Irreal

Marcin Borkowski (mbork) always has interesting posts that often give me good ideas. Recently, he posted on using Org-mode for SQL. I haven’t had to deal with SQL for a long time and I was never a heavy user so any complicated query always took me multiple tries. If, like me, you’re an interactive programming aficionado, what you’d like to do is to be able to build up your query piece by piece. SQL certainly doesn’t lend itself to interactive programming but Borkowski found a way. Actually, it’s pretty obvious once you see it.

His idea is to execute the query from inside an Org Babel block. He can continually execute and refine the query until he gets what he wants. The nice thing is that your last attempt doesn’t go away. It’s still there to edit and try again.

Take a look at Borkowski’s post for the details and a worked example. The post is worth reading if only to see his definition of “toy editor.”

Mike Zamansky is back with another video in his Using Emacs Series. This video discusses maintaining a simple Web site with Emacs and Org-mode. Usually when I write about this sort of thing, it’s in the context of publishing to a blog. Zamansky’s need is a bit different. He has a couple of Web sites he maintains that support some of his programs at Hunter College. These are simple sites that Zamansky had been writing directly in HTML by logging into the host machine and editing the file in situ.

Then he realized that rather than logging onto the host machine to edit the files, he could use Tramp and do his editing from the comfort of his own environment. That’s already an improvement but then he wondered why he was going through the pain of writing in HTML when we could write in Org-mode and export to HTML. The default look of the HTML exported from Org isn’t very pretty but it’s easy to disable the default CSS and add your own. You can even embed custom CSS in the Org source file if you need something special for that page.

Zamansky’s final realization was that rather than export to HTML and then transfer the HTML to the host machine, he could simply use Org-publish and have the whole process automated. All he needed to do was write the Org source and call Org-publish to export it to HTML and transfer it to the host machine. His solution is much like Sachin Patil’s that I wrote about last Friday. It’s a nice work flow that eliminates the boring rote work and lets him concentrate on the site’s content.

Follow the link to the video for more details on his set up and how he arrived at his solution. The video runs 16 minutes, 20 seconds so plan accordingly.

If you’re an Emacser and a blogger, you will, of course, want to write and publish your posts from within Emacs. If you’re using WordPress, org2blog provides an excellent solution. You write your posts in Emacs and Org-mode as usual and push a couple of keys to publish them to WordPress. Of course, not everyone likes WordPress and may prefer to maintain a website with static posts. In many ways that’s a lot simpler and if you don’t have complex requirements may be the best solution.

Over at Opensource.com, Sachin Patil has a nice post that explores publishing a blog completely from within Emacs. There are plenty of solutions that are sort of like this that use Django, Hugo, or Jekyll but Patil wanted a solution that required only Emacs. Not surprisingly, Emacs has a solution for that. The solution is to write your posts in Org-mode, leveraging all its formatting capabilities, and then use Org-publish to convert the posts to HTML and push then to the server.

There are ways to include and publish a site CSS file to make your site your own. Patil covers all this in his post. If you’re looking to start a blog and would like to handle everything in Emacs, take a look at Patil’s post. It’s amazing how much one can accomplish with just Org-publish. The nice thing about this approach is that you have total control over how you publish to your site and what the end result looks like.

I’m currently reading Sönke Ahrens’ How to Take Smart Notes, a book that describes Niklas Luhmann’s method of taking, storing, and retrieving notes about ideas he thought could be useful later. Luhmann kept his notes on what amounted to index cards and stored them in a wooden box. The idea, Zettelkasten in German, is usually translated as “slip-box” in English. I first came across the idea in this review and summary of Ahrens’ book by Tiago Forte.

Implementing Luhmann’s idea with physical index cards is still possible, of course, but most of us would prefer to use digital methods. There’s a whole site, Zettelkasten, supporting the idea as well as commercial applications such as Roam to help you implement the system.

As soon as I heard about the idea, I thought it would be an ideal application for Emacs and Org-mode. That’s hardly an original insight, of course, and a quick search yielded two excellent posts that discuss it. The first, by Jethro Kuan, describes his note taking work flow using the org-roam package that he wrote as a way to replicate Roam with Org-mode. Most of the post is a sort of précis of the ideas presented in Ahrens’ book.

The other post, How to Make Yourself Into a Learning Machine, describes the system used by Simon Eskildsen, Director of Production Engineering at Shopify. Eskildsen doesn’t use Org—he’s a Vimer—but he’s built his own Zettelkasten. I liked his post because he gives examples of actual notes and how he links them together.

Kuan points to a nice video by Nat Eliason that shows how he uses his (Roam-based) Zettelkasten to pull together an outline for an article he’s writing. It’s very enlightening and shows the power of the Zettelkasten method. It’s 15 minutes, 12 seconds long and definitely worth watching if you’re interested in the Zettelkasten idea.

I’m waiting to finish Ahrens’ book before I try my hand at using the method. It seems pretty straightforward to implement in Org but Kuan found he needed more so I’ll take a closer look at org-roam before I commit any serious effort to the project.