“RSS represents the antithesis of this new world [of proprietary APIs]: it’s completely open, decentralized, and owned by nobody, just like the web itself. It allows anyone, large or small, to build something new and disrupt anyone else they’d like because nobody has to fly six salespeople out first to work out a partnership with anyone else’s salespeople.” Jul 3, 2013
“Being interested in both natural and programming languages, I decided to expand on this analysis of profanity in git commit messages per programming language.” Jul 3, 2013
In a Bind: Word Processing in Linux Jul 3, 2013
The year is 2013 and as far as I can tell, there is no good solution for word processing in Linux—at least according to my modest standards. Perhaps this is unsurprising, given that some early Linux FAQs concerning word processors basically suggest that they’re unnecessary nuisances for people unfamiliar with markup languages. And while I’d happily forgo word processors all together, unfortunately, I’d have a hard time getting employers, collaborators, and publishers to agree to the suggestion. And so my search for a good open source word processor is ongoing.
In most contexts and about most things, I am a firm exponent of open source software generally and the Linux ecosystem in particular. But here is one area in which I am flummoxed. All I want are three simple things:
- To open and create MS Word-compatible comments and annotations. I may be willing to work in programs other than Microsoft Word, but I can’t count on my collaborators doing the same. If they mark up a document, or ask me for comments, it’s generally in .doc/.docx format.
- To export in a format acceptable to academic journals in the social sciences. In terms of the text itself, this generally means APA—or occasionally Chicago-style—headings, in-text citations, and references. With regard to filetype, it almost always means .doc or .docx.
- To use Emacs/*nix keybindings. The program should come with the option to use—or I should be able to implement—basic Emacs keybindings without creating horrible conflicts with other keyboard shortcuts.
This last point may seem nitpicky, but part of the joy of Linux for me is the ability to navigate my whole OS nimbly with the keyboard, and Emacs keybindings—along with vim’s—are pretty much as close as Linux comes to universal shortcuts. They’re present, or at least available, in just about every environment where one edits or works with text. And like many people who’ve used programming text editors along the lines of Emacs/vim or spent any length of time on the command line, these efficient keyboard shortcuts have ruined me when it comes to working in a word processor.
Here are the things I have tried, and how they stack up.
Comments + annotations. Until recently, comments and annotations were a bit lousy in LibreOffice, but as of version 4.0 it now has full support for MS Word annotations.
APA formatting. With formatting tools standard to pretty much any word processor and the ability to save in almost every popular document format, formatting for social science journals is no problem for LibreOffice.
Emacs keybindings. Here’s where things get sticky. Marcus Nitzschke has released a configuration file that ostensibly allows you to use simple Emacs keybindings with LibreOffice, but (a) it doesn’t always work as advertised (the CTL-B keybinding for moving back a space doesn’t work for me, even in the latest version of LibreOffice), and (b) there are conflicts all over the place. For example, CTL-A for jumping to the beginning of a line overrides the program’s existing shortcut for “select all,” and CTL-E for end-of-line overrides the program’s default shortcut for centering text. Remapping these other shortcuts only makes matters worse. For example, any alternative key combo that uses the Command/Windows/Super key is likely to conflict with one’s desktop shortcuts, and using the Alt key overrides yet more existing LibreOffice keybindings that have to be remapped in turn. All this gets messy very quickly. I imagine the person who finally and successfully remaps every LibreOffice shortcut to make it compatible with Emacs keybindings will find their online tip jar very full within a matter of hours. Comments + annotations. While it’s on the roadmap, Words sadly does not support MS Word-style comments and annotations.
APA formatting. Like LibreOffice, Words does a fine job when it comes to APA style and the creation of numerous document types.
Emacs keybindings. Admittedly, I haven’t searched very hard, since the annotations problem prevents me from using Words anyhow, but I’ve yet to find a good Emacs keybindings solution for this application. Comments + annotations. AbiWord supports comments, but it has its own unique system for comments and annotations that never seems to play well with MS Word.
APA formatting. Again, AbiWord does a fine job when it comes to APA style and the creation of numerous document types.
Emacs keybindings. Of all the Linux word processors, AbiWord is, surprisingly, the only one I’ve found that has built-in, out-of-the-box support for Emacs keybindings. If its system for comments and annotations were a bit more standard and more compatible with MS Word, I’d happily use it.
Now, you may ask, if I love Emacs so much, why don’t I use it as my word processor? Well, the answer is I’ve tried. Specifically, I’ve tried to use Emacs’ various TeX/LaTeX modes, as well as Org Mode‘s export features. While keybindings obviously cease to be an issue in these scenarios, other annoyances quickly become apparent.
Oh, to be a computer scientist, a mathematician, or a physicist, whose communities happily accept TeX files. Unfortunately, the whole point of TeX appears to be to allow folks who’re comfortable with Emacs and other programming editors to create beautifully typeset documents, while avoiding having to use a word processor altogether. Unfortunately for me, this means that reviewing MS Word comments, or exporting a TeX file to .doc/.docx format is a pain, and—for TeX’s core user base—beside the point. So, even though it’s relatively simple to create an APA-formatted TeX document, it’s hard to commute it into a file format that collaborators or social science journals will accept.
Org Mode is pretty awesome in that it allows you to write entire articles in Emacs and then export them in a format that can be opened in a word processor. It even works with a common citation manager. Unfortunately, it won’t as easily open or allow you to review comments on a document created in a word processor. What’s more, Org Mode’s core function is outlining, and an Org Mode document wants, with all its heart, to be exported in an outline format. It’s theoretically possible to override this behavior by editing the document’s settings, but Org Mode’s word processor compatibility still seems to be in its infancy and I’ve yet to be able to successfully configure a document such that it didn’t need a ton of reformatting after having been exported as a word processor document.
In short, I’m frustrated. If all you’ve ever used is Microsoft Word, you could probably make the jump to LibreOffice without skipping a beat. But once you’re hooked on those damn Emacs/shell keyboard shortcuts, it’s surprisingly hard to live without them.
So what word processor do I use? Sadly, in this one area I’ve forgone Linux and open source. I use Apple’s Pages. Ironically, despite the fact that Apple has gleefully blown up tons of Unix defaults, Pages (and other Apple software) support all the basic Unix keybindings, while open source word processors mostly fail miserably in this area.
If you have the solution here, please let me know. If there’s a Linux word processor that meets my three simple requirements, I’ll be immensely happy to be proven wrong.
My Linux Desktop(s) Jul 2, 2013
You know those days where something about your operating system is bugging you? The inability to drag and drop some piece of information from one place to another, some process that always wants to run in the background, slowing down your whole machine, or some other default behavior that annoys the crap out of you? Whether it’s Windows or Mac, I always encounter these yak shaving moments where a task that seems like it should be simple turns out to be needlessly complicated.
This is why I love Linux. Are Linux desktop interfaces perfect? Hell no. But when they bug me, I can switch between them at will. That’s because under Linux (as with many *nix systems), the graphical desktop interface is handled separately from lower level functions like file storage and device management. So when something bugs me about a particular desktop environment’s behavior, I just toggle over to a different one and continue working with my files and applications there. In fact, I keep three separate desktops on my Linux machine.
First, there’s Openbox, a very speedy, lightweight “window manager” that’s highly customizable and great for things like editing photos or other tasks that require manipulating lots of windows and applets at once:
Second, there’s HerbsluftWM, a “tiling window manager” that maximizes your screen real estate, is good at placing code snippets side by side, and requires very little mouse manipulation, allowing you to keep your hands on the keyboard at pretty much all times. It’s superior for programming tasks, and its ability to obviate use of the mouse probably saved me from a repetitive stress injury to my wrist earlier this year during grading season.
And lastly, for those times when I want an OS X-like “everything’s integrated” experience, I also have Gnome 3. It’s a very powerful and full-featured desktop environment. If Gnome 3 has a downfall, it’s that it’s less configurable than the others. But it’s also the most novice-friendly of the bunch and is straightforward for even first-time Linux users. Plus, since I’ve customized my other window managers to the point that they probably only make sense to me, Gnome 3 is nice to have around for those times when someone asks, “Can I use your computer for a second?”
[Image Credits: The screenshots are obviously mine, but I should mention that both of the visible desktop backgrounds come from Google+ Arch Linux Community user, Kemal E]
Back Up Your Google Reader Data Jun 29, 2013
Via Tiny Tiny RSS forum user cqrt, a solution for making a complete backup of your Google Reader data before the lights go out on July 1st:
This should probably belong in knowledge base but in a few days it will be obsolete. Former Google Reader engineer, Mihai Parparita, has created some python scripts that will enable you to download you entire Google Reader content, including subscriptions, notes, starred items, shared items, liked items, the list of followers, people you were following, the items shared by the people you were following and also including the entire content of the posts from your subscriptions. If you have a lot of subscriptions, make sure you have plenty of storage space available.
William Gibson’s “Pattern Recognition” Jun 29, 2013
I recently finished reading William Gibson’s novel, Pattern Recognition. Published in 2003, it’s a prescient social critique of surveillance marketing, the post-9/11 security state, and the uncomfortable blurring of the two. While reviews of the book when it came out apparently fretted at times as to whether it would seem dated within a few years, the topics it deals with, ranging from the discontents of globalization to NSA monitoring of Internet traffic, are in many places as relevant today or more so than they were a decade ago. A few of Gibson’s more memorable quotes from the book:
The site had come to feel like a second home, but she’d always known that it was also a fishbowl; it felt like a friend’s living room, but it was a sort of text-based broadcast, available in its entirety to anyone who cared to access it.
