Eric MacAdie

I Now Own Only Linux Machines

by

About nine months ago I got a small desktop computer running Linux from a company in Colorado named System76. I got the Meerkat. I still have to use Windows for work, but after about 21 years of either dual-booting or having one each of Linux and Windows, my machines are now free of Microsoft.

There is a saying that BSD is for people who like Unix, and Linux is for people who hate Windows. Either way works for me. One reason I went with System76 is that I wanted to buy something with Linux pre-installed [Note 1]. I think it is important to send a market signal that there is demand for Linux. Yes, I could buy something with Windows on it and install Linux. But I am tired of installing OSes. And it does not seem as easy as it used to be. I know that Unified Extensible Firmware Interface (UEFI) has replaced BIOS. I think I do not quite get the difference between a BIOS and a UEFI, and I do not care. I do not want to spend my time with low-level stuff. I have read some articles and comments that UEFI makes it harder to install Linux (or at least it did: see articles here and here). I think for a while Microsoft was the only UEFI signing authority.

Some people have suggested that I go for Apple. From one proprietary vendor to another. How much money does MS get for a laptop with Windows pre-installed? Maybe $50 or $100? Giving Apple an extra $1000 to not give MS $100 does not seem smart, especially when Apple is not what I really want anyway.

I also wanted to get away from Ubuntu. So far I have liked it, but I do not like working with snaps (see Hacker News posts here, here, here and here). They use a lot of disk space, sometimes they use up a lot of memory, and they tend to update whenever they want and not what the user wants, sort of like Windows. System76 has an Ubuntu derivative named Pop!OS (page here, subreddit here; and yes, the exclamation point and underscore are part of the name) and it does not use snaps. Another Ubuntu derivative named Linux Mint has also dropped snaps (HN threads here and here). If I had to install, I think I would go with Mint.

Before this I had two laptops: one I got used in 2017 from a local vendor named Discount Electronics that I installed Ubuntu on that I used as my main system, and one that I got in 2014 running Windows 8 that I mostly used to collect podcasts and watch videos. The Windows laptop was getting slower and slower, the battery stopped working a long time ago, and it would update whenever it wanted. When it tried to update, I used to try to tell it to update in 5 or so hours, but then it would bug me again five minutes later. Typical Windows: doing what it wants, and not what I want. At some point, it took so long to start up I thought that it was only a matter of time until it died and I should replace it soon.

I looked around for a few vendors. I asked on a local list about minis [Note 2] and laptops. I mentioned System76 and the Librem Mini by Purism. Someone else mentioned minis by Simply NUC. Other Linux vendors include KFocus by MindShare, Tuxedo Computers, Raptor Computing Systems (their systems run Linux on Power chips and are very expensive) and Starlabs Systems (their systems do not contain dark matter). Lenovo sells system in the US, but a lot of people do not trust them. Although I did not go with Ubuntu, they do have a list of hardware certified to run Ubuntu; it is a good resource to have if you want to install something a distro derived from Ubuntu.

I like the Meerkat. I had to buy a USB port because it did not have enough ports for me (sometimes I hook in a flash drive and headphones while my mouse and keyboard are hooked in). I also had to get an adapter to go into the USB port because the audio jack was not compatiable with my speakers. It seems to add OS updates more frequently than I expected (not necessarily kernel updates, but other packages it labels as part of the OS), and I am not able to make a few small configurations that I can on my Ubuntu machine. For some reason there are two Emacs installations on the system. And it still comes with vi installed. But other than that, I am happy with it and I would recommend it to anyone who wants a Linux system. I like the fact that OS updates might require a restart, but I can put that off as long as I want.

To sum up, I like System76. As ads would say in the days of yore, they are purveyors of fine Linux machines, and make systems that can handle your computing needs.

You’re welcome.

Note 1: I posted a question asking for recommendations for Linux machines. I said I wanted to buy something pre-installed. Someone asked me why. Why not? I am tired of installing OSes. I do not want to get something to have an excuse to install an OS. I have done more times than I wanted to, I never learned anything from it, and I am tired of doing it. I realize that somebody has to know how to do it. That does not mean everybody has to know it.

In the past I have asked people if there are routers with Linux pre-installed, mentioning that while I could install Linux, I have no desire to do so. Almost every time, some idiot says, “Why not install OpenWrt?” Because I am sick of installing OSes. If someone says, “I do not want to do X”, and you suggest, “Why not do X?”, you are not insightful, you are not funny, you are just being a jerk. Or at best obtuse

Note 2: Some jackass replied “Eric, you’ve got to make sure you keep your terms proper and straight. A mini-computer is one step up from a PC, and the next step to a main frame. These are very much like running a very big system.” First off, Dick, we are not friends. I have no idea who you are. Do not refer to me on a mailing list by my first name. Your parents must have seen the future when they decided to name you “Richard”. Secondly, the vendors call them minis. That works for me.

Most of the responses were constructive. Just not that guy.

For a community that talks a lot about freedom, some Linux people have very rigid ideas about how other people are supposed to act.

Image of the women at the tomb from Poussay Gospels, a 10th century manuscript now housed in the Bibliothèque Nationale de France, image from Biblissima Portal, assumed allowed under Public Domain. Telling people that death has been conquered seemed like an appropriate image.

2022-11-20: New Stuff I Learned: Grep and Databases

by

I recently learned a couple of new tricks, and I want to share a practice I engage in that I have gotten some compliments on.

