Where I outline the design of tags in Pijul, which are a first step towards a more general hybrid snapshot/patch version control system. The Sanakirja tricks presented here can also be applied outside the context of Pijul, to create efficient compressed databases.

It was a wild year for the database industry, with newcomers overtaking the old guard, vendors fighting over benchmark numbers, and eye-popping funding rounds. We also had to say goodbye to some of our database friends through acquisitions, bankruptcies, or retractions.

As the end of the year draws near, it’s worth reflecting and taking stock as we move into 2022. Here are some of the highlights and a few of my thoughts on what they might mean for the field of databases.

It seems as though an irrational fear of the mouse dominates amongst Un*x dogmatists. Be it because their tools don’t integrate, or because it is a symbol or rebellion, a means to differentiate their prior ignorance from the enlightenment they have attained between the standard input and the standard output stream, the self-emancipation from the mouse is a rite of passage or a shameful burden on those who fail1 2.

As might be clear by my tone, I have little regard for this position. It certainly is worthwhile knowing that and how it is possible to avoid using the mouse, when keyboard shortcuts more convenient3, but there is no need for the artificial asceticism of ready-to-use technology, for the sake of purity and dogma.

The denunciation of the mouse usually involves invoking concepts such as the “home row”, or the cumbersome migration of the hand between keyboard and mouse. These might all be well and good, if I were a typist and as such all I did was to type. But this isn’t the case, I ponder and perceive, more than I write.

When debugging code or studying a program, it is quite comfortable to depend on only one hand, and have the other free to scribble or relax. The issue isn’t using a mouse or only using a keyboard, it is context switching.

Read More

It took little under a decade for the headline feature developer Jon Skinner added to Sublime Text’s second version to become one of the defining features of this decade’s software.

“Goto Anything” is how it started, a search pane to jump to other files. Open a folder, press CMD/Ctrl+P, start typing to see a list of matching files, then press Enter to jump to it. Seems simple enough: You think of a file you need, and without leaving the keyboard can switch to that file and continue work.

Within months, that search pane gained a companion: The now-famous Command Palette. “The Command Palette provides a quick way to access commands that don’t warrant a key binding, and would usually be hidden away in a menu,” explained Skinner. This time, you’d press CMD/Ctrl+Shift+P and get a search bar, only here you’d search through program features.

Read More

I’ve been lurking the fish shell for a couple of years now (and the nushell but it is another story for another time). Not so long ago, I decided to try it, and it’s simply… amazing. If I had to state one feature that makes me like to use it, it’ll be the autocompletion, hands down. It’s the first time I just take a shell and without customization it’s pleasing to use.

During an idle Sunday evening, a few weeks ago, I played with JetBrains’ Projector: it’s IntelliJ — one of the most powerful IDEs around — installed on a remote server and accessible through the browser.

I wanted to see how I might offload some heavy computing tasks, like training and evaluating large machine learning models, off of my local computer, without compromising my coding user experience.

It’s a surprisingly easy setup to build. It took me less than an hour, from hitting “Create instance” on AWS to having a fully-fledged AI project running on my iPad Pro.

It felt like magic.

Read More
Nov 6, 2021

Emacs grew support for displaying colour emojis recently (and this is included in the release branch, which will become Emacs 28.1 in some months). This includes support for the grapheme cluster emojis (that consist of a number of Unicode code points, joined together with zero-width joiners and magic).

So finally Emacs can display all of these things. You can thank Robert Pluim, who did all the actual work.

Imagine if someone summoned a magical genie and wished for a perfect code editor. Since it is perfect, does that mean it provides you everything you ever need to code the optimal solution? Or since it is perfect, does it enable you to accomplish the coding aspect instantly?

Thus, the paradox:

Does the perfect code editor mean that you spend nearly 100% of your work time using the editor or does it mean you spend nearly 0% of your work time using the editor?

What metric can we even use to measure the perfect code editor? How will we know if and when we have it? Are we close to reaching that point?

Read More
Sep 20, 2021

Git is hard: screwing up is easy, and figuring out how to fix your mistakes is fucking impossible. Git documentation has this chicken and egg problem where you can’t search for how to get yourself out of a mess, unless you already know the name of the thing you need to know about in order to fix your problem.

So here are some bad situations I’ve gotten myself into, and how I eventually got myself out of them in plain english.

