I agree that it is probably most important to use an API like JCR instead accessing the file system directly.
About the implementation you might want to consider http://jackrabbit.apache.org/oak/docs/differences.html I am using myself https://github.com/wyona/yarep which I would consider a more lightweight API than JCR and implementation accordingly. HTH Michael Am 06.11.14 um 10:41 schrieb Thomas Mueller: > Hi, > > I would probably use the JCR API as much as possible. That way you can > more easily switch between implementations (and that includes not just > Jackrabbit JCR implementations). Some details: > > * I would not use same name siblings > > * I would not use many (more than about 1000) child nodes in a node (flat > hierarchies). > > * I would avoid using node types too much. > > > * As for queries, I would probably use XPath as it is simpler in many > cases. > > > As for Jackrabbit 2.x versus Jackrabbit Oak: > > * Jackrabbit 2.x is there since a longer time, has more documentation. > > * Jackrabbit Oak is relatively new, and probably still a bit "rough" > (documentation is not that clear for examle). Things are still changing > quite quickly. But eventually, it is or will be more flexible and scalable > (support larger repositories for example, more storage backends, more > indexes). > > Regards, > Thomas > > > > On 05/11/14 15:14, "Herrick, Rick" <herri...@mir.wustl.edu> wrote: > >> We're in the process of working on an archive management system for our >> medical imaging data platform (XNAT, http://www.xnat.org). Currently we >> just manage files on the hosting file system, with all the issues that >> implies. We've been considering using Jackrabbit to manage all of the >> data resources (MRI, CT, PET and similar imaging data, synthetic data > >from processing and analysis pipelines, research subject data, etc.), but >> we have a few concerns. >> >> There doesn't seem to have been too much activity on this list, most of >> the articles on the Jackrabbit articles page are from 2011 and earlier, >> and most of the Jackrabbit news is actually about Oak. >> >> So is Jackrabbit still an on-going and supported platform? Should we be >> looking at Oak instead? Basically we don't want to embark on a full-blown >> development effort on something that may not be maintained. Or is just >> that, because this is a back-end technology, there's just not that much >> traffic and that's actually a GOOD thing (i.e. it's basically done and it >> works and no one complains)? >> >> Any thoughts on this would be very helpful and greatly appreciated. >> Thanks! >> >> -- >> Rick Herrick >> Sr. Programmer/Analyst >> Neuroinformatics Research Group >> Washington University School of Medicine >> (314) 740-5961 >> >> ________________________________ >> >> The material in this message is private and may contain Protected >> Healthcare Information (PHI). If you are not the intended recipient, be >> advised that any unauthorized use, disclosure, copying or the taking of >> any action in reliance on the contents of this information is strictly >> prohibited. If you have received this email in error, please immediately >> notify the sender via telephone or return mail.
