Revision as of 06:20, 19 March 2006

GeoTools is one of the round of projects out of the gate, and as an rather old project (lots of background) with an active development community (who is busy hacking) does not expect to finish all that quickly.

Representative: Jody Garnett (jgarnett@refractions.net)

Initial Impressions

Judging from the email list the number one question is, what needs to be done? Rather wait we are going to wing it and hope for futher direction.

Community Feedback

The GeoTools community responded with the following questions:

rgould: What if I want to commit a patch or other code sent to me by someone else? Do they have to sign a CLA too?
- three line fix is okay
- three line w/ new functionality, or large change, requires a CLA
- If you can get a CLA
bryce: Public domain contributions (ie gov employees)
- individual code: can sign w/ no conflict as themselves
- gov code: already public domain - osgeo does not need a license
- treat on a case by case basis

What is in a Name

GeoTools is a common name, that many conferences, other open source projects and so on try to adopt. Yes we know it is obvious. Since the project has been going for 10 years (congrats) in a very public manner with out complaint we are going to keep on keeping on. If pressed we will adopt the name "GeoTools.org" and call it a day.

IP Check

We don't know if everyone who worked on geotools is even alive, let alone how to find the ones that are still ticking. This is a problem.

"svn blame" is almost an answer:

Create a perl script that can process svn blame
Get a list of names
Module maintainers will need to hunt down those involved and beat a Contributor Agreement out of them

But it is not a complete answer:

It only shows who has done what (per line) most revently
Svn is backed on to a berkly DB, and we can ask someone (hi paul) to query the heck out of this database and give us report, by module
If this is too hard I would be content with an SVN blame check based on the 2.0, 2.1 and 2.2 releases

If a module maintainer cannot document a change for a section of code they will need to re implement. Sad but true.

Property to Check

documentation

The documentation has always had a strict license, as indicated in the developers guide, however we break it down by maintainer:

Wiki
- news anouncements - can we ignore please?
- product branding logos - jody
Developers Guide - james (1st) and jody (2nd)
Users Guide - rueben
- Tutorials - rueben
- Snipits - rueben
- Articles - jody
RnD - jody
- Feature Model - gabriel
- Converage - bryce
- Rest - jody

We have lost our early history of news to a crash (I still am sad), and we have had around 6 websites. Some of the generated websites will be in version control and will show up as part of the code IP check.

code

The code has always been under LGPL: - 1.x code code was (c) leeds - jmacgill confirm? - 2.x code had (c) GeoTools PMC at the top.

module

The module section of code is kind of lame and has a long tangled history. It started off as "core" and contained implementations for all the interfaces we had ever defined. It may even of contained the interfaces. The conversion notes are in confluence.

Conversion? Yeah we converted it - into a single module called "main" (because we could not manage our build as everything was so brittle). This was done for GeoTools 2.0, for geotools 2.2 things are breaking up along different lines:

API - Jody Garnett - Capturing the (from main) interfaces that have been through a Q&A check
MAIN - PMC - sad but the definition of PMC was almost module maintainer of main for a while
- filter/expresison - cory
- rendering - jesse
REFERENCING - martin
LEGEND - andrea (may just want it deleted)
...

plugins

Plugins are bits of functionality that implement all the great stuff that makes geotools useful. They slot in at the various layers defined by the libary and are nice and targetted:

postgis: justin
oracle: marc
...

ext =

Extentions and Functionality built on GeoTools.

graph - justin
validation - jody

demo

Demos and Examples ...

property - jody

@@ Line 24: / Line 24: @@
 = IP Check =
 We don't know if everyone who worked on geotools is even alive, let alone how to find the ones that are still ticking.
 This is a problem.
@@ Line 31: / Line 30: @@
 # Create a perl script that can process svn blame
 # Get a list of names
-# Module maintainers will need to hunt down those involved and beat a [[Contributor Agreements]] out of them
+# Module maintainers will need to hunt down those involved and beat a [[Contributor Agreement]] out of them
 But it is not a complete answer:
@@ Line 43: / Line 42: @@
 === documentation ===
+The documentation has always had a strict license, as indicated in the developers guide, however we break it down by maintainer:
-The documentation has always had a strit license, as indicated in the developers guide, however we break it down by maintainer:
 * Wiki
 ** news anouncements - can we ignore please?
@@ Line 68: / Line 66: @@
 === module ===
-The module section of code is kind of lame and has a long tangled history. It started off as "core" and contained implemtnatoins for all the interfaces we had ever defined. It may even of contained the interfaces. The conversion notes are in confluence.
+The module section of code is kind of lame and has a long tangled history. It started off as "core" and contained implementations for all the interfaces we had ever defined. It may even of contained the interfaces. The conversion notes are in confluence.
 Conversion? Yeah we converted it - into a single module called "main" (because we could not manage our build as everything was so brittle). This was done for GeoTools 2.0, for geotools 2.2 things are breaking up along different lines:
@@ Line 80: / Line 78: @@
 === plugins ===
 Plugins are bits of functionality that implement all the great stuff that makes geotools useful. They slot in at the various layers defined by the libary and are nice and targetted:
 * postgis: justin

Difference between revisions of "GeoTools Incubation Progress"