Light

Factual Errors in “Git vs Mercurial: Why Git?” -- and corrections shown by example

Update 2016: Instead of fixing the article, the Atlassian web workers removed the comments which point out the misinformation in the article. *sigh*

Summary:

In the Atlassian Blog, a Git proponent spread blatant misinformation which the Atlassian folks are leaving uncommented even though the falseness has been shown by multiple people and even in examples in the article itself.

The claims and corrections:

  • Claim: Git never loses unreferenced data. Mercurial needs special handling to retrieve unreferenced data. Reality: Due to automatic garbage collection, history editing in git unpredictably loses unreferenced history while Mercurial stores permanent backups which can be retrieved with core commands.
  • Claim: Only git branches are namespaced. Reality: Mercurial bookmarks are namespaced with bookmark@path, when there could be confusion. This is equivalent to git’s use of path/branch, but only used where it is needed, while git forces the user to always make that distinction.
  • Claim: Only git can provide a staging area. Reality: Activating mercurial queues (mq) and the record extension provides a staging area like the git index — for those who want it.
  • Claim: Git is more powerful. Reality: Both have the same raw power (as proven by transparent access with Mercurial to Git repos via hg-git), but
  • its “cuddly command line” gives Mercurial an efficiency during actual usage which most people do not find in Git.

2 years ago, Atlassian developer Charles O’Farrell published the article Git vs. Mercurial: Why Git? in which he claimed to show "the winning side of Git”. This article was part of the Dev Tools series at Atlassian and written as a reply to the article Why Mercurial?. It was spiced with so much misinformation about Mercurial (statements which were factually wrong) that the comments exploded right away. But the article was never corrected. Just now I was referred to the text again, and I decided to do what I should have done 2 years ago: Write an answer which debunks the myths.

“I also think that git isn’t the most beginner-friendly program. That’s why I’m only using its elementary features” — “I hear that from many git-users …” — part of the discussion which got me to write this article

Wicked Words! on Patreon

John Wick is entering the patreon arena with the Wicked Words! Magazine: Adventures, GM Advice, Little Games, Stories, The Works!

There’s an update with a Happy ending on 1w6.org/english/wicked-words-patreon!

  Yay!

This is really good news for online publishing, because it shows by example how roleplaying games and shortstories enter a new stage on the web: Fan-Funded periodicals.

Write multiple images on a single page in org-mode.

How to add show multiple images on one page in the latex-export of emacs org-mode. I had this problem. This is my current solution.

Unicode char \u8:χ not set up for use with LaTeX: Solution (made easy with Emacs)

For years I regularly stumbled over LaTeX-Errors in the form of Unicode char \u8:χ not set up for use with LaTeX. I always took the chickens path and replaced the unicode characters with the tex-escapes in the file. That was easy, but it made my files needlessly unreadable. Today I decided to FIX the problem once and for all. And it worked. Easily.

Firstoff: The problem I’m facing is that my keyboard layout makes it effortless for me to input characters like ℂ Σ and χ. But LaTeX cannot cope with them out-of-the-box.

Publish a single file with emacs org-mode

I often write small articles on some experience I make, and since I want to move towards using static pages more often, I tried using emacs org-mode publishing for that. Strangely the simple usecase of publishing a single file seems quite a bit more complex than needed, so I document the steps here.

This is my first use of org-publish, so I likely do not use it perfectly. But as it stands, it works. You can find the org-publish version of this article at draketo.de/proj/orgmode-single-file.

How to show the abstract before the table of contents in org-mode

I use Emacs Org-Mode for writing all kinds of articles. The standard format for org-mode is to show the table of contents before all other content, but that requires people to scroll down to see whether the article is interesting for them. Therefore I want the abstract to be shown before the table of contents.

Top 5 systemd troubles - a strategic view for distros

systemd is a new way to start a Linux-system with the expressed goal of rethinking all of init. These are my top 5 gripes with it. (»skip the updates«)

Update (2016-09-28): Systemd is an exploit kit just waiting to be activated. And once it is active, only those who wrote it will be able to defuse it — and check whether it is defused. And it is starting: How to crash systemd in one tweet? Alternatives? Use OpenRC for system services. That’s simple and fast and full-featured with minimal fuss. Use runit for process supervision of user-services and system-services alike.

Update (2014-12-11): One more deconstruction of the strategies around systemd: systemd: Assumptions, Bullying, Consent. It shows that the attitude which forms the root of the dangers of systemd is even visible in its very source code.

Update (2014-11-19): The Debian General Resolution resulted in “We do not need a general resolution to decide systemd”. The vote page provides detailed results and statistics. Ian Jackson resigned from the Technical Committee: “And, speaking personally, I am exhausted.”

Update (2014-10-16): There is now a vote on a General Resolution in Debian for preserving the ability to switch init systems. It is linked under “Are there better solutions […]?” on the site Shall we fork Debian™? :^|.

Update (2014-10-07): Lennart hetzt (german) describes the rhetoric tricks used by Lennart Poettering to make people forget that he is a major part of the communication problems we’re facing at times - and to hide valid technical, practical, pragmatical, political und strategical criticism of Systemd.

Update (2014-09-24): boycott systemd calls for action with 12 reasons against systemd: “We do recognize the need for a new init system in the 21st century, but systemd is not it.”

Update (2014-04-03): And now we have Julian Assange warning about NSA control over Debian, Theodore Ts’o, maintainer of ext4, complaining about incomprehensible systemd, and Linus Torvalds (you know him, right?) rant against disrupting behavior from systemd developers, going as far as refusing to merge anything from the developers in question into Linux. Should I say “I said so”? Maybe not. After all, I came pretty late. Others saw this trend 2 years before I even knew about systemd. Can we really assume that there won’t be intentional disruption? Maybe I should look for solutions. It could be a good idea to start having community-paid developers.

Update (2014-02-18): An email to the mailing list of the technical committee of debian summarized the strategic implications of systemd-adoption for Debian and RedHat. It was called conspiracy theory right away, but the gains for RedHat are obvious: RedHat would be dumb not to try this. And only a fool trusts a company. Even the best company has to put money before ethics.

Update (2013-11-20): Further reading shows that people have been giving arguments from my list since 2011, and they got answers in the range of “anything short of systemd is dumb”, “this cannot work” (while OpenRC clearly shows that it works well), requests for implementation details without justification and insults and further insults; but the arguments stayed valid for the last 2 years. That does not look like systemd has a friendly community - or is healthy for distributions adopting it. Also an OpenRC developer wrote the best rebuttal of systemd propaganda I read so far: “Alternativlos”: Systemd propaganda (note, though, that I am biased against systemd due to problems I had in the past with udev kernel-dependencies)

Phoronix conclusions distort their results, shown with the example of GCC vs. LLVM/Clang On AMD's FX-8350 Vishera

Phoronix recently did a benchmark of GCC vs. LLVM on AMD hardware. Sadly their conclusion did not fit the data they showed. Actually it misrepresented the data so strongly, that I decided to speak up here instead of having my comments disappear in their forums. This post was started on 2013-05-14 and got updates when things changed - first for the better, then for the worse.

Update 3 (the last straw, 2013-11-09): In the recent most blatant attack by Phoronix on copyleft programs - this time openly targeted at GNU - Michael Larabel directly misrepresented a post from Josh Klint to badmouth GDB (Josh confirmed this1). Josh gave a report of his initial experience with GDB in a Kickstarter Update in which he reported some shortcomings he saw in GDB (of which the major gripe is easily resolved with better documentation2) and concluded with “the limitations of GDB are annoying, but I can deal with it. It's very nice to be able to run and debug our editor on Linux”. Michael Larabel only quoted the conclusion up to “annoying” and abused that to support the claim that game developers (in general) call GDB “crap” and for further badmouthing of GDB. With this he provided the straw which I needed to stop reading Phoronix: Michael Larabel is hostile to copyleft and in particular to GNU and he goes as far as rigging test results3 and misrepresenting words of others to further his agenda. I even donated to Phoronix a few times in the past. I guess I won’t do that again, either. I should have learned from the error of the german pirates and should have avoided reading media which is controlled by people who want to destroy what I fight for (sustainable free software).
Update 2 (2013-07-06): But the next went down the drain again… “Of course, LLVM/Clang 3.3 still lacks OpenMP support, so those tests are obviously in favor of GCC.” — I couldn’t find a better way to say that those tests are completely useless while at the same time devaluing OpenMP support as “ignore this result along with all others where GCC wins”…
Update (2013-06-21): The recent report of GCC 4.8 vs. LLVM 3.3 looks much better. Not perfect, but much better.


  1. Josh Klint from Leadwerks confirmed that Phoronix misrepresented his post and wrote a followup-post: » @ArneBab That really wasn't meant to be controversial. I was hoping to provide constructive feedback from the view of an Xcode / VS user.« » Slightly surprised my complaints about GDB are a hot topic. I can make just as many criticisms of other compilers and IDEs.« » The first 24 hours are the best for usability feedback. I figure if they notice a pattern some of those things will be improved.« » GDB Follwup «@Leadwerks, 2:04 AM - 11 Nov 13, 2:10 AM - 11 Nov 13 and @JoshKlint, 2:07 AM - 11 Nov 13, 8:48 PM - 11 Nov 13

  2. The first-impression criticism from Josh Klint was addressed by a Phoronix reader by pointing to the frame command. I do not blame Josh for not knowing all tricks: He wrote a fair account of his initial experience with GDB (and he said later that he wrote the post after less than 24 hours of using GDB, because he considers that the best time to provide feedback) and his experience can serve as constructive criticism to improve tutorials, documentation and the UI of GDB. Sadly his visibility and the possible impact of his work on free software made it possible for Phoronix to abuse a personal report as support for a general badmouthing of the tool. In contrast the full message of Josh Klint ended really positive: Although some annoyances and limitations have been discovered, overall I have found Linux to be a completely viable platform for application development. — Josh Klint, Leadwerks 

  3. I know that rigging of tests is a strong claim. The actions of Michael Larabel deserve being called rigging for three main reasons: (1) Including compile-time data along with runtime performance without clear distinction between both, even though compile-time of the full code is mostly irrelevant when you use a proper build system and compile time and runtime are completely different classes of results, (2) including pointless tests between incomparable setups whose only use is to relativate any weakness of his favorite system and (3) blatantly lying in the summaries (as I show in this article). 

Making websafe colors safe for colorblind people

I just made the colors of my plotting framework safe for colorblind people (thanks to Paul Tol’s notes) and I want to share a very nice result I got: How to make the really websafe colors safe for colorblind people with minimal changes.

mostly websafe and colorblindsafe websafe but NOT colorblind safe

(the colorblind-safe colors are on the left, the original websafe colors on the right)

To do so, I turned to Color Oracle (for simulation of colorblindness directly on my screen) and Emacs rainbow-mode (for seeing the colors while editing the hex-codes - as shown in the screenshots above) and tweaked the color codes bit by bit, until they were distinguishable in the simulation of Deuteranopia, Protanopia and Tritanopia.

Installing Scipy and PyNIO on a Bare Cluster with the Intel Compiler

2 years ago I had the task of running a python-program using scipy on our university cluster, using the Intel Compiler. I needed all those (as well as PyNIO and some other stuff) for running TM5 with the python shell on the HC3 of KIT.

This proved to be quite a bit more challenging than I had expected - but it was very interesting, too (and there I learned the basics of GNU autotools which still help me a lot).

But no one should have to go to the same effort with as little guidance as I had, so I decided to publish the script and the patches I created for installing everything we needed.1

The script worked 2 years ago, so you might have to fix some bits. I won’t promise that this contains everything you need to run the script - or that it won’t be broken when you install it. Actually I won’t promise anything at all, except that if the stuff here had been available 2 years ago, that could have saved me about 2 months of time (each of the patches here required quite some tracking of problems, experimenting and fixing, until it provided basic functionality - but actually I enjoyed doing that - I learned a lot - I just don’t want to be forced to do it again). Still, this stuff contains quite some hacks - even a few ugly ones. But it worked.

Installing GNU Guix 0.6, easily

Org-Source (for editing)

PDF (for printing)

“Got a power-outage while updating?
No problem: Everything still works”

GNU Guix is the new functional package manager from the GNU Project which complements the Nix-Store with a nice Guile Scheme based package definition format.

What sold it to me was “Got a power-outage while updating? No problem: Everything still works” from the Guix talk of Ludovico at the GNU Hacker Meeting 2013. My son once found the on-off-button of our power-connector while I was updating my Gentoo box. It took me 3 evenings to get it completely functional again. This would not have happened with Guix.

Update (2014-05-17): Thanks to zerwas from IRC @ freenode for the patch to guix 0.6 and nice cleanup!

Intro

Installation of GNU Guix is straightforward, except if you follow the docs, but it’s not as if we’re not used to that from other GNU utilities, which often terribly short-sell their quality with overly general documentation ☺

So I want to provide a short guide how to setup and run GNU Guix with ease. My system natively runs Gentoo, My system natively runs Gentoo, so some details might vary for you. If you use Gentoo, you can simply copy the commands here into the shell, but better copy them to a text-file first to ensure that I do not try to trick you into doing evil things with the root access you need.

In short: This guide provides the First Contact and Black Triangle for GNU Guix.

Communicating your project: honest marketing for free software projects

Communicating your project is an essential step for getting the users you want. Here I summarize my experience from working on several different projects including KDE (where I learned the basics of PR - yay, sebas!), the Hurd (where I could really make a difference by improving the frontpage and writing the Month of the Hurd), Mercurial (where I practiced the minimally invasive PR) and 1d6 (my own free RPG where I see how much harder it is to do PR, if the project to communicate is your own).

Since voicing the claim that marketing is important often leads to discussions with people who hate marketing of any kind, I added an appendix with an example which illustrates nicely what happens when you don’t do any PR - and what happens if you do PR of the wrong kind.

If you’re pressed for time and want the really short form, just jump to the questionnaire.

Use the source, Luke! — Emacs org-mode beamer export with images in figure

I just needed to tweak my Emacs org-mode to beamer-latex export to embed images into a figure environment (not wrapfigure!). After lots of googling and documentation reading I decided to bite the bullet and just read the source. Which proved to be much easier than I had expected.

This tutorial requires at least org-mode 8.0 (before that you had to use hacks to get figure without a caption). It is only tested for org-mode 8.0.2: The code you see when you read the source might look different in other versions.

default answer to “I want to connect with you on [hip unfree service]”

I just decided to give a default answer when I get some email from people asking me to connect to them on some new unfree service:

Hello [Person],

You asked me to connect with you on some unfree service. If you still want that, just use a status.net-server. Those are federated, so you can use a number of different providers and still be connected to everyone on any other server.

Surprising behaviour of Fortran (90/95)

1 Introduction

I recently started really learning Fortran (as opposed to just dabbling with existing code until it did what I wanted it to).

Here I document the surprises I found along the way.

If you want a quick start into Fortran, I’d suggest to begin with the tutorial Writing a commandline tool in Fortran and then to come back here to get the corner cases right.

As reference: I come from Python, C++ and Lisp, and I actually started to like Fortran while learning it. So the horror-stories I heard while studying were mostly proven wrong. I uploaded the complete code as base60.f90.

timezones of tccon stations

Timezones of most active TCCON stations in UTC+x (without daylight saving time (DST). Because I needed it and could not find a simple list quickly.

anmyondo: +9,
ascension: 0,
bialystok: +1,
bremen: +1,
caltech: -8,
darwin: 9, # Timezones2008 says 9 1/2???

Equal-Area Map Projections with Basemap and matplotlib/pylab

PDF (read as slides)

Org (reproduce)

Plotting global equal area maps with python, matplotlib/pylab and Basemap.

Table of Contents

information-disbalance creates a power-disbalance

→ a comment to You call it privacy invasion, I don't from Flameeyes.

What you state is a strong version of the “I’ve got nothing to hide” argument. If you’re interested in a thorough debunking, there is a very good article in the chronicle about that: Why Privacy Matters even if you have nothing to hide.

Thank you for your Flattr’s! | Danke für eure Flattr! | Dankon por vian Flattrn!

It’s always a great feeling to see a flattr - Thank you for your support!
You can find new free works for you enjoyment on draketo.de (infrequent and bursty) and 1w6.org (currently weekly but mostly in German).

ArneBab on Flattr

Es ist ein tolles Gefühl, geflattrd zu werden - Danke für eure Unterstützung!

Basic usecases for DVCS: Workflow Failures

If you came here searching for a way to set the username in Mercurial: just run hg config --edit and add
    [ui]
    username = YOURNAME <EMAIL>
to the file which gets opened. If you have a very old version of Mercurial (<3.0), open $HOME/.hgrc manually.

Update (2015-02-05): For the Git breakage there is now a partial solution in Git v2.3.0: You can push into a checked out branch when you prepare the target repo via git config receive.denyCurrentBranch updateInstead, but only if nothing was changed there. This does not fully address the workflow breakage (the success of the operation is still state-dependent), but at least it makes it work. With Git providing a partial solution for the breakage I reported and Mercurial providing a full solution since 2014-05-01, I call this blog post a success. Thank you Git and Mercurial devs!

Update (2014-05-01): The Mercurial breakage is fixed in Mercurial 3.0: When you commit without username it now says “Abort: no username supplied (use "hg config --edit" to set your username)”. The editor shows a template with a commented-out field for the username. Just put your name and email after the pre-filled username = and save the file. The Git breakage still exists.

Update (2013-04-18): In #mercurial @ irc.freenode.net there were discussions yesterday for improving the help output if you do not have your username setup, yet.

1 Intro

I recently tried contributing to a new project again, and I was quite surprised which hurdles can be in your way, when you did not setup your environment, yet.

So I decided to put together a small test for the basic workflow: Cloning a project, doing and testing a change and pushing it back.

I did that for Git and Mercurial, because both break at different points.

I’ll express the basic usecase in Subversion:

  • svn checkout [project]
  • (hack, test, repeat)
  • (request commit rights)
  • svn commit -m "added X"

You can also replace the request for commit rights with creating a patch and sending it to a mailing list. But let’s take the easiest case of a new contributor who is directly welcomed into the project as trusted committer.

dvcs-basic-svn.png

A slightly more advanced workflow adds testing in a clean tree. In Subversion it looks almost like the simple commit:

dvcs-basic-svn-testing.png

Babcore: Emacs Customizations everyone should have

Update (2017-05): babcore is at 0.2, but I cannot currently update the marmalade package. See lisplets/babcore.el

1 Intro

PDF-version (for printing)

Package (to install)

orgmode-version (for editing)

repository (for forking)

project page (for fun ☺)

Emacs Lisp (to use)

I have been tweaking my emacs configuration for years, now, and I added quite some cruft. But while searching for the right way to work, I also found some gems which I direly miss in pristine emacs.

This file is about those gems.

Babcore is strongly related to Prelude. Actually it is just like prelude, but with the stuff I consider essential. And staying close to pristine Emacs, so you can still work at a coworkers desk.

But before we start, there is one crucial piece of advice which everyone who uses Emacs should know:

C-g: abort

Hold control and hit g.

That gets you out of almost any situation. If anything goes wrong, just hit C-g repeatedly till the problem is gone - or you cooled off far enough to realize that a no-op is the best way to react.

To repeat: If anything goes wrong, just hit C-g.

wisp: Whitespace to Lisp

» I love the syntax of Python, but crave the simplicity and power of Lisp.«

display "Hello World!" ↦ (display "Hello World!")
define : factorial n     (define (factorial n)            
    if : zero? n       ↦     (if (zero? n)                
       . 1                      1                      
       * n : factorial {n - 1}  (* n (factorial {n - 1}))))

Wisp basics

»ArneBab's alternate sexp syntax is best I've seen; pythonesque, hides parens but keeps power« — Christopher Webber in twitter, in identi.ca and in his blog: Wisp: Lisp, minus the parentheses
♡ wow ♡
»Wisp allows people to see code how Lispers perceive it. Its structure becomes apparent.« — Ricardo Wurmus in IRC, paraphrasing the wisp statement from his talk at FOSDEM 2019 about Guix for reproducible science in HPC.
☺ Yay! ☺
with (open-file "with.w" "r") as port
     format #t "~a\n" : read port
Familiar with-statement in 25 lines.

 ↓ skip updates ↓

Update (2019-07-16): wisp-mode 0.2.5 now provides proper indentation support in Emacs: Tab increases indentation and cycles back to zero. Shift-tab decreases indentation via previously defined indentation levels. Return preserves the indentation level (hit tab twice to go to zero indentation).
Update (2019-06-16): In c programming the uncommon way, specifically c-indent, tantalum is experimenting with combining wisp and sph-sc, which compiles scheme-like s-expressions to c. The result is a program written like this:
pre-include "stdio.h"

define (main argc argv) : int int char**
  declare i int
  printf "the number of arguments is %d\n" argc
  for : (set i 0) (< i argc) (set+ i 1)
    printf "arg %d is %s\n" (+ i 1) (array-get argv i)
  return 0 ;; code-snippet under GPLv3+
To me that looks so close to C that it took me a moment to realize that it isn’t just using a parser which allows omitting some special syntax of C, but actually an implementation of a C-generator in Scheme (similar in spirit to cython, which generates C from Python), which results in code that looks like a more regular version of C without superfluous parens. Wisp really completes the round-trip from C over Scheme to something that looks like C but has all the regularity of Scheme, because all things considered, the code example is regular wisp-code. And it is awesome to see tantalum take up the tool I created and use it to experiment with ways to program that I never even imagined! ♡
TLDR: tantalum uses wisp for code that looks like C and compiles to C but has the regularity of Scheme!
Update (2019-06-02): The repository at https://www.draketo.de/proj/wisp/ is stale at the moment, because the staticsite extension I use to update it was broken by API changes and I currently don’t have the time to fix it. Therefore until I get it fixed, the canonical repository for wisp is https://bitbucket.org/ArneBab/wisp/. I’m sorry for that. I would prefer to self-host it again, but the time to read up what i have to adjust blocks that right now (typically the actual fix only needs a few lines). A pull-request which fixes the staticsite extension for modern Mercurial would be much appreciated!
Update (2019-02-08): wisp v1.0 released as announced at FOSDEM. Wisp the language is complete:
display "Hello World!"
↦ (display "Hello World!")

And it achieves its goal:
“Wisp allows people to see code how Lispers perceive it. Its structure becomes apparent.” — Ricardo Wurmus at FOSDEM
Tooling, documentation, and porting of wisp are still work in progress, but before I go on, I want thank the people from the readable lisp project. Without our initial shared path, and without their encouragement, wisp would not be here today. Thank you! You’re awesome!
With this release it is time to put wisp to use. To start your own project, see the tutorial Starting a wisp project and the wisp tutorial. For more info, see the NEWS file. To test wisp v1.0, install Guile 2.0.11 or later and bootstrap wisp:
wget https://bitbucket.org/ArneBab/wisp/downloads/wisp-1.0.tar.gz;
tar xf wisp-1.0.tar.gz ; cd wisp-1.0/;
./configure; make check;
examples/newbase60.w 123
If it prints 23 (123 in NewBase60), your wisp is fully operational.
If you have additional questions, see the Frequently asked Questions (FAQ) and chat in #guile at freenode.
That’s it - have fun with wisp syntax!
Update (2019-01-27): wisp v0.9.9.1 released which includes the emacs support files missed in v0.9.9, but excludes unnecessary files which increased the release size from 500k to 9 MiB (it's now back at about 500k). To start your own wisp-project, see the tutorial Starting a wisp project and the wisp tutorial. For more info, see the NEWS file. To test wisp v0.9.9.1, install Guile 2.0.11 or later and bootstrap wisp:
wget https://bitbucket.org/ArneBab/wisp/downloads/wisp-0.9.9.1.tar.gz;
tar xf wisp-0.9.9.1.tar.gz ; cd wisp-0.9.9.1/;
./configure; make check;
examples/newbase60.w 123
If it prints 23 (123 in NewBase60), your wisp is fully operational.
That’s it - have fun with wisp syntax!
Update (2019-01-22): wisp v0.9.9 released with support for literal arrays in Guile (needed for doctests), example start times below 100ms, ob-wisp.el for emacs org-mode babel and work on examples: network, securepassword, and downloadmesh. To start your own wisp-project, see the tutorial Starting a wisp project and the wisp tutorial. For more info, see the NEWS file. To test wisp v0.9.9, install Guile 2.0.11 or later and bootstrap wisp:
wget https://bitbucket.org/ArneBab/wisp/downloads/wisp-0.9.9.tar.gz;
tar xf wisp-0.9.9.tar.gz ; cd wisp-0.9.9/;
./configure; make check;
examples/newbase60.w 123
If it prints 23 (123 in NewBase60), your wisp is fully operational.
That’s it - have fun with wisp syntax!
Update (2018-06-26): There is now a wisp tutorial for beginning programmers: “In this tutorial you will learn to write programs with wisp. It requires no prior knowledge of programming.”Learn to program with Wisp, published in With Guise and Guile
Update (2017-11-10): wisp v0.9.8 released with installation fixes (thanks to benq!). To start your own wisp-project, see the tutorial Starting a wisp project. For more info, see the NEWS file. To test wisp v0.9.8, install Guile 2.0.11 or later and bootstrap wisp:
wget https://bitbucket.org/ArneBab/wisp/downloads/wisp-0.9.8.tar.gz;
tar xf wisp-0.9.8.tar.gz ; cd wisp-0.9.8/;
./configure; make check;
examples/newbase60.w 123
If it prints 23 (123 in NewBase60), your wisp is fully operational.
That’s it - have fun with wisp syntax!
Update (2017-10-17): wisp v0.9.7 released with bugfixes. To start your own wisp-project, see the tutorial Starting a wisp project. For more info, see the NEWS file. To test wisp v0.9.7, install Guile 2.0.11 or later and bootstrap wisp:
wget https://bitbucket.org/ArneBab/wisp/downloads/wisp-0.9.7.tar.gz;
tar xf wisp-0.9.7.tar.gz ; cd wisp-0.9.7/;
./configure; make check;
examples/newbase60.w 123
If it prints 23 (123 in NewBase60), your wisp is fully operational.
That’s it - have fun with wisp syntax!
Update (2017-10-08): wisp v0.9.6 released with compatibility for tests on OSX and old autotools, installation to guile/site/(guile version)/language/wisp for cleaner installation, debugging and warning when using not yet defined lower indentation levels, and with wisp-scheme.scm moved to language/wisp.scm. This allows creating a wisp project by simply copying language/. A short tutorial for creating a wisp project is available at Starting a wisp project as part of With Guise and Guile. For more info, see the NEWS file. To test wisp v0.9.6, install Guile 2.0.11 or later and bootstrap wisp:
wget https://bitbucket.org/ArneBab/wisp/downloads/wisp-0.9.6.tar.gz;
tar xf wisp-0.9.6.tar.gz ; cd wisp-0.9.6/;
./configure; make check;
examples/newbase60.w 123
If it prints 23 (123 in NewBase60), your wisp is fully operational.
That’s it - have fun with wisp syntax!
Update (2017-08-19): Thanks to tantalum, wisp is now available as package for Arch Linux: from the Arch User Repository (AUR) as guile-wisp-hg! Instructions for installing the package are provided on the AUR page in the Arch Linux wiki. Thank you, tantalum!
Update (2017-08-20): wisp v0.9.2 released with many additional examples including the proof-of-concept for a minimum ceremony dialog-based game duel.w and the datatype benchmarks in benchmark.w. For more info, see the NEWS file. To test it, install Guile 2.0.11 or later and bootstrap wisp:
wget https://bitbucket.org/ArneBab/wisp/downloads/wisp-0.9.2.tar.gz;
tar xf wisp-0.9.2.tar.gz ; cd wisp-0.9.2/;
./configure; make check;
examples/newbase60.w 123
If it prints 23 (123 in NewBase60), your wisp is fully operational.
That’s it - have fun with wisp syntax!
Update (2017-03-18): I removed the link to Gozala’s wisp, because it was put in maintenance mode. Quite the opposite of Guile which is taking up speed and just released Guile version 2.2.0, fully compatible with wisp (though wisp helped to find and fix one compiler bug, which is something I’m really happy about ☺).
Update (2017-02-05): Allan C. Webber presented my talk Natural script writing with Guile in the Guile devroom at FOSDEM. The talk was awesome — and recorded! Enjoy Natural script writing with Guile by "pretend Arne" ☺

presentation (pdf, 16 slides) and its source (org).
Have fun with wisp syntax!
Update (2016-07-12): wisp v0.9.1 released with a fix for multiline strings and many additional examples. For more info, see the NEWS file. To test it, install Guile 2.0.11 or later and bootstrap wisp:
wget https://bitbucket.org/ArneBab/wisp/downloads/wisp-0.9.1.tar.gz;
tar xf wisp-0.9.1.tar.gz ; cd wisp-0.9.1/;
./configure; make check;
examples/newbase60.w 123
If it prints 23 (123 in NewBase60), your wisp is fully operational.
That’s it - have fun with wisp syntax!
Update (2016-01-30): I presented Wisp in the Guile devroom at FOSDEM. The reception was unexpectedly positive — given some of the backlash the readable project got I expected an exceptionally sceptical audience, but people rather asked about ways to put Wisp to good use, for example in templates, whether it works in the REPL (yes, it does) and whether it could help people start into Scheme. The atmosphere in the Guile devroom was very constructive and friendly during all talks, and I’m happy I could meet the Hackers there in person. I’m definitely taking good memories with me. Sadly the video did not make it, but the schedule-page includes the presentation (pdf, 10 slides) and its source (org).
Have fun with wisp syntax!
Update (2016-01-04): Wisp is available in GNU Guix! Thanks to the package from Christopher Webber you can try Wisp easily on top of any distribution:
guix package -i guile guile-wisp
guile --language=wisp
This already gives you Wisp at the REPL (take care to follow all instructions for installing Guix on top of another distro, especially the locales).
Have fun with wisp syntax!
Update (2015-10-01): wisp v0.9.0 released which no longer depends on Python for bootstrapping releases (but ./configure still asks for it — a fix for another day). And thanks to Christopher Webber there is now a patch to install wisp within GNU Guix. For more info, see the NEWS file. To test it, install Guile 2.0.11 or later and bootstrap wisp:
wget https://bitbucket.org/ArneBab/wisp/downloads/wisp-0.9.0.tar.gz;
tar xf wisp-0.9.0.tar.gz ; cd wisp-0.9.0/;
./configure; make check;
examples/newbase60.w 123
If it prints 23 (123 in NewBase60), your wisp is fully operational.
That’s it - have fun with wisp syntax!
Update (2015-09-12): wisp v0.8.6 released with fixed macros in interpreted code, chunking by top-level forms, : . parsed as nothing, ending chunks with a trailing period, updated example evolve and added examples newbase60, cli, cholesky decomposition, closure and hoist in loop. For more info, see the NEWS file.To test it, install Guile 2.0.x or 2.2.x and Python 3 and bootstrap wisp:
wget https://bitbucket.org/ArneBab/wisp/downloads/wisp-0.8.6.tar.gz;
tar xf wisp-0.8.6.tar.gz ; cd wisp-0.8.6/;
./configure; make check;
examples/newbase60.w 123
If it prints 23 (123 in NewBase60), your wisp is fully operational.
That’s it - have fun with wisp syntax! And a happy time together for the ones who merge their paths today ☺
Update (2015-04-10): wisp v0.8.3 released with line information in backtraces. For more info, see the NEWS file.To test it, install Guile 2.0.x or 2.2.x and Python 3 and bootstrap wisp:
wget https://bitbucket.org/ArneBab/wisp/downloads/wisp-0.8.3.tar.gz;
tar xf wisp-0.8.3.tar.gz ; cd wisp-0.8.3/;
./configure; make check;
guile -L . --language=wisp tests/factorial.w; echo
If it prints 120120 (two times 120, the factorial of 5), your wisp is fully operational.
That’s it - have fun with wisp syntax!
Update (2015-03-18): wisp v0.8.2 released with reader bugfixes, new examples and an updated draft for SRFI 119 (wisp). For more info, see the NEWS file.To test it, install Guile 2.0.x or 2.2.x and Python 3 and bootstrap wisp:
wget https://bitbucket.org/ArneBab/wisp/downloads/wisp-0.8.2.tar.gz;
tar xf wisp-0.8.2.tar.gz ; cd wisp-0.8.2/;
./configure; make check;
guile -L . --language=wisp tests/factorial.w; echo
If it prints 120120 (two times 120, the factorial of 5), your wisp is fully operational.
That’s it - have fun with wisp syntax!
Update (2015-02-03): The wisp SRFI just got into draft state: SRFI-119 — on its way to an official Scheme Request For Implementation!
Update (2014-11-19): wisp v0.8.1 released with reader bugfixes. To test it, install Guile 2.0.x and Python 3 and bootstrap wisp:
wget https://bitbucket.org/ArneBab/wisp/downloads/wisp-0.8.1.tar.gz;
tar xf wisp-0.8.1.tar.gz ; cd wisp-0.8.1/;
./configure; make check;
guile -L . --language=wisp tests/factorial.w; echo
If it prints 120120 (two times 120, the factorial of 5), your wisp is fully operational.
That’s it - have fun with wisp syntax!
Update (2014-11-06): wisp v0.8.0 released! The new parser now passes the testsuite and wisp files can be executed directly. For more details, see the NEWS file. To test it, install Guile 2.0.x and bootstrap wisp:
wget https://bitbucket.org/ArneBab/wisp/downloads/wisp-0.8.0.tar.gz;
tar xf wisp-0.8.0.tar.gz ; cd wisp-0.8.0/;
./configure; make check;
guile -L . --language=wisp tests/factorial.w;
echo
If it prints 120120 (two times 120, the factorial of 5), your wisp is fully operational.
That’s it - have fun with wisp syntax!
On a personal note: It’s mindboggling that I could get this far! This is actually a fully bootstrapped indentation sensitive programming language with all the power of Scheme underneath, and it’s a one-person when-my-wife-and-children-sleep sideproject. The extensibility of Guile is awesome!
Update (2014-10-17): wisp v0.6.6 has a new implementation of the parser which now uses the scheme read function. `wisp-scheme.w` parses directly to a scheme syntax-tree instead of a scheme file to be more suitable to an SRFI. For more details, see the NEWS file. To test it, install Guile 2.0.x and bootstrap wisp:
wget https://bitbucket.org/ArneBab/wisp/downloads/wisp-0.6.6.tar.gz;
tar xf wisp-0.6.6.tar.gz; cd wisp-0.6.6;
./configure; make;
guile -L . --language=wisp
That’s it - have fun with wisp syntax at the REPL!
Caveat: It does not support the ' prefix yet (syntax point 4).
Update (2014-01-04): Resolved the name-clash together with Steve Purcell und Kris Jenkins: the javascript wisp-mode was renamed to wispjs-mode and wisp.el is called wisp-mode 0.1.5 again. It provides syntax highlighting for Emacs and minimal indentation support via tab. You can install it with `M-x package-install wisp-mode`
Update (2014-01-03): wisp-mode.el was renamed to wisp 0.1.4 to avoid a name clash with wisp-mode for the javascript-based wisp.
Update (2013-09-13): Wisp now has a REPL! Thanks go to GNU Guile and especially Mark Weaver, who guided me through the process (along with nalaginrut who answered my first clueless questions…).
To test the REPL, get the current code snapshot, unpack it, run ./bootstrap.sh, start guile with $ guile -L . (requires guile 2.x) and enter ,language wisp.
Example usage:
display "Hello World!\n"
then hit enter thrice.
Voilà, you have wisp at the REPL!
Caveeat: the wisp-parser is still experimental and contains known bugs. Use it for testing, but please do not rely on it for important stuff, yet.
Update (2013-09-10): wisp-guile.w can now parse itself! Bootstrapping: The magical feeling of seeing a language (dialect) grow up to live by itself: python3 wisp.py wisp-guile.w > 1 && guile 1 wisp-guile.w > 2 && guile 2 wisp-guile.w > 3 && diff 2 3. Starting today, wisp is implemented in wisp.
Update (2013-08-08): Wisp 0.3.1 released (Changelog).

