(feature request) supporting Dublin Core ontology #5
Comments
|
PPPS: (some background) I've experienced the move from paper tickets to "eTickets", in the airline industry: lots of savings and efficiencies, plus some "new & smart" functionalities paper could not support (from reissuing a ticket for a schedule change, up to real-time gate changes notifications). I used Tomcat+ServiceMix 3.x to implement both a service to extract data from "file printouts" (so that data from the GDS mainframe won't end up printed on paper, but stored on files, renamed after the data extracted from the files themselves), and the UI for accessing those files. I noticed that ESBs (sematic-less ones!) have been progressively adopted in the travel industry, not just GDSes and airlines, but also big-enough tour operators and travel agencies. Won't do any names but I know real examples. Finally, I tried implementing some document handling prototype on uAAL, but:
I hope this helps giving some context to the "bug request" I opened. Hoping to see the dev-friendly stuff out soon! cheers |
ghost
changed the title
|
hello @corradocampisano , Any way, I am probably skipping some background you should have on universAAL before getting to this level of integration. But essentially you'll need first to think about a top-level ontology for personal documents, then some semantical services to go with it, then integration with platform services (eg: storage, database, security, etc...) and finally a UI to go with it. Our efforts ( @universAAL/core ) towards dev-friendliness currently are focused on improving the documentation (you could help if you find any wiki page difficult to understand or just plain wrong), and providing a set of tools ( @universAAL/tool-developers , namely eclipse plugins as well as run time tools to help developers). If you find anything else we need to focus our attention on for making uAAL more dev-friendly, as well as any specific proposals, we are always open to suggestions. BR, |
|
HI Gil, Notice I picked DublinCore just because one of the most used and valued, as far as I know, for handling, generically, metadata about files, which then means all types of documents, even multimedia. I'm happy you also recognize these sort of needs, and that the new "Asset" concept is already in. Regarding the docs, I found it quite ok, except for the UI, for which maybe I have different expectations. What about a "complete", but simple, sample "full app with UI", which only allows you to log in (on a web UI, possibly "hosted" within Karaf itself) an see "what time it is" (with a service providing "exact time" to the bus)? I'll wait for the coming new features (those you mentioned) and the new dev-friendly stuff (pls don't let all this work end up like Nokia N900's maemo did!) BR |
|
There is a sample UI , to test it make sure you install the uAAL-SingleNode.Development feature in your karaf distro. You will also need to maven install it (build it) and copy the jar (in target) to the deploy folder of the karaf. There is also a small edit you'll need to do to the user main menu configuration of the ui.dm so that it appears in the main menu. All these procedures used to be documented, but I don't know where it is right now (added universAAL/platform#8 to remember my self to work on it). |
|
Hi Gil, I tried installing the feature:
Logs are below, screenshots are attached. Now:
thx |
|
So universAAL UI is capable of interacting with the user using different technologies, the "default" one is through Java Swing GUI, but using the UI.handler.WEB(page needs updating) will enable you to access the web format interacction. the "search page" is the universAAL main menu, which is empty since it has not been configured yet. when configured there should be a button per application/form for the user to select, the search is just in thecase the user has many buttons. I would suggest to open another Issue regading the UI set up / help either in the platform repository or the UI repository; as this conversation is steering in another direction as the original (Ontology). |
ghost
mentioned this issue
Hello people,
how long/how hard would it be to add an "ontology.dc" subproject here, so that at least some kind of "human documents" (from pdfs to emails) could be supported by the environment?
you know, with the move to "paperless" documents, how could (today's and tomorrow's) old people handle this change?
I'm not so old yet, but I'm already worried about how to safely store "my e-stuff", from blood test reports (my latest one was down-loadable as a pdf, not on paper), to bank accounts reports, and so on.
PS: (EU related) for the fundings, I'd recommend stressing out with EU representatives the fact that "We Europeans" are actually "saving" our docs and stuff using services run by private companies in "trump-land"... Just like using windows in EU institutions, that's rather a EU security issue, indeed.
PPS: (out of scope) I won't trust private or state cloud services, for actually storing the data. I'd rather go for a "distributed storage", where anyone with a node (a small qnap NAS, for example), who wants to "safely store 1G on the net", would also allow others from the net to store (at least) 1G on his node.
See napster/emule storage, for example, or bitcoin's one. Supporting storing documents using dublin core first (indeed: files and their metadata) will help implementing storing "crypted slices" from the net storage, later.
Wish you the best
Corrado Campisano
The text was updated successfully, but these errors were encountered: