comments:

• mingwpy support for Python-3.5 not ready yet
• numba doesn't need a C compiler at all, however lllvmlite requires the new Windows 10 Universal C Runtime to be installed (http://www.microsoft.com/en-us/download/details.aspx?id=48234)
• pipwin https://github.com/lepisma/pipwin could be a nice addition

Hi Carl,

Thanks for the update.

pipwin looks interesting indeed

May I suggest adding the PyMC library to the distribution? It is used widely for Bayesian stochastic modeling, and has got more and more attention. For example, the current stars on GitHub for pymc + pymc3 is 1420 (comparable with the popular statsmodels with 1173 stars). Link: https://github.com/pymc-devs

PyMC3 beta seems indeed interesting, but we shall wait that Christoph Gohlke gives its blessing wheels to this project,

There is also the problem that Theano last official release is becoming more and more irrelevant: pymc-devs/pymc#850 (comment)

@yw5aj

so, it just a matter of shifting to Theano master. Pymc3 seems much smaller in size than pymc2 (and compiles directly in winpython via a nice pip install https://github.com/pymc-devs/pymc3/zipball/master)

Today, Anaconda has released python 3.5 version, but without PyQt5 (https://www.continuum.io/downloads)

Will next version winpython support python 3.5 ?

Probably not, unless:

• mingwpy becomes available for Python 3.5 by that time (for Cython),
• Python 3.5.1 is released,
• 500 million persons have already upgraded their PC to Windows 10, after 'Threshold 2'

All these events should happen around february 2016, according to my cristal ball and marketing understandings.

@yw5aj which pymc are you talking (using) about ? pymc2 or pymc3 ?

@stonebig First of all, sorry for the slow response!

I was thinking about pymc2 because

• pymc3 is still in beta, pymc2 is the "official stable release"
• We do not have Christoph's build yet as you mentioned

But Christoph might be building it soon, so holding it until he builds it would be a terrific idea too.

Shawn

and does pymc2 works to your convenience with today WinPython ?

@stonebig Yes - pip install pymc doesn't work, but I have been downloading Christoph's wheels and using it. I am happy to use it this way - I proposed it because I thought that might be to the interest of a broader audience. It's totally fine with me as it is now :)

pipwin 0.2.2 with Python 3 support just landed.

trying to wake-up a pymc3 beta2, as pymc3 looks way more active, and it doesn't seem so far from rc.

build1 (2015-11-08: scikit-learn 0.17, matplotlib 1.5.0, pyzmq 15, PythonQwt/guiqwt/guidata refresh)

main features (since previous release):

• scikit-learn 0.17, matplotlib 1.5.0 final,
• a few Jupyter-related upgrades: nbconvert, ipyparallel, pyzmq
• a few Qt-related upgrades: PythonQwt/guidata/guiqwt
• some bayesian packages: emcee, astroML
• Flavors and Builds are better integrated into the Build process.

next build effort:

• pandas 0.17.1, Ipython 4.1
• experimental:
  • pymc3 beta2, lasagne (and so pick a Theano commit)
  • WinPythonSlim: how slim while still usefull can we be ?

History of changes for WinPython 3.4.3.7Qt5Build1

The following changes were made to WinPython distribution since version 3.4.3.6Qt5.

Python packages

New packages:

• astroml 0.3 (tools for machine learning and data mining in Astronomy)
• emcee 2.1.0 (Kick ass affine-invariant ensemble MCMC sampling)

Upgraded packages:

• guidata 1.7.0b6 → 1.7.1 (Automatically generated graphical user interfaces for easy data set edition and display)
• guiqwt 3.0.0b7 → 3.0.0rc1 (Efficient curve/image plotting and other GUI tools for scientific data processing software development)
• ipyparallel 4.0.2 → 4.1.0 (Interactive Parallel Computing with IPython)
• ipywidgets 4.1.0 → 4.1.1 (IPython HTML widgets for Jupyter)
• joblib 0.9.2 → 0.9.3 (Lightweight pipelining: using Python functions as pipeline jobs.)
• lmfit 0.9.1 → 0.9.2 (Least-Squares Minimization with Bounds and Constraints)
• markdown 2.6.2 → 2.6.3 (Python implementation of Markdown.)
• matplotlib 1.5.0rc3 → 1.5.0 (2D plotting library (embeddable in GUIs created with PyQt))
• nbconvert 4.0.0 → 4.1.0 (Converting Jupyter Notebooks)
• numexpr 2.4.4 → 2.4.6 (Fast evaluation of array expressions elementwise by using a vector-based virtual machine)
• pymongo 3.0.3 → 3.1 (Python driver for MongoDB http://www.mongodb.org)
• pyparsing 2.0.3 → 2.0.5 (A Python Parsing Module)
• pythonqwt 0.4.0 → 0.5.0 (Qt plotting widgets for Python)
• pytz 2015.6 → 2015.7 (World Timezone Definitions for Python)
• pyzmq 14.7.0 → 15.0.0 (Lightweight and super-fast messaging based on ZeroMQ library (required for IPython Qt console))
• rpy2 2.7.2 → 2.7.3 (Python interface to the R language (embedded R))
• scikit-learn 0.16.1 → 0.17 (A set of Python modules for machine learning and data mining)
• setuptools 18.4 → 18.5 (Download, build, install, upgrade, and uninstall Python packages - easily)
• simplejson 3.8.0 → 3.8.1 (Simple, fast, extensible JSON (JavaScript Object Notation) encoder/decoder)
• sqlparse 0.1.16 → 0.1.18 (Non-validating SQL parser)
• twine 1.6.3 → 1.6.4 (Collection of utilities for interacting with PyPI)

build2 (2015-11-10: experimenting a Slim version of Winpython)

main features (since previous 'Qt5' build1):

• Simplify Build maintenance,
• Experimental 'WinPythonSlim' (32bit):
  • 50% smaller when installed (650 Mo and 23k files)
  • still hopefully complete for French educational programs
    ==> 50% smaller installation time & disk footprint

Feedback on this experiment is welcomed

next build effort:

• pandas 0.17.1, Ipython 4.1
• pymc3 beta2, lasagne (and so pick a Theano commit)

History of changes for WinPython 3.4.3.7Slim (build2)

The following changes were made to WinPython distribution since version 3.4.3.7Qt5 (build1)

Python packages

New packages:

• html5lib 0.9999999 (HTML parser based on the WHATWG HTML specification)

Upgraded packages:

• baresql 0.7.2 → 0.7.3 (playing SQL directly on Python datas)
• pythonqwt 0.5.0 → 0.5.1 (Qt plotting widgets for Python)
• tornado 4.2.1 → 4.3 (Scalable, non-blocking web server and tools (required for IPython notebook))

Removed packages:

• adodbapi 2.6.0.7 (A pure Python package implementing PEP 249 DB-API using Microsoft ADO.)
• alabaster 0.7.6 (A configurable sidebar-enabled Sphinx theme)
• astroml 0.3 (tools for machine learning and data mining in Astronomy)
• babel 2.1.1 (Internationalization utilities)
• bcolz 0.11.4 (columnar and compressed data containers.)
• blaze 0.8.3 (Blaze)
• blosc 1.2.8 (Blosc data compressor)
• bloscpack 0.9.0 (Command line interface to and serialization format for Blosc)
• bokeh 0.10.0 (Statistical and novel interactive HTML plots for Python)
• bottleneck 1.0.0 (Fast NumPy array functions written in Cython)
• bqplot 0.4.2 (Interactive plotting for the Jupyter notebook, using d3.js and ipywidgets.)
• brewer2mpl 1.4.1 (Connect colorbrewer2.org color maps to Python and matplotlib)
• castra 0.1.6 (On-disk partitioned store)
• certifi 2015.9.6.1 (Python package for providing Mozilla's CA Bundle.)
• cffi 1.3.0 (Foreign Function Interface for Python calling C code.)
• click 5.0 (A simple wrapper around optparse for powerful command line utilities.)
• cvxopt 1.1.7 (Convex optimization library)
• cx-freeze 4.3.4 (Deployment tool which converts Python scripts into stand-alone Windows executables (i.e. target machine does not require Python or any other library to be installed))
• cyordereddict 0.2.2 (Cython implementation of Python's collections.OrderedDict)
• cython 0.23.4 (Cython is a language that makes writing C extensions for the Python language as easy as Python)
• cytoolz 0.7.4 (Cython implementation of Toolz: High performance functional utilities)
• dask 0.7.5 (Minimal task scheduling abstraction)
• datashape 0.4.7 (A data description language)
• dill 0.2.4 (serialize all of python (almost))
• docopt 0.6.2 (Pythonic argument parser, that will make you smile)
• docutils 0.12 (Text processing system for processing plaintext documentation into useful formats, such as HTML or LaTeX (includes reStructuredText))
• emcee 2.1.0 (Kick ass affine-invariant ensemble MCMC sampling)
• flask 0.10.1 (A microframework based on Werkzeug, Jinja2 and good intentions)
• greenlet 0.4.9 (Lightweight in-process concurrent programming)
• holoviews 1.3.2 (Composable, declarative data structures for building complex visualizations easily.)
• husl 4.0.3 (Human-friendly HSL (Hue-Saturation-Lightness))
• itsdangerous 0.24 (Various helpers to pass trusted data to untrusted environments and back.)
• joblib 0.9.3 (Lightweight pipelining: using Python functions as pipeline jobs.)
• julia 0.1.1.8 (Python interface to the Julia language)
• keras 0.2.0 (Theano-based Deep Learning library)
• llvmlite 0.6.0 (lightweight wrapper around basic LLVM functionality)
• lmfit 0.9.2 (Least-Squares Minimization with Bounds and Constraints)
• locket 0.2.0 (File-based locks for Python for Linux and Windows)
• lxml 3.4.4 (Powerful and Pythonic XML processing library combining libxml2/libxslt with the ElementTree API.)
• markdown 2.6.3 (Python implementation of Markdown.)
• mingwpy 0.1.0b3 (the python friendly windows compiler toolchain)
• mpld3 0.2 (D3 Viewer for Matplotlib)
• multipledispatch 0.4.8 (A relatively sane approach to multiple dispatch in Python)
• mysql-connector-python 2.0.4 (MySQL driver written in Python)
• netcdf4 1.2.1 (python/numpy interface to netCDF library (versions 3 and 4))
• networkx 1.10 (Python package for creating and manipulating graphs and networks)
• nltk 3.0.5 (The Natural Language Toolkit (NLTK) is a Python package for natural language processing.)
• nose 1.3.7 (nose is a discovery-based unittest extension (e.g. NumPy test module is using nose))
• numba 0.20.0 (compiling Python code using LLVM)
• numpydoc 0.5 (Sphinx extension to support docstrings in Numpy format)
• oct2py 3.3.3 (Python to GNU Octave bridge --> run m-files from python.)
• odo 0.3.4 (Data migration in Python)
• param 1.3.2 (Declarative Python programming using Parameters.)
• partd 0.3.2 (Appendable key-value storage)
• pg8000 1.10.2 (PostgreSQL interface library)
• pkginfo 1.2.1 (Query metadatdata from sdists / bdists / installed packages.)
• pulp 1.5.9 (PuLP is an LP modeler written in python. PuLP can generate MPS or LP files and call GLPK, COIN CLP/CBC, CPLEX, and GUROBI to solve linear problems)
• py 1.4.30 (library with cross-python path, ini-parsing, io, code, log facilities)
• pyaudio 0.2.9 (Bindings for PortAudio v19, the cross-platform audio input/output stream library.)
• pycparser 2.14 (C parser in Python)
• pymongo 3.1 (Python driver for MongoDB http://www.mongodb.org)
• pyopengl 3.1.1b1 (Cross platform Python binding to OpenGL and related APIs)
• pyserial 2.7 (Library encapsulating the access for the serial port)
• pystache 0.5.4 (Mustache for Python)
• pytest 2.8.2 (pytest: simple powerful testing with Python)
• python-hdf4 0.9 (Python-HDF4: Python interface to the NCSA HDF4 library)
• pywin32 219 (Python library for Windows)
• pyyaml 3.11 (YAML parser and emitter for Python)
• redis 2.10.3 (Python client for Redis key-value store)
• reportlab 3.2.0 (The PDF generation library)
• requests-toolbelt 0.4.0 (Requests is an Apache2 Licensed HTTP library, written in Python, for human beings.)
• rpy2 2.7.3 (Python interface to the R language (embedded R))
• rx 1.2.3 (Reactive Extensions (Rx) for Python)
• scikit-image 0.11.3 (Image processing toolbox for SciPy)
• scikit-learn 0.17 (A set of Python modules for machine learning and data mining)
• scilab2py 0.6 (Python to Scilab bridge)
• snowballstemmer 1.2.0 (This package provides 16 stemmer algorithms (15 + Poerter English stemmer) generated from Snowball algorithms.)
• sphinx 1.3.1 (Tool for generating documentation which uses reStructuredText as its markup language)
• sphinx-rtd-theme 0.1.9 (Tool for generating documentation which uses reStructuredText as its markup language)
• sympy 0.7.6.1 (Symbolic Mathematics Library)
• theano 0.7.0 (Optimizing compiler for evaluating mathematical expressions on CPUs and GPUs.)
• toolz 0.7.4 (List processing tools and functional utilities)
• twine 1.6.4 (Collection of utilities for interacting with PyPI)
• twitter 1.17.1 (An API and command-line toolset for Twitter (twitter.com))
• werkzeug 0.10.4 (The Swiss Army knife of Python web development)
• wheel 0.26.0 (A built-package format for Python.)
• xlsxwriter 0.7.7 (A Python module for creating Excel XLSX files.)
• xray 0.6.1 (N-D labeled arrays and datasets in Python)

build3 (2015-11-11: experimenting a Slim version of Winpython for Python3.5 64bit)

main features (since previous build2):

• Simplify Build maintenance,
• Tweak Build system for Python3.5,
• Experimental 'WinPythonSlim' (64bit) for Python3.5:
  • nearly identical to yesterday Slim version for 3.4,
  • differences are:without guiqwt, + Python 3.5 64 bit instead of Python 3.4 32 bit.

Feedback on this early Python3.5 experiment is welcomed:

• users with windows 10 may have a good chance to have no problem,
• others may have to install KB2999226 of Microsoft.

WinPython 3.5.0.1Slim

The following packages are included in WinPython v3.5.0.1Slim build3.

Tools

Python packages

build4 (2015-11-14: experimenting a full Qt4 version of Winpython for Python3.5 64bit)

main features (since previous build3Slim):

• update: guiqwt/guidata/seaborn
• new: pymc2
• remove from winpython_checker what we can't have working in 3.5 (mingwpy / Theano / Pulp / pywin...)

Feedback on this last Python3.5 dedicated build experiment is welcomed:

• users with windows 10 may have a good chance to have no problem,
• others may have to install KB2999226 of Microsoft.

Next builds:

• more cleanup in build system
• back to Ipython / Spyder / Pandas fundamental upgrades
• considering a Python 3 only package to somewhat mitigate Pulp not in 3.5 (pyeda)

WinPython 3.5.0.1

The following packages are included in WinPython v3.5.0.1 build4.

Tools

Python packages

Here is a link for the Universal C Runtime for an OS-specific, smaller download:
https://support.microsoft.com/en-us/kb/2999226

Since WinPython Release 2015-07 is the end of Python 2 support, I suggest offering "still" and "fresh" variants of Python 3.4. This is similar to LibreOffice, where "still" implies stability and bug fixes, while "fresh" implies bleeding-edge, new features and high risk for bugs.

Very important to me is stability in Spyder. The Spyder team is very busy making changes to support Qt5 and Python 3.5, but they haven't yet made a release supporting these. I also saw a Spyder commit saying that pylint doesn't yet support Python 3.5. I would rather stay with Qt4 and Python 3.4 for a few months until Spyder stabilizes. I am not clear on whether the latest numba, which requires the new Universal C Runtime, is too risky for the "still" variant.

My own whish is to maintain WinPython 3.4 until July 2016.

That said, we can't really see further than next release, like most open-source projects.

build5 (2015-11-21: pandas 0.17.1, MKL update, formlayout for Qt5)

Main features (since previous build4):

• update: Spyder, Pandas, Holoviews, all MKL binaries + pymc on WinPython3.4
• new: formlayout is now compatible with Qt5,
• more cleanup in build system , allowing next experimental line ...
• ... WinPythonZero: smaller than Python, (slightly) bigger on the inside (no Qt, no Spyder, ...)

Feedback welcome on:

• Spyder3.0 bugs, Holoviews bugs,
• Real use case for a WinPythonZero.

Next builds:

• Python 3.5.1rc1, Holoviews 1.4
• Ipython 4.1 on the horizon ?

WinPython 3.4.3.7Zero

The following packages are included in WinPython v3.4.3.7Zero build5.

Tools

Python packages

build6 (2015-11-21/22/23: pandas 0.17.1, MKL update, formlayout for Qt5)

Main features (since previous build4):

• update: Spyder, Pandas, Holoviews, all MKL binaries + pymc on WinPython3.4
• new: formlayout is now compatible with Qt5,
• more cleanup in build system , allowing next experimental line ...
• ... WinPythonZero: smaller than Python, (slightly) bigger on the inside (no Qt, no Spyder, ...)
• build 6 = build5 with the latest WinPython wheel

Feedback welcome on:

• Spyder3.0 bugs, Holoviews bugs,
• Real use case for a WinPythonZero.

Next builds:

• Python 3.5.1rc1, Holoviews 1.4
• Ipython 4.1 on the horizon ?

WinPython 3.4.3.7Zero

The following packages are included in WinPython v3.4.3.7Zero build6.

Tools

Python packages

History of changes for WinPython 3.4.3.7Qt5

The following changes were made to WinPython distribution since version 3.4.3.6Qt5.

Python packages

New packages:

• astroml 0.3 (tools for machine learning and data mining in Astronomy)
• emcee 2.1.0 (Kick ass affine-invariant ensemble MCMC sampling)
• formlayout 1.1.0 (Module for creating form dialogs/widgets to edit various type of parameters without having to write any GUI code)
• html5lib 0.9999999 (HTML parser based on the WHATWG HTML specification)
• pymc 2.3.6 (Markov Chain Monte Carlo sampling toolkit.)

Upgraded packages:

• baresql 0.7.2 → 0.7.3 (playing SQL directly on Python datas)
• bcolz 0.11.4 → 0.12.0 (columnar and compressed data containers.)
• cffi 1.3.0 → 1.3.1 (Foreign Function Interface for Python calling C code.)
• guidata 1.7.0b6 → 1.7.2 (Automatically generated graphical user interfaces for easy data set edition and display)
• guiqwt 3.0.0b7 → 3.0.0 (Efficient curve/image plotting and other GUI tools for scientific data processing software development)
• holoviews 1.3.2 → 1.4.1a1 (Composable, declarative data structures for building complex visualizations easily.)
• ipyparallel 4.0.2 → 4.1.0 (Interactive Parallel Computing with IPython)
• ipywidgets 4.1.0 → 4.1.1 (IPython HTML widgets for Jupyter)
• joblib 0.9.2 → 0.9.3 (Lightweight pipelining: using Python functions as pipeline jobs.)
• lmfit 0.9.1 → 0.9.2 (Least-Squares Minimization with Bounds and Constraints)
• lxml 3.4.4 → 3.5.0 (Powerful and Pythonic XML processing library combining libxml2/libxslt with the ElementTree API.)
• markdown 2.6.2 → 2.6.4 (Python implementation of Markdown.)
• matplotlib 1.5.0rc3 → 1.5.0 (2D plotting library (embeddable in GUIs created with PyQt))
• nbconvert 4.0.0 → 4.1.0 (Converting Jupyter Notebooks)
• numexpr 2.4.4 → 2.4.6 (Fast evaluation of array expressions elementwise by using a vector-based virtual machine)
• pandas 0.17.0 → 0.17.1 (Powerful data structures for data analysis, time series and statistics)
• patsy 0.4.0 → 0.4.1 (Describing statistical models using symbolic formulas)
• pymongo 3.0.3 → 3.1.1 (Python driver for MongoDB http://www.mongodb.org)
• pyparsing 2.0.3 → 2.0.6 (A Python Parsing Module)
• pythonqwt 0.4.0 → 0.5.1 (Qt plotting widgets for Python)
• pytz 2015.6 → 2015.7 (World Timezone Definitions for Python)
• pyzmq 14.7.0 → 15.0.0 (Lightweight and super-fast messaging based on ZeroMQ library (required for IPython Qt console))
• redis 2.10.3 → 2.10.5 (Python client for Redis key-value store)
• rpy2 2.7.2 → 2.7.4 (Python interface to the R language (embedded R))
• rx 1.2.3 → 1.2.4 (Reactive Extensions (Rx) for Python)
• scikit-learn 0.16.1 → 0.17 (A set of Python modules for machine learning and data mining)
• setuptools 18.4 → 18.5 (Download, build, install, upgrade, and uninstall Python packages - easily)
• simplejson 3.8.0 → 3.8.1 (Simple, fast, extensible JSON (JavaScript Object Notation) encoder/decoder)
• sqlparse 0.1.16 → 0.1.18 (Non-validating SQL parser)
• tornado 4.2.1 → 4.3 (Scalable, non-blocking web server and tools (required for IPython notebook))
• twine 1.6.3 → 1.6.4 (Collection of utilities for interacting with PyPI)
• werkzeug 0.10.4 → 0.11.2 (The Swiss Army knife of Python web development)
• winpython 1.2 → 1.3 (WinPython distribution tools, including WPPM (package manager))

build7 (2015-11-24+: Python3.5.1rc1)

Main features (since previous build6):

• python3.5: update to python3.5.1rc1 changelog + include a pyvenv.cfg
• spyder 3.0.0b2-20151127 (just before the small 'big split')

History of changes for WinPython 3.5.1.1

The following changes were made to WinPython distribution since version 3.5.0.1.

Python packages

New packages:

• lazy-object-proxy 1.2.1 (A fast and thorough lazy object proxy.)
• prompt-toolkit 0.54 (Library for building powerful interactive command lines in Python)
• wcwidth 0.1.5 (Measures number of Terminal column cells of wide-character codes)
• wrapt 1.10.5 (A Python module for decorators, wrappers and monkey patching.)

Upgraded packages:

• astroid 1.3.6 → 1.4.1 (Rebuild a new abstract syntax tree from Python's ast (required for pylint))
• bcolz 0.12.0 → 0.12.1 (columnar and compressed data containers.)
• certifi 2015.9.6.1 → 2015.11.20.1 (Python package for providing Mozilla's CA Bundle.)
• click 5.0 → 6.2 (A simple wrapper around optparse for powerful command line utilities.)
• holoviews 1.4.1a1 → 1.4.0a1 (Composable, declarative data structures for building complex visualizations easily.)
• ipykernel 4.1.1 → 4.2.0 (IPython Kernel for Jupyter)
• ipython 4.0.0 → 4.0.1 (Enhanced Python shell)
• markdown 2.6.4 → 2.6.5 (Python implementation of Markdown.)
• psutil 3.2.2 → 3.3.0 (Provides an interface for retrieving information on all running processes and system utilization (CPU, disk, memory, network) in a portable way)
• py 1.4.30 → 1.4.31 (library with cross-python path, ini-parsing, io, code, log facilities)
• pylint 1.4.3 → 1.5.0 (Logilab code analysis module: analyzes Python source code looking for bugs and signs of poor quality)
• pytest 2.8.2 → 2.8.3 (pytest: simple powerful testing with Python)
• Python 3.5.0 → 3.5.1 (Python programming language with standard library)
• pyzmq 15.0.0 → 15.1.0 (Lightweight and super-fast messaging based on ZeroMQ library (required for IPython Qt console))
• qtconsole 4.1.0 → 4.1.1 (Jupyter Qt console)
• requests-toolbelt 0.4.0 → 0.5.0 (Requests is an Apache2 Licensed HTTP library, written in Python, for human beings.)
• setuptools 18.5 → 18.7 (Download, build, install, upgrade, and uninstall Python packages - easily)
• sphinx 1.3.1 → 1.3.2 (Tool for generating documentation which uses reStructuredText as its markup language)

Removed packages:

• logilab-common 1.1.0 (Collection of low-level Python packages and modules used by Logilab projects (required for pylint))

Spyder v2.3.8 has been released, which includes a fix for compatibility with matplotlib v1.5.

Don't Panic !

Spyder is chasing a (too) wide range of targets: Python2/3, Linux/Windows/Apple. the fix/bug was for Python 2 / 'linux' combination.

Thanks for the clarification.

I notice that NumPy v1.10.x has been removed from here:
http://www.lfd.uci.edu/~gohlke/pythonlibs/#numpy
Maybe there is a serious bug with it, so I am waiting for at least v1.10.2 before testing.

Upgrading numpy too early caused me issues (work) last year.
The same happened this year with matplotlib and Ipython.
==> I will be now extra cautious for these three.

pylint v1.5.1 is now available, which is especially important for Python 3.5:
https://pypi.python.org/pypi/pylint/1.5.1
Some of the changes:
http://docs.pylint.org/changelog.html#changelog-for-pylint

ok

Because Wiki is not practical to me to post changelogs and package index, I will jus post them as .md on the source. It shows result as if it was in The Wiki, with much less efforts.