So a couple of days ago I promised a Nautilus + Zeitgeist hack… Thanks to Randy it is reality (and I am not a liar)…It is just a proof of concept, we will take time next week to look into details and make things really shiny… Read his blog for more info

Oh yeah here is the sexy stuff if you are to lazy to click and read his blog…

Ladies and Gentlehackers,

I would like to congratulate (Dr. to be) Thorsten Prante for taking up on blogging… His first blogpost, surprise surprise, is about some ideas behind Zeitgeist… Also a paper about Zeitgeist has been accepted for CHI Workshop… Please enjoy….

OK this post is from Ian Cylkowski aka izo.
please please have a look!
http://www.design-by-izo.com/2010/02/27/deconstructing-nautilus-and-rebuilding-it-better/
It shoes some nice mockups

So if someone can link me with a tutorial how to extend the nautlius sidebar I will add the “commonly accessed” functionality within an hour or 2!

PROMISE

OK here is a video…

First the Release Announcement:

GNOME Activity Journal 0.3.3 – “Patrick Star meets Lady Gaga”
==============================================

The GAJ & Zeitgeist team, are proud to announce the second development release of GNOME Activity Journal, codenamed “Patrick Star meets Lady Gaga”.

What is GNOME Activity Journal?
=========================

GNOME Activity Journal is not a File Browser but an Activity Browser. It uses the ZeitgeistFramework to display what you did and introduces a better way of finding the things that you were doing quickly.

Where?
=====

* Downloads: http://edge.launchpad.net/gnome-activity-journal/0.3/0.3.3/+download/gnome-activity-journal-0.3.3.tar.gz
* About Zeitgeist: http://zeitgeist-project.com
* Wiki: http://live.gnome.org/GnomeActivityJournal

What is done so far:
==============

Since the last release we included some new features (but postponed others)
Alongside the “Pretty layout”, “Pinning of stuff, “Calendar scrubbing” and the experimental “Tracker Search” we are introducing:

* “Detailed Day View” accessible over left clicking the day label.
* “Thumbnail View” accessible over right clicking the day label.
* Remove Items from the Journal

Also a lot of bugs have been fixed alongside caching and performance improvements.

What to expect with the next releases:
===========================

* Improve “right-click” menu (Display related files).
* Add “right-click” menu to the new views.
* Search and interaction.
* Tags.
* Remove activities from the journal (permanently).
* Fixing all open bugs .

—-
Cheers
Seif

Now for Screenshots of the new views (accessible by clicking the day label)

• Detailed View (Accessible over left click):

• Thumbnail View (Accessible over right click):

The next release will allow more interactions and make use of Zeitgeist’s “Most used With” and “A priori Sets”. Zeitgeist 0.3.3 will be out soon…

With the framework being in a stable form now and Siegfrieds work on a “dataprovider register” (he will blog about it soon i hope) I decided to take a stab at something new that has not really been used on the desktop before except in GNOME-Do.
However what we are doing is a bit more complicated.
With Zeitgeist knowing about your activities and events i tried to apply the classic “A Priori” Algorithm to detect associated files based on usage.
Unlike our current implementation of finding files related to other explicit given files. This one looks at a whole timerange and creates sets of files used together. It is kind of like detecting usage patterns.

Taken from wikipedia here is a little example:

A large supermarket tracks sales data by SKU (item), and thus is able to know what items are typically purchased together. Apriori is a moderately efficient way to build a list of frequent purchased item pairs from this data. Let the database of transactions consist of the sets {1,2,3,4}, {2,3,4}, {2,3}, {1,2,4}, {1,2,3,4}, and {2,4}. Each number corresponds to a product such as “butter” or “water”. The first step of Apriori to count up the frequencies, called the supports, of each member item separately:
Item Support
1 | 3
2 | 6
3 | 4
4 | 5
We can define a minimum support level to qualify as “frequent,” which depends on the context. For this case, let min support = 3. Therefore, all are frequent. The next step is to generate a list of all 2-pairs of the frequent items. Had any of the above items not been frequent, they wouldn’t have been included as a possible member of possible 2-item pairs. In this way, Apriori prunes the tree of all possible sets..
Item Support
{1,2} | 3
{1,3} | 2
{1,4} | 3
{2,3} | 4
{2,4} | 5
{3,4} | 3
This is counting up the occurrences of each of those pairs in the database. Since minsup=3, we don’t need to generate 3-sets involving {1,3}. This is due to the fact that since they’re not frequent, no supersets of them can possibly be frequent. Keep going:
Item Support
{1,2,4} | 3
{2,3,4} | 3

In much larger datasets, especially those with huge numbers of items present in low quantities and small numbers of items present in large quantities, the gains made by pruning the possible pairs tree like this can be very large.

Now imagine the supermarket being all files on ur computer and the transactions being sets of events happening to the files per 30 minutes.

Using this we will be able to determine alot of cool associations between files and with some Tracker magic we could do this on a metadata level.

Now i pushed the implementation with this test case and another into a branch waiting for the team to review it and expose it properly over D-Bus. The test cases work though which kinda gets me excited to work on new algorithms and generate adaptive transaction since right now i assume every 30 minutes is a transaction. I have a lot of ideas and also I will be applying two other algorithms Winepi and Minepi to compare results.

Expect some cool stuff to come up soon…
Releases of the engine and GAJ are being prepared…
Cheers

Out of personal and convenience issues i decided not to board the train today.
So I wont attend FOSDEM. However 2 Zeitgeisters will be there.
Please don’t hesitate to contact Hylke Bons (hbons) concerning the GNOME Activity Journal and Siegfried Gevatter (RainCT) concerning Zeitgeist Framework!
Have fun! And lay off the booze!

With Codethink sponsoring the prize for the upcoming to be announced competition for the coolest app using Tracker and RDF, I took the liberty to sum up the “fuss” around Tracker in 2 blog posts.

• Part I: User Stories and Problems of the current desktop.
• Part II: Tracker To The Rescue and Providing New User Experience

So here we go with Part I:

User stories:

• Tony is making some changes to a document. To clarify some issues, he wants to refer back to the email threads or IM conversations he’s had about that document.
• Jenna uploaded her vacation pictures from Brazil to her desktop. She would like to tag people in the picture with contacts in her address book. Later when she uploads the pictures to Flickr she expects to have the tags available there too.
• John has to work on his Michael Jackson research. Since he worked using webapps such as “Google docs” and some local data, he needs to be able to globally search for all his data on the web and locally.
• Rob is trying to land a contract with ACME Inc. he wants to get a list of all of his contacts who have a relationship to ACME Inc.
• Tony met a hot girl at a house party. He wants to search which of his freinds know her.
• April is working on a project. She wants to make a ‘folder’ containing all the research materials she’s using on the project. Her research materials include webpages, downloaded files, e-books, emails and IMs.

Problem:

After a some dirty hacks and little battle with dependencies. The Zeitgeist team got the engine running on Maemo…

Now for the UI and some Zeitgeist voodoo

BTW if u have any N900 to throw or donate at the Zeigeist developers side please contact us on irc.freenode.net #zeitgeist.