Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 13 additions & 0 deletions doc/source/devel/devguide.rst
Original file line number Diff line number Diff line change
Expand Up @@ -108,3 +108,16 @@ the algorithms with implications to the performance or validity of results.

It may list references to 3rd party bugtrackers, in case the reported bugs
match the criteria listed above.

.. _mission_and_values:

.. _community_guidelines:

.. _code_of_conduct:

Community guidelines
====================

Please see `our community guidelines
<https://github.com/nipy/nibabel/blob/master/.github/CODE_OF_CONDUCT.md>`_.
Other projects call these guidelines the "code of conduct".
156 changes: 156 additions & 0 deletions doc/source/devel/governance.rst
Original file line number Diff line number Diff line change
@@ -0,0 +1,156 @@
.. _governance:

==============================
Governance and Decision Making
==============================

Abstract
========

Nibabel is a consensus-based community project. Anyone with an interest in the
project can join the community, contribute to the project design, and
participate in the decision making process. This document describes how that
participation takes place, how to find consensus, and how deadlocks are
resolved.

Roles And Responsibilities
==========================

The Community
-------------

The Nibabel community consists of anyone using or working with the project
in any way.

Contributors
------------

Any community member can become a contributor by interacting directly with the
project in concrete ways, such as:

- proposing a change to the code or documentation via a GitHub pull request;
- reporting issues on our
`GitHub issues page <https://github.com/nipy/nibabel/issues>`_;
- discussing the design of the library, website, or tutorials on the
`mailing list <https://mail.python.org/mailman/listinfo/neuroimaging>`_,
or in existing issues and pull requests; or
- reviewing
`open pull requests <https://github.com/nipy/nibabel/pulls>`_,

among other possibilities. By contributing to the project, community members
can directly help to shape its future.

Contributors should read the :ref:`contributor_guide` and our :ref:`code_of_conduct`.

Core Developers
---------------

Core developers are community members that have demonstrated continued
commitment to the project through ongoing contributions. They
have shown they can be trusted to maintain Nibabel with care. Becoming a
core developer allows contributors to merge approved pull requests, cast votes
for and against merging a pull request, and be involved in deciding major
changes to the API, and thereby more easily carry on with their project related
activities. Core developers appear as team members on the `Nibabel Core Team
page <https://github.com/orgs/nipy/teams/nibabel-core-developers/members>`_ and
can be messaged ``@nipy/nibabel-core-developers``. We expect core developers to
review code contributions while adhering to the :ref:`core_dev`.

New core developers can be nominated by any existing core developer. Discussion
about new core developer nominations is one of the few activities that takes
place on the project's private management list. The decision to invite a new
core developer must be made by “lazy consensus”, meaning unanimous agreement by
all responding existing core developers. Invitation must take place at least
one week after initial nomination, to allow existing members time to voice any
objections.

.. _steering_council:

Steering Council
----------------
The Steering Council (SC) members are core developers who have additional
responsibilities to ensure the smooth running of the project. SC members are
Comment on lines +71 to +72
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
The Steering Council (SC) members are core developers who have additional
responsibilities to ensure the smooth running of the project. SC members are
The Steering Council (SC) members are current or former core developers who
have additional responsibilities to ensure the smooth running of the project.
SC members are

expected to participate in strategic planning, approve changes to the
governance model, and make decisions about funding granted to the project
itself. (Funding to community members is theirs to pursue and manage.) The
purpose of the SC is to ensure smooth progress from the big-picture
perspective. Changes that impact the full project require analysis informed by
long experience with both the project and the larger ecosystem. When the core
developer community (including the SC members) fails to reach such a consensus
in a reasonable timeframe, the SC is the entity that resolves the issue.

Steering Council members appear as team members on the `Nibabel Steering
Council Team page
<https://github.com/orgs/nipy/teams/nibabel-steering-council/members>`_ and
can be messaged ``@nipy/nibabel-steering-council``. Core

Decision Making Process
=======================

Decisions about the future of the project are made through discussion with all
members of the community. All non-sensitive project management discussion takes
place on the project
`mailing list <https://mail.python.org/mailman/listinfo/neuroimaging>`_
and the `issue tracker <https://github.com/nipy/nibabel/issues>`_.
Occasionally, sensitive discussion may occur on a private list.

Decisions should be made in accordance with our :ref:`mission_and_values`.

Nibabel uses a *consensus seeking* process for making decisions. The group
tries to find a resolution that has no open objections among core developers.
Core developers are expected to distinguish between fundamental objections to a
proposal and minor perceived flaws that they can live with, and not hold up the
decision making process for the latter. If no option can be found without
an objection, the decision is escalated to the SC, which will itself use
consensus seeking to come to a resolution. In the unlikely event that there is
still a deadlock, the proposal will move forward if it has the support of a
simple majority of the SC. Any proposal must be described by a Nibabel :ref:`biap`.

Decisions (in addition to adding core developers and SC membership as above)
are made according to the following rules:

- **Minor documentation changes**, such as typo fixes, or addition / correction
of a sentence (but no change of the Nibabel landing page or the “about”
page), require approval by a core developer *and* no disagreement or
requested changes by a core developer on the issue or pull request page (lazy
consensus). We expect core developers to give “reasonable time” to others to
give their opinion on the pull request if they’re not confident others would
agree.

