues.apache.org/jira/browse/MARMOTTA-593 [5] >> >> Links: >> ------ >> [1] http://rdf4j.org/ >> [2] http://www.rdfhdt.org/manual-of-the-java-hdt-library/ >> [3] http://www.w3.org/Submission/2011/SUBM-HDT-Implementation-20110330/ >> [4] http://www.rdfhdt.org/hdt-internals/ >> [5] https://issues.apache.org/jira/browse/MARMOTTA-593 >> > -- Sergio Fernández Partner Technology Manager Redlink GmbH m: +43 6602747925 e: [email protected] w: http://redlink.co
