Another SIOC browser
Hi all,
Here's a first version of my SIOC browser / explorer:
The idea, as discussed with Uldis when I was in Galway, is:
- Crawl different SIOC blogs
- Put data in a 3store
- Query this data w/SPARQL over HTTP
- Get some clean interfaces to view results
The browser is available here: http://apassant.net/hack/2006/06/sioc-browser/tags.php
I'll explain the concepts and the implementation later as I'd like to improve / redesign some parts of the code first - will do that next week -, but you still can try it now. (In a few words, it uses JSON / AJAX / PHP)
Unfortunately, Joseki - the 3store used at the moment - crashes sometimes (seems to be a MySQL jdbc driver problem), so I hope you can have a look at it before it's down :)
Cheers,
Alex.
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "SIOC-Dev" group.
To post to this group, send email to sioc-dev@googlegroups.com
To unsubscribe from this group, send email to sioc-dev-unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sioc-dev
-~----------~----~----~----~------~----~------~--~---
- Login to post comments
This resource is supported by Science Foundation Ireland under grant number SFI/02/CE1/I131 (Líon) and SFI/08/CE/I1380 (Líon 2) at the Digital Enterprise Research Institute, National University of Ireland, Galway.
Syndicate
Another SIOC browser
Hi all,
Here's a first version of my SIOC browser / explorer:
The idea, as discussed with Uldis when I was in Galway, is:
- Crawl different SIOC blogs
- Put data in a 3store
- Query this data w/SPARQL over HTTP
- Get some clean interfaces to view results
The browser is available here: http://apassant.net/hack/2006/06/sioc-browser/tags.php
I'll explain the concepts and the implementation later as I'd like to improve / redesign some parts of the code first - will do that next week -, but you still can try it now. (In a few words, it uses JSON / AJAX / PHP)
Unfortunately, Joseki - the 3store used at the moment - crashes sometimes (seems to be a MySQL jdbc driver problem), so I hope you can have a look at it before it's down :)
Cheers,
Alex.
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "SIOC-Dev" group.
To post to this group, send email to sioc-dev@googlegroups.com
To unsubscribe from this group, send email to sioc-dev-unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sioc-dev
-~----------~----~----~----~------~----~------~--~---
Another SIOC browser
Alex, there is a acknowledged and CVS-fixed problem with MySQL
disconnection Joseki after 8 hourse (default), even with ?
autoreconnect=true in the jdbc url.
The fix made by AndyS tries to reconnect Joseki once if a query failed.
CaptSolo seems to have another problem with joseki...
Christoph
Am 15.06.2006 um 00:29 schrieb Alexandre Passant:
> Unfortunately, Joseki - the 3store used at the moment - crashes
> sometimes (seems to be a MySQL jdbc driver problem), so I hope you
> can have a look at it before it's down :)
--
Christoph Görn
http://B4mad.Net/FOAF/goern.rdf#goern
Usability schmusability... where's the part where we talk about how
this helps users kick ass?
Another SIOC browser
> Here's a first version of my SIOC browser / explorer:
>
> The idea, as discussed with Uldis when I was in Galway, is:
> - Crawl different SIOC blogs
> - Put data in a 3store
> - Query this data w/SPARQL over HTTP
> - Get some clean interfaces to view results
>
> The browser is available here:
> http://apassant.net/hack/2006/06/sioc-browser/tags.php
Amazing! It looks great! :)
[GNU] started with the first AJAX SIOC browser and this now is another
step forward.
Seeing that you and your viewers also use my "on-the-fly" viewer to
see individual posts - please give me some feedback what needs to be
improved.
Do you have any ideas how to implement the 2nd part we talked about -
graphical link / relation browser?
> Unfortunately, Joseki - the 3store used at the moment - crashes sometimes
> (seems to be a MySQL jdbc driver problem), so I hope you can have a look at
> it before it's down :)
Hmm. Are you still experiencing problems? My store (also Joseki, as
you know) seems to be more stable now. Maybe what we can do is have 2
(or more) stores with mirrored content - and build into your
application switching from one store to another in case if the first
is down.
The store's SPARQL endpoint is at http://sparql.captsolo.net/joseki/query.html.
Only thing needed is to load it with new data - the data now in are quite old.
Keep up the good work! :)
Uldis
[ http://captsolo.net/info/ ]
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "SIOC-Dev" group.
To post to this group, send email to sioc-dev@googlegroups.com
To unsubscribe from this group, send email to sioc-dev-unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sioc-dev
-~----------~----~----~----~------~----~------~--~---
Another SIOC browser
Hi,
Thanks :)
Indeed, It's the first time I play with AJAX / SPARQL and it's really powerful.
Not yet. We've got a flah graph tool here in my company, I'll see if we can release it so that it could be used in SIOC Browser.
But I've found a nice API if wa want to make graphs about the data:
http://www.maani.us/charts/index.php
One of the feature I'll setup up is allowing users to choose the store they wan to query. So indeed, we could have mirrored 3stores. (or even if they're not mirrored, we can use different SIOC stores, eg: one for personal weblogs, one for irishblogs, and even one for the whole data
Regarding the crash problems, I'll check CVS as Christoph mentionned in of his reply.
Yes, we also have to find a way to update the data. But since we can do that, I'll try to setup a weekly tarball or SIOC data on my server, so that we can update the stores without crawling the data each other.
Alex
Another SIOC browser
This is cool!! From a visual point of view what I really want to do.
Even on Safari (MacOSX it produced good (not perfect jet) visual
results)
Feels like the time is right for a first consolidation of work isnt it?