I have worked on a few Java web applications, and instead of searching with the IDE, I use grep in cygwin. Sometimes the term I search for is not just in a Java file, but it is also in a Javascript file that has no line breaks. (Seriously, why do people make files like this?) This makes the results harder to read. One possibility is to pipe the results of the grep to another grep:

grep -riIn someString * | grep -v '.min.js' | grep -v '.js.map'

The disadvantage of this method is you lose the color in the output. I read the man page, and I found there is an exlusion flag: –exclude. Even better, you can use it more than once:

grep -riIn --exclude=*min.js --exclude=*.js.map someString *

I also found out the -I option can exclude binary files, which were also polluting the output.

The database we use is Oracle. I started using the Explain Plan to analyze some queries. You can significanly reduce the cost of a query by using upper to compare strings. Going from this:

where some_column like 'some string%'

to this:

where upper( some_column ) like upper ( 'some string%' )

reduced the cost of a few queries from 20,000 to 40,000 to between 30 to 40. I thought that using “upper” would take longer, since it has to run that comparison on the whole table, and that using equals would be faster, but using “upper” is faster. I do not know if this also holds true for other databases. On Oracle “upper” seems to be faster than “lower”.

Another thing I want to mention is that I have gotten a few compliments on how I format insert statements. I line the column names up, and I put in comments in the values() part of the statement for the name of each column.

You have probably seen insert statements like this:

insert into us_state( state_id, state_name, capital,largest_city, abbreviation, population, median_age, persons_sq_mile,nickname, area_sq_mile, avg_elevation, north_latitude_deg, north_latitude_min, south_latitude_deg, south_latitude_min, east_longitude_deg, east_longitude_min, west_longitude_deg, west_longitude_min, created_date, created_by, modified_date, modified_by)
values (func_to_get_next_id(), 'Illinois', 'Springfield', 'Chicago', 'IL', 12882135, 39, 231, 'Prairie State', 57915, 600, 42, 30, 36, 58, 87, 30, 91, 31, sysdate(), user_id, sysdate(), user_id);

I have some that are worse than that. Imagine that statement with no spaces after the commas.

Compare that to this:

insert into us_state( 
    state_id,           state_name,         capital,            largest_city,       abbreviation,       population,         
    median_age,         persons_sq_mile,    nickname,           area_sq_mile,       avg_elevation,      north_latitude_deg, 
    north_latitude_min, south_latitude_deg, south_latitude_min, east_longitude_deg, east_longitude_min, west_longitude_deg, 
    west_longitude_min, created_date,       created_by,         modified_date,      modified_by 
) values (
    func_to_get_next_id(), -- state_id
    'Illinois',            -- state_name
    'Springfield',         -- capital
    'Chicago',             -- largest_city
    'IL',                  -- abbreviation
    12882135,              -- population
    39,                    -- median_age
    231,                   -- persons_sq_mile
    'Prairie State',       -- nickname
    57915,                 -- area_sq_mile
    600,                   -- avg_elevation,
    42,                    -- north_latitude_deg
    30,                    -- north_latitude_min
    36,                    -- south_latitude_deg
    58,                    -- south_latitude_min
    87,                    -- east_longitude_deg
    30,                    -- east_longitude_min
    91,                    -- west_longitude_deg
    31,                    -- west_longitude_min
    sysdate(),             -- created_date
    user_id,               -- created_by
    sysdate(),             -- modified_date
    user_id                -- modified_by
);

It is a little more typing up front, but it is less thinking later on [1].

I also reformat numbers in my notes. I keep notes as I work on stories/defects. If I get an id or a long number from a database, I will copy it, and then add dashes every third number:

4198275630/419-827-563-0
5832064197/583-206-419-7

Years ago I worked on a project where the IDs were a string based on the date down to the nanosecond with no dashes or separators. So I was dealing with very long strings of about 30 characters, half of which I did not need. I got tired of parsing long numbers in my head. More typing means less thinking. The fewer plates I have to spin in my head, the better. Granted, a lot of things that could reduce my coginitve load are usually out of my control, but I change what I can change.

You’re welcome.

Note 1: The formatting is lined up in the WordPress editor; sometimes it looks crooked in the final version. Things like this are why I sometimes think about going with a static site.

Image from Golden Gospels of Henry III, aka Codex Aureus of Speyer or Speyer Gospels (Speyerer Evangeliar), an 11th-centry manuscript housed in the Royal Site of San Lorenzo de El Escorial; image from Wikimedia assumed allowed under public domain.

Learning Languages Test-first and Project-first

by

Sometimes I wonder why don’t books and tutorials teach programming languages closer to the way web apps are taught and made: By making projects up front, and by incorporating tests right away.

I am going through some Pluralsight tutorials for Go and Elixir, and while there is a lot of information in them, in most of them the code is not well organized. Only one of the dozen or so Go tutorials that I have been through makes a project. Why not do that every time? That is how we make web apps, especially since after Rails most frameworks are created with generators. You can make projects with most languages, but for some reason it is not emphasized. Granted, Go is a language where you can run code outside of a project: You can just put code into “some_file.go”, and then type “go run some_file.go”, but in my opinion it is a bad practice.

Several years ago, I went through a Rails bootcamp. After I was done, another alumni and I started a small study group to go through the RSpec book by David Chelimsky et al. We only had a few meetings, and I do not remember how far in the book we got. We thought that testing was not covered enough in the bootcamp, and we thought it was odd that some languages (like Ruby) have a strong “test-first” culture but when it comes to learning the language tests are treated like an afterthought. If we tell people to write their apps tests-first (or at least simultaneously with their code), why not teach languages test-first? A lot of books wait until halfway through before talking about tests and/or projects, or leave it until the end.

Even Dave Thomas holds off on projects and testing. I thought one of the guys famous for being part of the agile/extreme/TDD movement would start out with tests right away. The pickaxe book and his Elixir book do not get to them until halfway through; it is chapter 13 out of 25 in the Elixir book. Chapter 13 starts with: “Let’s stop hacking and get serious.” Why not start out serious?

It also might help in learning a language. If you are going through a tutorial, I think it is better to have the different sections as files in a project as opposed to just one-off files or snippets floating in space.

It might seem like it is too much for beginners. But what is another chapter in a 200+ page book? And people should know it anyway. If we want people to be competent professionals, they should learn the proper way to do things from the beginning. I think we need to get rid of this idea that a smart developer can learn a language in a weekend. Yes, syntax can be picked up quickly. But to really be good at a language, you need to know the libraries, frameworks and tools. And how to structure your code so that other people can understand and use it.

To introduce people to projects and tests, a tutorial could make a package (or namespace or whatever your language calls them) that just has a few functions/methods that take a couple of numbers, sums them, and adds one to the sum. And another method that subtracts one from the sum. And you could have another package that does the same for multiplication.

I might re-do some of the Pluralsight tutorials by making them into projects. I noticed there is now one for the Gin web framework. Perhaps someone at Pluralsight read my post a few months back on things I do not like about Javascript. They have a tutorial on Go web services that includes a massive Angular app; it takes up more than 500 MB on my hard drive. I wrote it would have been easier (and better for learning Go) to just make a web app in Go.

I was looking for a way to make a project with Go, when I found a video that creates a project while using VS Code. This tutorial follows a page on the Go site. One thing I do not like about Go is that “go test” does not run tests in all the sub-directories by default. Maybe it does and I am just not as familiar with it. But “mvn test” or “gradle test” or “lein test” will run all the tests.

Keep in mind, I am not a TDD purist. I do not always write failing tests first. I have no issue with writing tests after code; Ken Kousen calls it GDD: Guilt Driven Development. I also have no problem with tests hitting the database. A presenter at Austin Ruby years ago pointed out that you can get mocks and stubs to do what you want. If your code has a lot of mocks and stubs, are your tests telling you that your code does what you think it should do, or are they telling you that you are good at mocking and stubbing? Some TDD people say it is fine to use an in-memory database. Why not use an actual database? You are still going outside your code either way.

If you are a TDD purist, do not bother arguing with me. You will not change my mind, and NOTHING ever changes the mind of a purist. Not even the fact that purists are more likely to turn people off to something than turn them on to it. I announced my Groovy Email Server on a list somewhere, and someone got pissy with me because the first thing they did was run tests, and they failed because there was no database connection. Even though the README clearly states that you need to set up a database first. Like Will Rogers said, some people read the sign, and some people just gots to touch the electric fence.

You’re welcome.

Image from León Antiphonary, a 10th-century manuscript hosted at Virtual Library of Bibliographic Heritage, image allowed under Creative Commons CC BY 4.0.

2022-11 Austin Emacs Meetup

by

There was another meeting a week ago of EmacsATX, the Austin Emacs Meetup group. For this meeting there was a predetermined topic.

The topic was “Mastering Crafted Emacs”. The speaker was one of the developers in OKC (not the professor), aka The Artist Formerly Known As #3. He gave a more formal talk similar to the off-the-cuff one he gave the previous month on Crafted Emacs, particularly a module implementing configuration options covered in the book Mastering Emacs.

Fourteen people were on the call. We did not go around the horn and introduce ourselves. I did look at a few of the Meetup profiles of the respondants. I think we have a person in India, one in Germany, and one in Nigeria. So now all we need is someone from South America and Australia, and our global dominance is complete. A few people dropped off during the presentation; for the international members, it was very early or very late.

This time, he presented the information in a different order. He went over the basic concept of Crafted Emacs (not as complete as Prelude or Doom, but not leaving you high and dry either; you can still do it yourself without having to do it all yourself). He talked about Mastering Emacs, and again recommended that everyone get it. Even an experienced Emacs user can learn something, and you get free updates for every Emacs release as long as he keeps updating it.

The module can be found here on Github. It provides completion, window configuration, buffer switching customization, and turns on the following modes: winner-mode, flyspell-mode, ibuffer, and uses hydra to configure dumb-jump. He made a VM on his machine, and started with a fresh install of Emacs. He then added Crafted Emacs, and enabled different parts of the file and demonstrated what capabilities it gave him.

There was some general discussion after the presentation. Most of the questions were answered by The Artist Formerly Known As #3. Someone mentioned Tree Sitter (site here, Github repo here, Emacs module site here, Emacs module Github repo here). Per its website, Tree-sitter is a parser generator tool and an incremental parsing library. It can build a concrete syntax tree for a source file and efficiently update the syntax tree as the source file is edited. It can be used for syntax highlighting, and it can prevent your syntax highlighting from breaking when you start typing. Most language modes would need to be re-written since they currently use regular expressions. It would allow Emacs do for other languages what Paredit does for the Lisp family: you can hide sections of the code, like loops and if/else branches, instead of just folding methods or functions. Combining it with Language Server Protocol could make Emacs more competitive against IDEs. Quite a few members were excited about it. I thought it was another example of other languages finally being able to do something that Lisp could do for years, which is a good summary of the past 40 years of computer science.

Another member asked about require and use-package. The basic idea is that use-package is a macro that calls require, and throws in some housekeeping like error-handling, as well as installing a package on your system if it is not already there. Unlike a package manager, use-package will not keep packages up to date.

A couple of members also talked about corfu and company, which both handle completion. There is an article comparing them here. One member said that understanding search and search results can make you faster if you are good at them.

You’re welcome.

Don’t forget: EmacsConf 2022 is happening online on 2022-12-03 and 2022-12-04.

I give people numbers since I do not know if they want their names in this write-up. Think of it as the stoner’s version of the Chatham House Rule. I figured that numbers are a little clearer than “someone said this, and someone else said that, and a third person said something else”. Plus it gives participants some deniability. People’s numbers are based on the order they are listed on the call screen, and the same person may be referred to by different numbers in different months.

I am not the official spokesperson for the group. I just got into the habit of summarizing the meetings every month, and adding my own opinions about things. That said, if you like something in this post, I will take credit; if you don’t, in this case you get what you pay for.

Image from MS. Auct. E. 5. 11, a 10th century manuscript housed at the Bodleian Library, University of Oxford; image under Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)

2022-10 Austin Emacs Meetup

by

There was another meeting a week ago of EmacsATX, the Austin Emacs Meetup group. For this meeting there was no predetermined topic.

#1 was one of the organizers, and is active in the Austin Clojure group.

#2 was one of the developers in OKC (not the professor), aka The Artist Formerly Known As #3.

#3 was one of the organizers, and formerly worked for the City of Austin.

#2 did most of the talking. There were two general themes: Crafted Emacs and the book Mastering Emacs. He recommended that everyone read Mastering Emacs. When he read it he had been doing Emacs for thirty years, and he still learned something from it, even though it is geared towards beginners. I think I was the only one who had not read it. The group thought that it would be great if the book talked about Org Mode, but it is still worthwhile. The book explains how to get information from Emacs itself, and on customizing it. It emphasizes using the customization buffer to make changes as opposed to using Emacs Lisp. I think Emacs Lisp is not covered in the book; #1 has given a few presentations on Emacs Lisp. #1 and #2 thought the output of the customize interface was difficult to parse.

#2 went over a module he added to Crafted Emacs that incorporates a lot of the information in Mastering Emacs. He also compared a few external packages to their corresponding built-in alternatives, such as how Emacs help is different than the “helpful” package. (Is there a package out there called “useless”?) Helpful provides more information (hence the name), but each call opens a new buffer, while with the built-in help you can control the windows and frames with your configuration. #2 mentioned dedicated windows, which is a term I had never heard. A few other packages that were mentioned were ibuffer (see here, here and here), dumb-jump and ripgrep. One of them mentioned flyspell (see here and here); a lack of spell checking is starting to bug me in my Emacs usage.

#2 used a LOT of key chords. I do not like to memorize key chords, but I admit it looks more impressive when someone is going through buffers, windows and frames and making changes with key chords as opposed to M-x wait-for-me-to-find-what-I-want.

At one point #2 admitted he pointed at screen with his finger while on the call; nice to know I am not the only person who does that. He also used “a priori” in a sentence outside of a philosophy class. You don’t hear that very often. In 1992, I heard someone use “i.e.” (as in “id est“, or “that is (to say)”, not Internet Explorer). Pedantic points for everyone.

There might be a topic next month.

Don’t forget: EmacsConf 2022 is happening online on 2022-12-03 and 2022-12-04.

You’re welcome.

I give people numbers since I do not know if they want their names in this write-up. Think of it as the stoner’s version of the Chatham House Rule. I figured that numbers are a little clearer than “someone said this, and someone else said that, and a third person said something else”. Plus it gives participants some deniability. People’s numbers are based on the order they are listed on the call screen, and the same person may be referred to by different numbers in different months.

I am not the official spokesperson for the group. I just got into the habit of summarizing the meetings every month, and adding my own opinions about things. That said, if you like something in this post, I will take credit; if you don’t, blame somebody else.

Image from the Gero Codex, a 10th-century Ottonian manuscript housed at the University and State Library Darmstadt; image under Creative Commons CC0 1.0 Universal (CC0 1.0) Public Domain Dedication.

2022-09 Austin Emacs Meetup

by

There was another meeting a week ago of EmacsATX, the Austin Emacs Meetup group. For this meeting there was no predetermined topic.

#1 was one of the organizers, and is active in the Austin Clojure group.

#2 was one of the organizers, and formerly worked for the City of Austin.

#3 was the professor in OKC.

#4 was a new guy in India. I do not think he said what part of India he was in (or I did not put it in my notes).

#5 was a new guy in Warsaw, Poland.

Two months ago we had an attendee in Japan. This meeting we had attendees from both Europe and Asia. If we get one each from South America, Africa and Australia, then we will have attendees from each populated landmass. Someday we might get someone out there who has used Emacs for scientific research in Antarctica.

I googled “Emacs Antarctica”, and I found out there is another software package called EMACS: Electromagnetic Airport Control and Survey. It is used to monitor electromagnetic radiation at airports. I also found an article from Linux Journal from 1994 about Australian researchers using Linux all over Australia, even on their bases in Antarctica. There is only a brief mention of Emacs in the article. It is listed with other software that was installed “to make life a little easier”. I guess they did not like vim.

As vim boosters love to point out, you should use vim because it comes with Linux. Kind of like how your Luddite relatives use Windows because that is what was on the machine when they bought it at Best Buy. In all seriousness, there is nothing wrong with buying stuff at Best Buy, but if you are going to spend all your time in a text editor, inertia is a terrible criteria.

As usual, there was no set topic.

When I joined, #1 said that he updated his Mac laptop and everything broke. I am not a Mac person, so I just waited until the topic changed. Someone told #1 to try the Emacs For Mac OS X site to get new binaries.

There was some discussion about features coming in the next version of Emacs. #5 mentioned pgtk, or pure GTK. From googling, the benefit is that there are a few GUI backends to Emacs for different OSes, and pgtk should make Emacs look better and run more quickly on multiple OSes. I use Emacs with the –no-desktop command line switch, so I might be wrong about pgtk’s benefits. #3 started using nightly builds, and found out there is a new command line switch: –init-directory, that might make chemacs2 obsolete. #1 pointed out that native compilation was supposed to be in 28, but got pushed to 29.

