- W2K and Chandler 0.4
- Notes from Oct, 2004
- V 0.3
- What is Chandler?
- Why I care about Chandler
- Keeping up with what's happening with Chandler
- Setting up Chandler/Getting started
- Curent thoughts about the Chandler UI
- Other people working actively with Chandler
- Parcels
- Chandler data model
- List o' links
W2K and Chandler 0.4
Chandler 0.4 (![[WWW]](/www/moin/htdocs/classic/img/moin-www.png){kind=small}
Chandler Download 0.4) doesn't work on W2K?: Re: Re: Chandler 0.4 Could not run on Win 2000
Notes from Oct, 2004
V 0.3
The ChandlerProject is extremely promising as a foundation for the ScholarsBox. It's at version 0.3 right now (just released). Exciting and timely stuff:
-
0.3 is the last of our architecture-focused releases as described in our
ProductRoadmap. It is not intended for end-users but rather targets developers who want an early preview into our architecture as we are developing it. Our architecture will continue to evolve, but starting with 0.3, we are finally able to start developing end-user features. The two biggest architecture advancements in this release are the Chandler Presentation and Interaction Architecture (CPIA) and the Repository. This release marks the debut of CPIA, which is a UI layer in our architecture that is uniquely adapted for item-centric applications based on our repository. Not only does it abstract away implementation-specific UI widgets, but CPIA elements have direct access to our Repository via data-driven queries. In 0.3, our Repository now implements a transaction and threading model, and is a lot more robust and scalable.
I have written previous notes about the ChandlerProject -- but it was geared to version 0.2, which I believe is substantially different from 0.3 in many ways.
What is Chandler?
From Chandler Vision statement:
-
Our product (code-named "Chandler" after the great detective novelist Raymond Chandler,) is a Personal Information Manager (PIM) intended for use in everyday information and communication tasks, such as composing and reading email, managing an appointment calendar and keeping a contact list. Because of the ease with which Chandler users can share information with others, Chandler might be called the first Interpersonal Information Manager. (The term PIM was first used in conjunction with the product Lotus Agenda in the 1980's. Chandler is the spiritual descendant of Agenda (and has a common designer in Mitch Kapor.)
MIT Tech Review (Nov 2003) has a long overview of the Chandler Project.
Why I care about Chandler
I've long been interested in this OpenSource PersonalInformationManager for two reasons:
-
Although EccoPro (my favorite program and the center of my computing world) has served me and its community of loyal users for a long time, it's time to find/build a worthy successor that takes into account the world of computing post-1997 (the year the music ended for EccoPro)
-
Chandler seems to the best foundation I've seen upon which to build the ScholarsBox
So far, I've been following the project on a high level. With the release of version 0.2 of Chandler, I've decided that this is a good time to get my hands dirty and feet wet -- and start building (even in a preliminary fashion) something to work with Chandler. Maybe it's still too early -- but I'll see.
Here I plan to document what I learn and find.
Keeping up with what's happening with Chandler
OSAF weekly Status Manager Reports provides detailed reports on the latest, based on OSAF's internal categories. There are also more high-level updates.
The Chandler mailing lists are also a prime vehicle, such as the dev list.
http://wiki.osafoundation.org/twiki/bin/view/Main/WebHome is the Chandler wiki -- lots of action there. The RSS feed works. ![[Chandler]](/www/moin/htdocs/classic/img/moin-inter.png){kind=small}
WebChanges seems to be the list of most recent changes.
Setting up Chandler/Getting started
One of the first choices I am making after reading GettingChandler is which version of Chandler to download. I'm not currently set up to download the entire distribution and build it myself -- though I might want to go in that direction. Let me try the Developer distribution.
Curent thoughts about the Chandler UI
Other people working actively with Chandler
I have some vague memory that TomHoffman has been working with Chandler.
Thanks to TomHoffman for responding on 2003-10-17 20:50:22
-
Right now, I'm working with several of the building blocks of Chandler and loosely tracking its development without spending much time with Chandler itself. That is, I am makng extensive use of Python, wxWindows and RDF. At this point, I'm dealing much more directly with RDF than one probably will in developing Chandler apps--which on the whole is probably a good thing. When I first took a crack at Chandler parcels, immediately after 0.1 came out, I had never done any GUI programming at all, so it was a bit tough. I think I will be easier when I come back to it, now that I've got some wxPython experience under my belt.
I need to look at the mailing lists to figure out who is working on what. Who is in the Chandler community? There might even be others at UcBerkeley working on it.
Parcels
/ChandlerParcels are the extension mechanism for Chandler.
Chandler data model
I read DataModelIntro to figure out the Chandler data model. I find the ChandlerGlossary helpful to get me oriented to the vocabulary of the ChandlerProject.
What is the data model? (DataModelIntro#What_is_a_data_model_):
-
Some people use the term data model as a synonym for schema. When we talk about our data model at OSAF, we usually mean something slightly different. We're not always consistent, but we generally use the term data model to describe the set of building blocks that schema authors can use to describe schemas. My rough understanding. There are Items and Items can be of a given Type (multiple Type?) which means that there are certain standard Attributes
More about what this data model does:
-
The Chandler data model describes Chandler's basic notions about how data is represented, in terms of items and attributes, and kinds and types, and small set of other building blocks.
The essay introduces some desires on behalf of the project:
-
Users will be able to define their own classes of data, for keeping track of anything they want, from favorite movies to class assignments to bug lists.
We want Chandler to be designed for storing both structured data and unstructured data. A Chandler user should be able to enter info in an unstructured way and gradually add structure later, if and when they feel like it.
And even if an item is an item of some kind, that doesn't mean that the item is limited to only having the attributes typically associated with items of that kind.
Maybe the best way forward is to look at the current code base: Sept 2003 status of data model.
Noteworthy items
-
There is the Chandler PIM Schema that comes built with Chandler.
-
Key terms: Item, Attribute, Type, Kind -- I'm trying to figure out the relationships among these terms
MitchKapor's What's New About Chandler is interesting for many commentators said that they are using the same Item/Attributes data model in their own products: DannyAyers, SnipSnap, etc. -- so the question is what is truly distinctive about Chandler is not fully answered in the post.
Question: Will we be able to relate the Kind(s) of Item(s) in Chandler to "kinds" in formal DataFormatRegistries?
List o' links
Meaty documents:
