thehackerwithin / berkeley Goto Github PK

Hi, not sure if this will be of interest in general for the group. pandoc is already involved in academic writing for example as the backend of rmarkdown in R, and ipython notebook in Python; and also Authorea which is an alternative of ShareLaTeX and Overleaf.

pandoc base on the markdown syntax, with extensions such as raw html and raw LaTeX, and is a program that input and output md, docx, tex, html, epub, etc.

I'm kind of a heavy user and contributor of pandoc. I'm not sure if I'm the right person to give a talk though (I'm a newbie here). The creator of pandoc is a Berkeley professor. I could contact him and see if he's interested in giving the talk. But the more important question is if the audiences will like it.

Shiny, Cesium, Flask, Jupyter extensions

Or other tools, NextFlow?

hi @cypranowska , thanks for this awesome tutorial, I bummed I missed it.

I found two tiny typos in the tutorial as I was trying to follow it, both in this section:

• missing closing quotation mark for "height"
• extra unnecessary curlycue brace in cx function

thanks!

Two separate sessions

Add this link
to 2017-10-04-github-oss-f17.md

Working on my talk for this week's meeting, and doing most of the demos in a separate repo since I'm using an actual C library as an example. I'm thinking of adding it as a submodule under an appropriate folder here. Anyone opposed to that plan, if I put in a README comment explaining how to make git do the right thing?

Python package for regressions, etc

Since we have some community members who don't have formal Berkeley affiliations. I've seen other slack channels do this where you can request access and a moderator confirms yes you're a real person to approve it, but I'm not sure where in the slack settings or what other steps you need to do to enable that.

Hi @katyhuff — So when I signed up for giving the LaTeX presentation, I hadn't realized how many deadlines for postdoc positions, grants and conferences were going to appear right at the beginning of October.

@jljones @paciorek would you be ok moving to another week? If not, I don't think I will be able to help prepare a talk with ya'll…But, I will do mini lightning talks on the more foundational/theoretical aspects of tex/LaTeX once my plate clears up a bit (since that's what I'd expected to cover… e.g., brief history of font formats, what metafont is, what TeX vs. LaTeX are, the glue and box model, &c.).

@stefanv, @rossbar or @chick are any of you able to swap (assuming @jljones and @paciorek can swap, of course)?

website has previous posts listed as '2015-2016', but there are definitely some dating back to 2014. :)

gave a 30-second pass through and couldn't find where to change that, and thus this issue was opened.

Upvote if interested?

I know a very a little amount, but oddly useful things, like cropping and annotating images from the command line (which can be used to process 100s of images in seconds, if that's your thing) and looping gifs. This would take a fair amount of effort to prepare as a tutorial, so I'd like to judge interest before committing.

Report errors :

• Bogetich is wrong, correct version Bogetic
• mispelled name of Ms. Munk, "Madicken" is the corrected version

As a digression from #108. Warning: just a brainstorm here, the idea may be too crazy.

One obstacle in hacker-within is that people joining the sessions are often having vastly different environments, experiences, and even OSes. And unavoidably, experiences on some topics might not be uniform across different OSes (e.g. bash, make, docker, CI, etc.) In some situations it is also unavoidable for beginners to find themselves needing to migrate to the UNIX world in certain fields. So the idea of this topic is to help these people to turn their OS into a UNIX/GNU machine (more on GNU later.)

Ideas:

• Windows
  • Cygwin (native but limited)
  • Windows Subsystem for Linux (64-bit only, has system call overhead)
  • Install Ubuntu using Hyper-V (64-bit and the "Pro-variants" of Windows only, lowest overhead.)
  • Other VM solutions (may cost $, more overhead)
    • may also mention hacking VMWare to run macOS in a virtual machine
• macOS
  • macOS is already UNIX, but point out differences (e.g. BSD variant, GPLv2 limitations, etc.)
  • GNU-ize : because of the limitation of GPLv2, macOS shipped with older GNU tools (e.g. sed, make, bash) or replace them with non GNU tools (fake gcc, etc.), fortunately through brew (or macports), newest GNU tools can be installed. The goal is to minimize the difference between macOS and Linux, and hence people can develop on macOS but deploy on Linux (majority of the clouds and virtually all supercomputers)
  • other tricks is to not keeping the OS being the latest release to be able to use some tools (e.g. valgrind, Intel's tools) that aren't updated immediately after an OS launched.
  • in brew, override default compilers to use the GNU's variants (rather than clang)
  • VMs could be mentioned too
  • Hackintosh: this is mentioned because beginners might not have a Mac. Migrating to Linux means they are going to lose most of the propriety tools they need. macOS can be a sweet spot of being UNIX but with many commerical appications avaiable. But it can't run on their existing hardware so hackintosh is needed if they don't want to buy a new machine, or just want to get a taste of it first. Making it working can be challenging or impossible though, depending on the hardwares.
  • Darwine and friends: for those who need to migrate from Windows to macOS, Darwine could be helpful. See below for Wine.
• docker: it is an alternative way to run UNIX/Linux applications on Windows and Mac. They uses hypervisor technology to minimize overhead. This is the ultimate way to done away the differences between Windows, macOS and Linux, not to mention very useful on Linux as well.
• Linux
  • Wine: while Linux being Linux, there's no need to modify it to get the UNIX/GNU experience, Wine could be helpful for those needing to migrate from Windows to Linux. It is a compatibility layer to translate the system calls of an Windows application to make them run natively on a Linux host (WSL steal this idea to run Linux application natively on Windows.) It ain't perfect though, and there's a long list showing which Windows applications might be run using Wine.
  • VM solution also avaiables here to install Windows and macOS
• Other interesting cross-breeding:
  • running Linux alongside Chrome OS (meaning one could buy a cheap Chrome OS mahcine if they want to buy a Linux machine.)
  • Ubuntu Phone: there's a limited no. of Android phones/tablets that can install Ubuntu. Also a good candidate for cheap Linux machine.

add name, Andrey Mironyuk

Posting an issue about this as part of the live demo.

My last name is spelled "Rowland", not "Rowlend".

add name to ORIGEN... post, name is Tenzing Joshi

Hi, I'd be happy to co-lead a session on d3. Coming from geography, I could put an emphasis on maps, but not necessarily.
Caroline is looking for a co-leader...?

The installation instructions for NLTK+Jupyter for the October 11, 2016 THW session are confusing and difficult and assumes a lot of prior knowledge about how to set up and install things like Java, and anaconda, and... lots of stuff!

Maybe the Cloud Working Group can help?

Installation
We are using this Jupyter notebook in the thehackerwithin/berkeley repo, master branch, nltk folder.

For installation of Python and NLTK follow these instructions

If you installed anaconda:

conda install nltk

Otherwise:

pip install nltk

Lastly, the NER wrapper requires the Java Stanford NER here: Note: do not download the extension, just Download Stanford Named Entity Recognizer version 3.6.0

Should be 2 sessions in a row

Might be a bunch of IDEs

Co-leaders:

• We shall get together this week (e.g. on Thursday) to decide on the materials and logistics.

• Some jump start material is at http://swcarpentry.github.io/python-novice-inflammation/; but it is a four hour lecture. We can't do that much.

• We shall spend some time setting up a Python distribution for the attendees. (e.g. anaconda) A GUI is useful, so Jupyter notebook.

For the actually 'Introduction to Python programming', a few things that popped into my mind are:

• What is computer programming? How does it differ and assimilate with mathematics?

• The confused equal sign. L-value and R-value.

• Identity of objects, internal state and immutability

• Functions & Recursion

• Control structures. Conditionals and Loops. -- How does this fit into a functional perspective of programming?

• Dealing with Tensors (Arrays)

• Strings and their representations, encoding, unicode, bytes.

Probably way too heavy for an intro ...

I realize that this is slightly last-minute, so I do apologize...

For the IPython talk on February 25, would it be possible to go over some of the complete and total basics of the IPython notebooks (creating one, launching it in-browser, etc.)?

@minrk @takluyver

The master vs. gh-pages branch has long been a complicated thing for people to figure out and switch between. GitHub pages now supports being built from a /docs subfolder in a master branch. Any thoughts on refactoring the folders in the master branch into something like /code and gh-pages in /docs ?

I'd like to request a session on Docker and Singularity.

In particular, I'm interested in understanding how to manipulate images if you don't have the original Dockerfiles, and I'm interested in Singularity for cluster computing reasons.

problem:

there is no LICENSE

proposal:

add BSD-3 using github builtin license utility in community

for that matter, there's no code of conduct or contribution guidelines either.