#1 was having some issues with git and Emacs. He uses dired instead of his OS’s file explorer interface, but when he moves a file in dired he loses the git history. He asked the other attendees if there was a solution. Someone suggested he try magit. I will ask him next month if he resolved his issue. The magit site is here, the Magit docs are also on EmacsDocs.

#1 and #2 talked about Javascript for a few minutes (another topic that does not interest me). They compared the relationship of ClojureScript to Javascript to that of Clojure to Java: Clojure started out with a lot of wrappers around Java libraries, but now seems more “filled out” (#2’s term), which ClojureScript does not seem as advanced. They said a lot of ClojureScript projects that they are close to have been re-written in Javascript. I pointed out that the Brave Clojure jobs list seems to have more ClojureScript jobs than Clojure jobs, which depressed me and surprised the other two.

At the meeting before the prior meeting, The Artist Formerly Known As #3 talked about his experience w/XEmacs. At some point I came across SXEmacs. Some developers forked the XEmacs source code and are trying to keep the dream alive. I did not try it out, and I have no idea if it can work with a current GNU Emacs config file. The last commit was a year ago. I don’t know if that means the developers have it at a point where they are happy, or if they have abandoned the project. You have to join the mailing lists to view the archives, and I did not feel like joining just to see if it is still active.

It looks like a one-man show. This reminded #5 of TempleOS. I don’t think it is a fair comparison. The TempleOS guy really did go crazy. I think it’s funny that the maintainer of SXEmacs says he will not “tollerate” fools (page here, Wayback Machine snapshot from November 28, 2020 here). I have to admit, for a long time, one of my fears was misspelling “intelligent”. Perhaps it’s time to add spell check to SXEmacs.

Someone pointed the group to Rust Emacs, an attempt to port Emacs to Rust. It has not been maintained. The language stats for the project are interesting:

Emacs Lisp 71.5%
C 16.7%
Roff 5.7%
Rust 1.3%
Objective-C 0.8%
M4 0.7%
Other 3.3%

 

I admit I do not know a lot about Rust. I always thought that Emacs was written mostly in Emacs Lisp, with the Lisp interpreter written in C. I assume that the main task would be to re-write the interpreter in Rust. But they have 10 times as much C as Rust. #2 thought it is odd that the Rust community seems so gung-ho about writing everything in Rust. Then again, when you are trying to replace C, you have to be gung-ho.

I asked #4 and #5 to introduce themselves since they were new. #4 is a developer in India. He uses Python, R, and Org mode, and is interested in literate programming in Org. #5 is a PHP developer in Poland. He has been using Emacs for four years, and intends on diving in and learning more about Emacs, Lisp and Clojure. His goal is to live in Org mode. He likes Emacs because he can execute PHP in Emacs. #1 told me it is possible due to Org Babel.

#4 then shared his screen and showed us how he does literate programming in Org. He was able to put images in his Org files and resize them. This is similar to how #3 in OKC uses Org, and he was very interested in doing more with images. They were exchanging ideas that were over my head, since I do not use Org the way they do.

At first #4 was not able to share his screen. Then he switched from Firefox to Chrome, and we could see his screen. We were using Webex for the meeting. So if you ever have trouble sharing with Webex, try a different browser.

You’re welcome.

I give people numbers since I do not know if they want their names in this write-up. Think of it as the stoner’s version of the Chatham House Rule. I figured that numbers are a little clearer than “someone said this, and someone else said that, and a third person said something else”. Plus it gives participants some deniability. People’s numbers are based on the order they are listed on the call screen, and the same person may be referred to by different numbers in different months.

I am not the official spokesperson for the group. I just got into the habit of summarizing the meetings every month, and adding my own opinions about things. That said, if you like something in this post, I will take credit; if you don’t, blame somebody else.

Image from Dobrilovo Gospel, a 12th century manuscript housed at the Russian State Library; image from Wikimedia, assumed to be under public domain.

2022-08 Austin Emacs Meetup

by

There was another meeting a couple of days ago of EmacsATX, the Austin Emacs Meetup group. For this meeting there was no predetermined topic.

I will tell you up front there was a lot of talk about Doom and running Emacs on Windows, neither of which interest me. Then again, you can get Emacs to do anything you want, so this may interest you.

#1 was one of the organizers, and is active in the Austin Clojure group.

#2 was a developer in the Dallas-Fort Worth area. He tends to talk quite a bit. Kind of like our non-professor developer in OKC. Now that I think about it, I don’t those two have ever been at the same meeting. Coincidence? I think not.

#3 was one of the organizers, formerly worked for the City of Austin. Now your complaints must be taken elsewhere. He did not say much while I was on the call.

#4 had just come back from Europe, and is a developer in Austin. Everyone else seemed to know him, but I did not recognize his face, his voice or his handle. At one point he made a comment that he got a programming job “off the street”. I do not know if that means he scraped by getting a CS degree, or if he was in a non-STEM related field and decided to do something different with his life.

I called in a bit late. #1 was sharing his screen and trying to get emacsclient working on his Mac.

Then #2 and #4 talked about getting Emacs working on Windows with Windows Subsystem for Linux. If you can use Emacs, then you should have no problem with Linux. And why not just run straight Linux? I do not like MS, and I do not trust MS. I got a mini from System 76 with PopOS pre-installed, and it works flawlessly. MS might consider people running Linux to be a problem; I think that is the solution. WSL might be solving a problem MS has, it is not solving a problem I have. Maybe the guy is a hardcore gamer. Still, I think most developers can shell out enough cash for another system. I would never want to discourage people from running Emacs, but it looks like #2 and I see some things differently. (Perhaps that is inevitable with something as customizable as Emacs.)

Granted, he has a job where he uses Emacs all day, so he is doing something right. #2 also mentioned they recently converted a few vim users to Doom Emacs, and one of them said it changed their life.

#2 said running Doom Emacs on Windows was difficult. Apparently Doom uses some C libraries that are not available on Windows. I have no interest in Doom, so I did not ask for details. Later #2 mentioned that he was able to run Doom with multiple profiles without using Chemacs. I would not be surprised if multiple profiles becomes a feature in future versions of Emacs.

#2 also mentioned MS Powertoys. At first I thought he was being facetious and making fun of something called “Power Tools”, but I checked, and, no, it is actually called PowerToys. The jokes write themselves. The page on MS’s site describes PowerToys as “Utilities to customize Windows”. Unfortunately, MSPT would not let #2 make the adjustment he REALLY wanted to make, which sounds like Peak Microsoft.

#4 said they use Emacs for Ruby and Go, and that led to several minutes of people offering their opinions on Go. I have tried it out a bit, and constantly checking if something is nil gets old fast. #4 had heard people say bad things about it, but when he tried it he liked it. #2 and #4 both use Doom, and like #1 use Emacs for Clojure.

#2 mentioned that a recent change in Emacs resolved the “long lines” issue. I think he might be referring to so-long, which was mentioned on Mastering Emacs (also see EmacsWiki page, ELPA page, Savannah page and source page).

The last topic was Brushes With Greatness. #2 had a talk with Simon Peyton Jones at Strange Loop where SPJ gave a talk on “Shaping our children’s education in computing“. #2 mentioned a few things about Haskell that he did not like, and realized it was a mistake. SPJ was not a jerk, and he actually seemed interested in making Haskell better and hearing what #2 had to say, but he was asking questions at a level of detail that #2 was not able to answer.

#1 talked with Rich Hickey at ClojureConj a few years ago. RH uses a minimal setup: Just paredit (see pages on EmacsWiki and WikiEmacs) and Inferior Lisp mode. #1 asked RH if he did any web development, and RH said that he thought web development was too complicated. #1 and #2 thought it was interesting that some big names in development who work on some important projects have some of the simplest setups.

You’re welcome.

I give people numbers since I do not know if they want their names in this write-up. Think of it as the stoner’s version of the Chatham House Rule. I figured that numbers are a little clearer than “someone said this, and someone else said that, and a third person said something else”. People’s numbers are based on the order they are listed on the call screen, and the same person may be referred to by different numbers in different months.

Image from Harley MS 5647, an 11th century manuscript housed at the British Library, assumed to be under public domain.

2022-07 Austin Emacs Meetup

by

There was another meeting a couple of weeks ago of EmacsATX, the Austin Emacs Meetup group. For this meeting there was no predetermined topic.

#1 was one of the organizers. He told us he no longer works for the City of Austin, and now works for a payment processor.

#2 was the other organizer, and active in the Austin Clojure group.

#3 was one of our developers in OKC.

#4 was a developer in Austin, and also a member of the Austin Clojure group.

#5 was an expat in Japan. This is the first time we had an attendee from overseas. He is a Haskell programmer.

#6 was our other developer in OKC (the professor).

The first topic of conversation was some new things in Emacs 28. There are a lot of new changes. The pace of Emacs development has accelerated in the past few years along with interest in Emacs. A few people on the call were getting into the new built-in project management solution called simply “Project”: see the page on Mastering Emacs (there are a LOT of things listed on this page; Project is more than halfway down) or the GNU site. This can be used as a replacement for Projectile (Github page here, documentation page here). Number 3 has replaced Projectile with Project.

Number 3 has had a finger on Emacs development for quite a while. He said that Bozhidar Batsov approached the FSF about integrating Projectile into Emacs, but they did not get back to him for a couple of years. By the time they did, Projectile had changed and also had accrued more contributors. Mr Batsov did not think that all the contributors would sign their code over to the FSF (or chose not to bother), and that was the end of it. At some point the FSF decided to make their own. Number 3 said this is in line with Crafted Emacs (formerly Rational Emacs): use or enhance what is built-in as much as possible, with using something external as a last resort. He admitted he might be biased, but he said that there is a general trend of moving towards the built-in stuff.

I will have to add looking into Emacs 28 features (like ibuffer and fido) to my ever-expanding to-do list (which of course is in Org mode). There are a few new functions in Emacs 28 that deal with outlines. Since I tend to use M-x commands rather than key chords I will definitely look into those.

I asked #5 and #6 how they learned Emacs (since I am interested in helping evangelize Emacs to newbies). #6 gave a presentation a few months ago, and while he had been using Emacs for two years when he presented, he learned a lot in his first few months. By the time this topic was over, I think almost everyone had given their Emacs origin story and waxed nostaligic for OSes and systems long gone into the ether. I will start with mine.

I worked at a firm doing analytics from 1998 to 2000. All the developers had Sun workstations (I think mine was a SPARCStation, but it might have been an Ultra). So you had to learn Emacs or vi. Someone showed my vi, and while I got the idea of modes, it just seemed stupid to me at the time. It feels like the Unix version of the Microsoft Paper Clip: “I see you are editing text. Do you want the text to actually appear in this file?” I went to one of the people who knew Emacs, and I asked them: “What happens in Emacs when you press the ‘a’ key?” The guy said: “The letter ‘a’ is added to the file wherever the cursor is. Just like almost ever other text editor on the planet.” I told him to show me more Emacs. Later in 2000 I went to Bank Of America, and I had to use Windows. But I did have a PA-RISC workstation under my desk running HP-UX.

Over the years, my Emacs usage waned. I started running Linux on my home machines, but I tended to use JEdit more than Emacs. There was an Emacs group in Austin, but they met at lunchtime. Given how spread out Austin is and how spread out the jobs are, a lunchtime meetup just seems like a bad idea. Growing up near Chicago, the Loop felt like the center of the universe. It had the highest density of jobs in the region. There are jobs throughout the region, but the Loop has a lot more. Even then, a lunchtime meetup would not have gotten a lot of traction. Doing that in Austin just makes no sense at all.

Number 2 took over, and started having meetups in the evenings. It has been blue skies and no blue screens ever since. The first evening meetup was on Thursday, September 22, 2016, at the now-closed Cafe Express (they still have locations in Houston). It was at one of the meetups at Cafe Express that I first heard about ido and smex. Someone asked what happens if two modules have the same keychord. Another member mentioned smex. It changed my Emacs experience. When I learn new modes, I use the M-x function names and not the key chords. Granted, it can make following along a bit cumbersome. Usually the function names describe what they do.

Until I started going to the Emacs meetups, all I really knew were the basic commands for editing, searching, killing and yanking. I did not do much with modes, buffers or dired. I did not know much, but I got a lot of mileage out of what I did know.

I finally ditched JEdit and went all-Emacs outside of work around October/November 2021 when I found out about the Emacs desktop, and how to only use it for one open session and not others.

I do have Emacs installed on my work laptop, even though it is not a standard tool. I would like to work a job where I can be in Emacs all day. I used to keep notes about all sorts of things at work and at home in text files. Now I use Org. One day I googled if it was possible to do outlining in Notepad++. One suggestion is to “use a UDL (User Defined Language) and just have parentheses to do the folding for you.” Using your editor to use parentheses to make your own language. Sounds familiar.

There are a few other suggestions here, here and here. One of those suggestions lead to this post on OutlinerSoftware. They only have 8 topics to a page (as of 2022-07-27 there are 3,760 pages).

I use the outlining feature in Org a lot. I am on Teams calls with our testers at work a lot, and some of them write their scenarios in Notepad. Not Notepad++, but Notepad. I have no idea why. Anyway, they will have a list of preconditions in a run-on sentence on one line. Making these sections in an outline would make them easier to keep track of. I know PowerPoint has soured people on bullet points, but they are good for organizing thoughts. I am reluctant to talk to them about using an editor with outlining features because I have a feeling they would all go to Word. Note to corporate America: Innovation does not mean “Let’s do everything in Excel” or “Let’s do everything in Javascript.”

Number 1 used vi at his first job. Everyone else used Emacs, but he resisted for a long time. He started doing Lisp, Scheme, and started using Emacs for that. He looked into Perl, and when he found out you could run the perl debugger through Emacs he was hooked.

Number 2 used Emacs on and off. He was working at a company that could not get licenses for UltraEdit, so one day he sat down at a cafe with Number 4 and started learning vanilla Emacs.

Number 3 started using Emacs in 1990 on an AIX machine without info files for reference. He switched to MicroEMACS for a while, and eventually got GNU Emacs running on OS/2. He started getting serious about Emacs in 1995, and kept using Emacs wherever he went. For a period he used SlickEdit to configure Emacs. I told him if we had geek points to award that we should give them to him. Most people have never heard of AIX. He might be the first person I have spoken to in a long time who has actually used it.

Number 5 was a vim user for eight years. He found a vim plugin for org-mode and liked it. He decided to try Emacs, but had trouble with key chords, so he uses Evil mode. He started eight months ago, and has been using Emacs every day for about two months (which will be three months by the time I publish this). He prefers Emacs since it is more customizable. He said Emacs and vim are so different he does not know why people lump them together.

Number 6 started two years ago, and committed a year ago to using Emacs everyday. He started with a book, I think he said it was Learning GNU Emacs, and said it was a good overview. He tried Spacemacs, but found having to deal with the extra layer offputting. Then he just went with the default. He said the main thing is to have a reason to use Emacs and weave it into your daily work. I think for him Org mode was the killer app, although at first he tried to do too much and found success by starting over and incorporating Org more slowly into his workflow. Number 6 uses Org, overleaf, Atomic Chrome and a lot of LaTeX.

There was some discussion about keyboards and ergonomics. Number 5 uses evil because Emacs is hard on his wrist. Number 1 has learned to use both control keys instead of just one. I just move my hand an inch. Speed is nice, but it is not the most important thing. Numbers 3 and 5 both tried Dvorak keyboards with Emacs, and both thought it was a train wreck.

We talked about the changes that have been happening in the Emacs community over the past few years. XEmacs came about because some people thought new versions of GNU Emacs were not coming out quickly enough. XEmacs got some traction. I used it for a few years, but the last release was in 2009. There was some turnover in the GMU Emacs leadership, and there have been more frequent releases and improvements to GNU Emacs. Number 3 said he submitted a small patch to both GNU Emacs and XEmacs. GNU Emacs incorporated it without a fuss, but this small change caused a lot of arguments on the XEmacs mailing list. He said it turned him off to XEmacs, and was not surprised that it eventually died.

You’re welcome.

I give people numbers since I do not know if they want their names in this write-up. Think of it as the stoner’s version of the Chatham House Rule. I figured that numbers are a little clearer than “someone said this, and someone else said that, and a third person said something else”. People’s numbers are based on the order they are listed on the call screen, and the same person may be referred to by different numbers in different months.

Image from the Susteren Gospel Book, an 11th or 12th century manuscript housed in Susteren Abby in the Netherlands; image from Wikimedia, assumed to be under public domain.

2022-07-13 Update

by

I know I said I wasn’t going to write about my site, but I have been spending a lot of time on it, and looking into static site generators and moving away from WordPress (or as some people call it, “WordMess”).

I have a few sites that do not get much traffic (which is fine with me). But I do use the search features on WordPress a lot. It is nice to have a record of my thoughts without relying on Google. I got locked out of one of them recently. I don’t know if it was the site host or WordPress that locked things down. The lockout said it would expire in fifteen minutes, but it lasted an hour. I googled for a solution, and one was to change the slug for the login page. Which required a plugin. So I installed the plugin and changed the URL to a random word. I suppose I could have used a UUID.