Here’s what I learned from starting to read the source code to fff - in particular, the main function.

fff is “a simple file manager written in Bash”. As I’m always on the lookout to learn more about Bash, that description got my attention immediately. It’s a small but perfectly formed offering, complete with man page and even a Makefile for installation. And the file manager executable* itself is a single Bash script.

*I use this term deliberately, and it does make me stop and think every time I see scripts in a bin directory (where “bin” stands for binary). But that’s a conversation for another time.

The author, Dylan Araps has produced other interesting pieces of software (such as neofetch) as well some great documents such as the pure bash bible as well as the pure sh bible. He’s also the creator of Kiss Linux. He has a reputation for writing great Bash code, so this seems like an opportunity too good to miss to learn from better programmers.

It seems that recently Dylan has disappeared off the radar, I don’t know what the situation is but I wish him well.

Anyway, I wanted to take a first look at fff to see what I could discern. I’m reviewing the code as it stands at the latest to-date commit, i.e. here.

Where I can, I link to reference material so you can dig in further to Bash details that take your fancy. This reference material includes the following sites (and there are more of course):

Read More

Many people wonder how Emacs Org mode is used in “the right way”.

Here is my short answer to that: There isn’t one right way for using Org mode features.

If this is obvious to you, you can skip the rest of this article. If not, you really should read this until the end. It may answer many questions you probably do have in your head. Especially when you’re rather new with the Org mode universe.

Of course, the same rationale is true for all kind of advanced software solutions that offer a certain level of flexibility to its users. However, my examples here are based on Emacs Org mode.

Read More
Aug 12, 2021

The concept of the staging area is one which many newcomers struggle with when starting to learn git, and the fact that this concept is barely mentioned in the official documentation (i.e. man pages) doesn’t help at all. Instead the official name is “the index” which has absolutely nothing to do with the way users interact with it. That’s the reason most people who teach git prefer to use the term “staging area” instead, and in fact that’s the term used in the best available documentation: the Pro Git book (which is not official), and also pretty much in all online documentation, including tutorials and blog posts (e.g. Atlassian saving changes, code refinery).

Attempts to move away from the incorrect term “the index” towards one that most native and non-native English speakers can grasp without the need for further explanation have been attempted for more than a decade, and even though there’s universal consensus that “staging area” is by far the best alternative, attempts to use it officially in the documentation and user interface have been blocked because one person cannot be convinced.

Read More

There’s a keyboard layout I’ve been using for the past 8 or 9 months. It’s called the workgirl layout. It may look familiar to some of you:

By no coincidence, this layout happens to have the exact same key placement as the workman keyboard layout. Rest assured though, it is not the same layout.

Oh, you’re still wondering what the difference is? The name, silly!

If you spend all day looking at code, letters, and characters—why not make it fun?

It is a well-known “fact”, in some circles, that running grep in the C locale is much faster than in UTF-8 locales, the latter being a common default on current client systems.

Indeed, just the other day, a colleague of mine was running something akin to:

grep ‘“event-type-id”:4727’ app.log

where app.log is a multi-gigabyte file of JSON lines1, the goal being to quickly see whether a newly-added type of event is seen. Almost immediately, someone jumped in to recommend running grep in the C locale, instead of the default UTF-8 locale, for Guaranteed Extra Speediness™.

I wondered whether this was indeed good advice. (TL;DR: using the C locale does not help performance in general, or jump to conclusions.)

Read More
Jul 12, 2021

Alf is a modern, powerful implementation of relational algebra. It brings relational algebra where you don’t necessarily expect it: in shell, in scripting and for building complex software. Alf has an rich set of features. Among them, it allows you to:

  • Query .json, .csv, .yaml files and convert from one format to the other with ease,

  • Query SQL databases with a sounder and more powerful query language than SQL itself,

  • Export structured and so-called “semi-structured” query results in various exchange formats,

  • Query multiple data sources as if they were one and only one database,

  • Create database viewpoints (mostly read-only viewpoints for now), to provide your users with a true database interface while keeping them away from data they may not have access to,

  • Enjoy a rich set of relational operators and even define your own high-level and domain-specific ones.

Read More
Programming Tools

This is where all the vim VS emacs posts should go.

Created on Oct 19, 2020
By @gurlic