- **Code changes and major documentation changes** require agreement by *one*
core developer *and* no disagreement or requested changes by a core developer
on the issue or pull-request page (lazy consensus).

- **Changes to the API principles** require a :ref:`biap` and follow the
decision-making process outlined above.

- **Changes to this governance model or our mission and values** require
a :ref:`biap` and follow the decision-making process outlined above, *unless*
there is unanimous agreement from core developers on the change.

If an objection is raised on a lazy consensus, the proposer can appeal to the
community and core developers and the change can be approved or rejected by
escalating to the SC, and if necessary, a BIAP (see below).

.. _biap:

Enhancement Proposals (BIAPs)
=============================

Any proposals for enhancements of Nibabel should be written as a formal BIAP
following the template :doc:`biap-template`. The BIAP must be made public and
discussed before any vote is taken. The discussion must be summarized by a key
advocate of the proposal in the appropriate section of the BIAP. Once this
summary is made public and after sufficient time to allow the core team to
understand it, they vote.

The workflow of a BIAP is detailed in :ref:`biap0`.

A list of all existing BIAPs is available :ref:`here <biap_list>`.

Acknowledgments
===============

Many thanks to Jarrod Millman, Dan Schult and the Scikit-Image team for the
`draft on which we based this document
<https://networkx.github.io/documentation/latest/developer/nxeps/nxep-0001.html>`_.
2 changes: 2 additions & 0 deletions doc/source/devel/index.rst
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,8 @@ Developer documentation page
:maxdepth: 2

devguide
governance
roadmap
add_test_data
add_image_format
devdiscuss
Expand Down
91 changes: 91 additions & 0 deletions doc/source/devel/roadmap.rst
Original file line number Diff line number Diff line change
@@ -0,0 +1,91 @@
#######
Roadmap
#######

The roadmap is intended for larger, fundamental changes to the project that are
likely to take months or years of developer time. Smaller-scoped items will
continue to be tracked on our issue tracker.

The scope of these improvements means that these changes may be controversial,
are likely to involve significant discussion among the core development team,
and may require the creation of one or more NIAPs (Nibabel Increased
Awesomeness Proposals).

==========
Background
==========

Nibabel is a workbench that provides a Python API for working with images in
many formats. It is also a base library for tools implementing higher level
processing.

Nibabel's success depends on:

* How easy it is to express common imaging tasks in the API.
* The range of tasks it can perform.

An expressive, broad API will increase adoption and make it easier to teach.

Expressive API
==============

Axis and tick labels
--------------------

Brain images typically have three or four axes, whose meanings depend on the
way the image was acquired. Axes have natural labels, expressing meaning, such
as "time" or "slice", and they may have tick labels such as acquisition time.
The scanner captures this information, but typical image formats cannot store
it, so it is easy to lose metadata and make analysis errors; see
`https://github.com/nipy/nibabel/wiki/BIAP6`_

We plan to expand Nibabel's API to encode axis and tick labels by integrating
the Xarray package: `http://xarray.pydata.org`_. Xarray simplifies HDF5
serialization, and visualization; see `https://napari.org`_.

An API for labels is not useful if we cannot read labels from the scanner
data, or save them with the image. We plan to:

* Develop HDF5 equivalents of standard image formats, for serialization of
data with labels.
* Expand the current standard image format, NIfTI, to store labels in a JSON
addition to image metadata: `https://github.com/nipy/nibabel/wiki/BIAP3`_.
* Read image metadata from DICOM, the standard scanner format.

Reading and attaching DICOM data will start with code integrated from Dcmstack,
by Brendan Moloney: `https://github.com/moloney/dcmstack`_; see:
`https://github.com/nipy/nibabel/wiki/BIAP4`_.

DICOM metadata is often hidden inside "private" DICOM elements that need
specialized parsers. We want to expand these parsers to preserve full metadata
and build a normalization layer to abstract vendor-specific storage locations
for metadata elements that describe the same thing.

API for surface data
--------------------

Neuroimaging data often refers to locations on the brain surface. There are
three common formats for such data: GIFTI, CIFTI and Freesurfer. Nibabel can
read these formats, but lacks a standard API for reading and storing surface
data with metadata; see `https://github.com/nipy/nibabel/issues/936`_,
`https://github.com/nilearn/nilearn/issues/2171`_. We plan to develop
a standard API, apply it to the standard formats, and design an efficient
general HDF5 storage container for serializing surface data and metadata.

Range
=====

Spatial transforms
------------------

Neuroimaging toolboxes include spatial registration methods to align the
objects and features present in two or more images. Registration methods
estimate and store spatial transforms. There is no standard or compatible
format to store and reuse these transforms, across packages.

Because Nibabel is a workbench, we want to extend its support to read
transforms calculated with AFNI, FreeSurfer, FSL, ITK/ANTs, NiftyReg, and SPM.

We have developed the NiTransforms project for this task; we plan to complete
and integrate NiTransforms into Nibabel. This will make transforms more
accessible to researchers, and therefore easier to work with, and reason about.