You ‘know’ in your limbic brain. The seat of instinct. The mammalian brain. Deeper, wider, beyond logic. That is where advertising works, not in the upstart cortex. What we think of as ‘mind’ is only a sort of jumped-up gland, piggybacking on the reptilian brainstem and the older, mammalian mind, but our culture tricks us into recognizing it as all consciousness. The mammalian spreads continent-wide beneath it, mute and muscular, attending its ancient agenda. And makes us buy things.
Everything Lenin taught us of communism was false, and everything he taught us of capitalism, true.
Merriam-Webster on “Irregardless” Jun 28, 2013
I watched the explanation. This word still ticks me off regardless.
“When Unix was created and when it formed its cultural values, there were no end users. Computers were expensive, CPU time was expensive, and learning about computers meant learning how to program. It’s no wonder that the culture which emerged valued things which are useful to other programmers.” Jun 28, 2013
In this essay, Spolsky reviews Eric S. Raymond’s recent book, and examines differences between the cultural values of Windows and Unix programmers and how that translates into different user experiences. I like the essay quite a bit—it has a nice “communitarianism meets software development” vibe to it.
“The problem with most Web video startups is that you’ve seen them before. Another compilation of cheap clips posing as the next big cable channel. Another solution for nonexistent ‘discovery’ problems. Another ‘Instagram for video’ that won’t be.” Jun 27, 2013
Obligatory Google Reader Exodus Post Jun 26, 2013
Since Google announced it would be shutting down Google Reader everyone who makes regular use of RSS, and who’d succumbed to Google’s hegemony over feeds, has scrambled to find a good alternative. I’ve been following other folks’ posts about the various options they’ve chosen, and it’s been interesting to see the diversity in how people make use of their RSS readers. Some simply use them to read content, others use them to share content, and still others curate their own feeds of interesting or useful items.
In making my own switch from Reader, I toyed with a broad variety of tools including but not limited to Thunderbird, NetNewsWire, Pulse, Netvibes, Yoleo, and Feedly. There were so many directions to go that I ultimately had to take a step back, take a break from all the marketing messages enshrouding each service, and ask myself what I wanted out of an RSS reader.
Here’s what I decided I needed:
- Access from anywhere. I switch frequently between my personal Linux machine and my OS X work computer, and I wanted access to my feeds from both devices. That ruled out most desktop applications, and particularly platform-dependent ones. I figured it also wouldn’t hurt to be able to access my reader from my tablet or other devices on the go.
- Ability to publish and curate my own feeds. Over the years, I’ve greatly appreciated the ability to publish my own RSS feeds made up of curated links from items in my reader, as well as self-authored items. I’ve used curated feeds with my classes and also as a way of feeding custom content into web development projects. Once upon a time, before I fell into Google Reader, I used an excellent open source web application called reBlog for curating feeds, but that project was orphaned and it unfortunately broke with the introduction of PHP 5.
- Control over my data. If ever the service or application ceased to exist, I wanted to be able to leave with my subscriptions and all my data intact.
- Extensibility and control over the software. I’d prefer some level of control over upgrades and functionality, as opposed to being one of the lowly chorus of users who complains helplessly about changes to a service (e.g., “I don’t like the new Facebook/Gmail/Twitter/etc.”).
After looking at a bunch of different options and consulting with my inner Dave Winer, I finally decided on Tiny Tiny RSS (TT-RSS for short). TT-RSS is a self-hosted web application, meaning I run it via my own server. This means my data will always belong to me, and I can export it at any time in any format I need. And since it’s web-based, I can access it from anywhere on any device.
It’s an open source application and it has an extensible plugin architecture, meaning that where necessary I can modify it to suit my needs, as well as take advantage of tweaks provided by other users. It’s written in PHP, which for better or for worse is the server-side scripting language with which most of my core competencies lie.
And, last but not least, it has some very flexible tools and methods for publishing and curating your own RSS feeds. TT-RSS’ interface also includes a bookmarking feature for adding and remembering content from anywhere on the web, and as such I find that it’s rapidly supplanting my need for tools like Del.icio.us or Scuttle.
So, that’s it. It’s the solution that worked for me. I’m not planning on running a hosting service, but if you’re a friend or colleague who’s interested in TT-RSS, but not quite technical enough to install it yourself, I’d be happy to create an account for you on my server. Just email me and I can get you set up.