Christoph
Am 15.06.2006 um 00:48 schrieb CaptSolo:
>> The browser is available here:
>> http://apassant.net/hack/2006/06/sioc-browser/tags.php
--
blogging at: http://B4mad.Net/datenbrei
info at: http://B4mad.Net/datenbrei/index.php?author=2/foaf.rdf
gpg key: http://pgpkeys.pca.dfn.de:11371/pks/lookup?
op=get&search=0xB10DFF8D88FD746C
x509 root ca certificate: http://b4mad.net/CA/
Another SIOC browser
Yep !
I didn't test it on Safari yet, but I experienced some stylesheet problems on IE.
I planned to release its code anyway, so maybe I can put it on gna.org ot other management system so that we can work together on it.
Oh, and BTW, I made a mistake in my first post, the browser base URL is http://apassant.net/hack/2006/06/sioc-browser
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "SIOC-Dev" group.
To post to this group, send email to sioc-dev@googlegroups.com
To unsubscribe from this group, send email to sioc-dev-unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sioc-dev
-~----------~----~----~----~------~----~------~--~---
Another SIOC browser
Hi Guys!
> > The browser is available here:
> > http://apassant.net/hack/2006/06/sioc-browser/tags.php
Alex: Wow really good work! I really like it.
When using it I found that I make a mistake (not really a mistake but
I forgot something in my implementation): all the Talk Digger "
conversations" are tagged by users when they start tracking them. So,
why don't I use the most common tags and use them as sioc:Topics?
Dunno, but I'll fix that "problem" over the week end.
That way, your SIOC explorer could be able to index them and
aggregate them in the "topic-browser" section.
I also thought about something else. If you wish, we could think
about implementing some type of "pinging" system between your SIOC
browser and Talk Digger. That way, each time a new "conversation"
start to be tracked by Talk Digger, it could ping your server and
then it could crawl Talk Digger to index the new content.
It could be interesting because you could prove the concept of cross-
community interaction of the SIOC ontology.
There is good XML-RPC server/client vailable for PHP, so it would not
be hard and time expensive to develop such a pinging system.
> Amazing! It looks great! :)
Uldis, just a little suggestion for your SIOC browser: would it be
possible to add a link to the raw RDF data from your interface? Or it
would only clutter it?
Keep up the good work guys!
Salutations,
--
Frederick Giasson
www.fgiasson.com
www.talkdigger.com
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "SIOC-Dev" group.
To post to this group, send email to sioc-dev@googlegroups.com
To unsubscribe from this group, send email to sioc-dev-unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sioc-dev
-~----------~----~----~----~------~----~------~--~---
Another SIOC browser
Indeed, I wish we could have a way to easilly update the stores for all the sioc:Site it contains.
The first idea I got was to use RSS feed of each blog (comments + posts) to see what have changed, and then to crawl the new data.
Pb: we won't get information about new users, we must be sure we won't lost any information (crawl every 30 minutes ?), and how can we deal with removed posts/comments ?
Another idea could be, as you also suggested, to implement a ping feature in any SIOC site, with params as "NEW", "DEL", "UPDATE", followed by the URI of the post/comment/user... to consider.
Then the store/browser could ping different services - if the service ping the browser, he has to know every store that use its data (?) - to know what's new since last crawl.
But it's a bit more tricky regarding the implementation,especially for people who want to write export plugins actually, as they'll also have to write some ping parameters.
What do you think about this ?
Alex.
Another SIOC browser
Hi Alex,
> The first idea I got was to use RSS feed of each blog (comments +
posts) to
> see what have changed, and then to crawl the new data.
> Pb: we won't get information about new users, we must be sure we won
't lost
> any information (crawl every 30 minutes ?), and how can we deal
with removed
> posts/comments ?
Not sure that it is a good idea. First the SIOC content publisher
will have to create/manage such feeds (we have to keep it as simple
as possible for adoption) and as you said: will you have to crawl it
every 30 minutes to make sure that you don't loose anything?
> Another idea could be, as you also suggested, to implement a ping
feature in
> any SIOC site, with params as "NEW", "DEL", "UPDATE", followed by
the URI of
> the post/comment/user... to consider.
> Then the store/browser could ping different services - if the
service ping
> the browser, he has to know every store that use its data (?) - to
know
> what's new since last crawl.
> But it's a bit more tricky regarding the implementation,especially
for
> people who want to write export plugins actually, as they'll also
have to
> write some ping parameters.
>
> What do you think about this ?
Right now I was thinking about a way to make Talk Digger (and
possibly a couple of other websites) communicating with your SIOC
browser. Considering that, we have much more flexibility to implement
a working solution.
What I propose is that we agree on a method to communicate and that
we implement it. I think we have to do that so we will be able to
find out what could and couldn't work in the *real* world (in a world
of large adoption).
What I was proposing is the *Technorati way*:
1- Something changed on Talk Digger (update, delete, creation), so I
Ping (XML-RPC) you SIOC browser. It tell him: He, something changed
at that URL, could you take a look and update your databases
accordingly?
2- Your browser receive the ping, check the URL and crawl it.
3- It change its story accordingly to the new information.
4- each week or so (everything depends on your available bandwidth),
you could re-crawl the pings URL you received to check if any changes
have been made. What we could also do is to create a simple pinging
protocol, as you suggested, to let me tell you if something have been
added, deleted or updated.
You can check my blog (my last posts) to see some simple SIOC
document created by Talk Digger.
Thanks,
Salutations,
Fred
--
Frederick Giasson
www.fgiasson.com
www.talkdigger.com
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "SIOC-Dev" group.
To post to this group, send email to sioc-dev@googlegroups.com
To unsubscribe from this group, send email to sioc-dev-unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sioc-dev
-~----------~----~----~----~------~----~------~--~---