Going from a simple Makefile to Autotools

Table of Contents

Links

Intro

I recently started looking into Autotools, to make it easier to run my code on multiple platforms.

Naturally you can use cmake or scons or waf or ninja or tup, all of which are interesting in there own respect. But none of them has seen the amount of testing which went into autotools, and none of them have the amount of tweaks needed to support about every system under the sun. And I recently found pyconfigure which allows using autotools with python and offers detection of library features.

Warning 2016: Contains some cargo-cult-programming — my current setup is cleaner thanks to using AC_CONFIG_LINKS in configure.ac.

Tail Call Optimization (TCO), dependency, broken debug builds in C and C++ — and gcc 4.8

TCO: Reducing the algorithmic complexity of recursion.
Debug build: Add overhead to a program to trace errors.
Debug without TCO: Obliterate any possibility of fixing recursion bugs.

“Never develop with optimizations which the debug mode of the compiler of the future maintainer of your code does not use.”°

UPDATE: GCC 4.8 gives us -Og -foptimize-sibling-calls which generates nice-backtraces, and I had a few quite embarrassing errors in my C - thanks to AKF for the catch!

1 Intro

Tail Call Optimization (TCO) makes this

def foo(n):
    print(n)
    return foo(n+1)
foo(1)

behave like this

def foo(n):
    print(n)
    return n+1
n = 1 while True: n = foo(n)

Test of the hg evolve extension for easier upstreaming

1 Rationale

PDF-version (for printing)

orgmode-version (for editing)

repository (for forking)

Currently I rework my code extensively before I push it into upstream SVN. Some of that is inconvenient and it would be nicer to have easy to use refactoring tools.

hg evolve might offer that.

This test uses the mutable-hg extension in revision c70a1091e0d8 (24 changesets after 2.1.0). It will likely be obsolete, soon, since mutable-hg is currently moved into Mercurial core by Pierre-Yves David, its main developer. I hope it will be useful for you, to assess the future possibilities of Mercurial today. This is not (only) a pun on “obsolete”, the functionality at the core of evolve which allows safe, collaborative history rewriting ☺

The dynamics of free culture and the danger of noncommercial clauses

NC covered works trick people into investing in a dead end

Free licensing lowers the barrier of entry to creating cultural works, which unlocks a dynamic where people can realize their ideas much easier - and where culture can actually live, creating memes, adjusting them to new situations and using new approaches with old topics.

But for that to really take off, people have to be able to make a living from their creations - which build on other works.

Motivation and Reward

Debunking the myth that you can increase the performance of creative workers with carrot and stick.

Update: Good fixed income and long term contracts are a tool to allow people to work full-time without reducing their motivation. They avoid the harmful effect performance-based payment can have on performance while enabling people to work full-time on a project. An empirical study found, that the source and intensity of motivation of free software developers does not differ significantly between people who work for hire and people who work without payment, so many companies employing free software developers seem to do it right (or only the companies who do it right can keep their free software programmers).1

Executive Summary

For creative tasks, the quality of performance strongly correllates with intrinsic motivation: Being interested in the task itself.

This article will only talk about that.

The main factors which are commonly associated with intrinsic motivation are:

  • Positive verbal feedback which increases intrinsic motivation.
  • Payment independent of performance which actually has no effect.
  • Payment dependent on performance which reduces the motivation on the long term.
  • Negative verbal feedback which directly reduces intrinsic motivation.
  • Threatening someone with punishment which strongly reduces intrinsic motivation.

To make it short: Anything which diverts the focus from the task at hand towards some external matter (either positive or negative) reduces the intrinsic motivation and that in turn reduces work performance.

If you want to help people perform well, make sure that they don’t have to worry about other stuff besides their work and give them positive verbal feedback about the work they do.

Note: In the paper »Why Hackers Do What They Do: Understanding Motivation and Effort in Free/Open Source Software Projects« from 2005, Karim R. Lakhani and Robert G Wolf showed empirically that the payment people get to work in free software projects has no detrimental effect on their intrinsic motivation. In their sample 40% of the developers were paid for their work on free software projects and their intrinsic motivation was as high as the motivation of unpaid developers.


  1. We find […], that enjoyment-based intrinsic motivation, namely how creative a person feels when working on the project, is the strongest and most pervasive driver. The source and intensity of motivation of free software developers does not differ significantly between people who work for hire and people who work without payment. From Why Hackers Do What They Do: Understanding Motivation and Effort in Free/Open Source Software Projects by Karim R. Lakhani* and Robert G Wolf** from the * MIT Sloan School of Management | The Boston Consulting Group and ** The Boston Consulting Group. 

Neither Humble nor Indie Bundle

Update 2016: Later Bundles seem to have gotten better again.

Comment to New Humble Bundle Is Windows Only, DRM Games.

The new Humble Indie Bundle is no longer free, indie, cross-plattform or user-respecting.

When the first bundle had a huge boost in last-minute sales after the devs offered to free the source of 4 of the 5 games, I had hoped, they would keep that.

How to make companies act ethically

→ comment on Slashdot concerning Unexpected methods to promote freedom?

Was it really Apple who ended DRM? Would they have done so without the protests and evangelizing against DRM? Without protesters in front of Apple Stores? And without the many people telling their friends to just not accept DRM?

That “preaching” created a situation where Apple could reap monetary gain from doing the right thing.

Insert a scaled screenshot in emacs org-mode

@marjoleink asked on identi.ca1, if it is possible to use emacs org-mode for showing scaled screenshots inline while writing. Since I thought I’d enjoy some hacking, I decided to take the challenge.

It does not do auto-scaling of embedded images, as far as I know, but the use case of screenshots can be done with a simple function (add this to your ~/.emacs or ~/.emacs.d/init.el):


  1. Matthew Gregg: @marjoleink "way of life" thing again, but if you can invest some time, org-mode is a really powerful note keeping environment. → Marjolein Katsma: @mcg I'm sure it is - but seriously: can you embed a diagram2 or screenshot, scale it, and link it to itself? 

  2. For diagrams, you can just insert a link to the image file without description, then org-mode can show it inline. To get an even nicer user-experience (plain text diagrams or ascii-art), you can use inline code via org-babel using graphviz (dot) or ditaa - the latter is used for the diagrams in my complete Mercurial branching strategy

Inhalt abgleichen
Willkommen im Weltenwald!
((λ()'Dr.ArneBab))



Beliebte Inhalte

sn.1w6.org news

Kommunikation im Internet ist in Gefahr!

Liebe Besucher,
leider müssen wir euch kurz unterbrechen:

Das EU-Parlament stimmt nächste Woche über eine Urheberrechtsreform ab,
die alle kleinen Kommunikationsplattformen in ihrer Existenz bedroht.

Offener Brief der Forenbetreiber

Wir bitten euch daher darum, dass ihr

Vielen Dank für eure Unterstützung!

Euer Draketo

Wir sind Teil des #321EUOfflineDay