Re: [Tracker] Need guidance
- From: Philip Van Hoof <philip codeminded be>
- To: Pallavi Jha <pallavikumarijha gmail com>
- Cc: tracker-list gnome org
- Subject: Re: [Tracker] Need guidance
- Date: Wed, 23 May 2012 16:07:03 +0200
On Fri, 2012-04-20 at 19:32 +0530, Pallavi Jha wrote:
Hi Pallavi Jha,
I am doing documentation on documents-tracer.page. I would like to
know what all should I include while describing about "How to turn on
tracker" so that both non technical and technical user can get best
out of it?
Great!
In my opinion is Tracker, much like D-Bus, a middleware component to be
used mostly by application developers. Non-technical users benefit from
that by means of their applications being integrated, sharing relevant
data with each other, not having to reinvent content & metadata storage
themselves, etcetera.
Other than some of its included tools like tracker-needle developed by
Martyn will a non-technical user have limited use for Tracker
Just searching for files isn't the only thing Tracker can do. More
importantly, especially for application developers, is documentation on
SPARQL + our extensions, SPARQL INSERT + our extensions, GraphUpdated
signal for signal-on-change notifications, the Nepomuk ontology,
libtracker-sparql and QSparql.
Also important is general documentation on RDF and what it is, how it
works and how other bodies are using it for storing, among others,
social information in graphs. It's integrating with that that ought to
be a future goal for any future development with, for and on Tracker in
my opinion.
<subjective opinion>Sadly are most people heavily focused on just the
end-user or non-technical user's use-cases for Tracker, ie. when
criticizing that Tracker's file mining or indexing is using a lot of CPU
(recent criticism on FB by somebody, I forgot who ..) and with that
losing track of what the actual goal and/or purpose of a local
RDF/SPARQL endpoint is or would be (integration with, among others,
social media, for example).</subjective opinion>
Kinda like D-Bus ... but then instead of process communication, graph
communication (or whatever you'd call it) between graph sources.
Anyway .. (I know it's a hard sell)
Kind regards,
Philip
--
Philip Van Hoof
Software developer
Codeminded BVBA - http://codeminded.be
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]