Then I was getting email reports about attempted logins due to some xmlrpc issue. Google searches showed that the preferred solution was another plugin, but I was able to take care of this issue with the first plugin.

And then there are the comments. I might just turn them off. I have not gotten a useful or relevant comment in a long time. And since Russia invaded Ukraine (which is NOT anyone’s fault but Russia’s) I have gotten a lot more comments in Cyrillic. And I still get ads from people offering me SEO capabilities, even though my contact page says this is not a business and I do not need any services from you.

And then there is the possibility that the custom editor will stop working. And then one time it went into “Maintenance mode”, and did not come out. I had to delete a file from the server via SFTP.

Also on my post about CJUG presentations I was not able to get the formatting the way I want. A few times I got it to look the way I want in the preview, but after hitting the “Update” button the system changed it.

So I have started looking into static site generators. This is a larger project than I anticipated, and will take longer.

I would like a static site generator that can make my site like it is now: showing full posts ten to a page, with subsequent posts on subsequent pages. I do not want one page listing all the posts. I would like to have the files in different directories as well. I have a lot of posts, and I do not like hundreds of files in a directory if I can avoid it. I want each post to have links to the posts before and after it. I want an SSG with themes in case I do not like the default. And I want an SSG with an active forum or mailing list in case I have questions that I cannot answer from looking at the documentation.

Here are some notes for future reference:

  • The Python contender is Pelican (Github link here). I have not tried this one out yet. They do have an IRC channel. Those are not the best ways to get answers. I prefer forums or mailing lists. Their Github wiki has pages not on their main site. I do not want to have to look in multiple places for information.
  • The Clojure contender is Cryogen (Github link here). I could not find any mention of a way to interact with the community. In general that would be a point against, but I am pretty familiar with Clojure so I might be able to get it to work the way I want without assistance. Plus a lot of Clojure projects Just Work(tm).
  • The Java contender is JBake (Github link here, themes on Github here and here). I tried this out with a few posts. I do not have to have all the posts in one directory. However I do not know if all of the archive posts have to go on one page. I will keep looking into this one.
  • The Ruby contender is Jekyll (Github link here, forum here, themes site here). I had to jump through a lot of hoops installing Ruby via rvm, but I got it working. This is one system that seems to require all the files being in one folder. One point in favor of Jekyll is I did find a WordPress plugin that converts WP content to Markdown pages.
  • The Go contender is Hugo (Github link here, forum here, themes site here). Like Jekyll, this has a lot of features. It could probably do what I want, but it would take a lot of time to look into it.

You’re welcome.

Image is a scene from the life of 5th century Desert Father Arsenius the Great from an 11th century menologium manuscript housed in the State Historical Museum in Moscow, image from Wikimedia, assumed allowed under Public Domain.

2022-06 Austin Emacs Meetup

by

There was another meeting this week of EmacsATX, the Austin Emacs Meetup group. For this meeting there was no predetermined topic.

Most of the attendees were in Austin, but we had a few from other states.

#1 is one of the organizers. I think he works for the City of Austin, and does not use Emacs much at his job.

#2 has a math degree and makes machine learning software.

#3 is a student in Massachusetts. He did not specify which school. He uses Emacs to program in Haskell (which was the first language he learned), Common Lisp, and Perl.

#4 is a sysadmin in Tennessee. He was one of the few who lives the dream of using Emacs for his job. Now that I think about it, I should have asked him why he likes Emacs, since a lot of sysadmins prefer Vim. Granted, he could use both. I recently logged into a system that had only Vim and I was able to make a small change. I was not converted.

#5 is another machine learning developer in Austin.

#6 presented a few meetings ago on Rational Emacs (now called Crafted Emacs). He is a software developer in Oklahoma City. I asked him if he knew the presenter from last month before the presentation, and he said he did not. (I know there are around 700,000 people in OKC, but it was a possibility.) We did spend a few minutes chatting about Oklahoma City in general. Like its economy, the OKC tech scene is dominated by oil companies. The capitol building is on top of an oil field, one of the few in the country in an urban area. #6 uses Emacs for several languages for his job, and is also living the dream. I do not remember what languages he uses. I think Python is one of them.

I was hoping the presenter from last month would be here. I think he learned Emacs not too long before he submitted his talk to EmacsConf. If he is a beginner and able to put executable source code into an Org file, then whatever tutorial or book he used was effective.

#7 is a local Austin developer, living the golden dream: Using Emacs to write Clojure. He showed us running Emacs on his iPad using the iSH app. A few people wondered how long it will be until Apple cracks down on this, and whether you can run Emacs on Android.

#3 mentioned shx for Emacs (or “shell extras”), which extends Comint mode. Per the Emacs manual, Comint mode is “a general-purpose mode for communicating with interactive subprocesses”. This led to some discussion about shell modes in Emacs. #6 said that since there are modes for every programming language he uses, he does not do much in the shell since the modes have commands for most things he needs to do. “There’s a mode for that.” I have been spending time with eshell. #1 pointed out eshell works wherever Emacs does, and you can run ELisp code in it, so it is portable and powerful.

The discussion then pivoted to comparing Lisps. The consensus was that Clojure might be The Lisp That Sticks, and that since it is on the JVM it handles dependencies better than most Lisps. I pointed out that a lot of Racket packages do not have version numbers or version numbers for their dependencies, which horrified a couple of people. A few of us agreed that Dr. Racket looks cartoonish, and Racket Mode is much better.

#7 mentioned that Edsger W. Dijkstra wrote a letter to the UT CS department complaining about the decision to drop Haskell as the introductory language. The more I read about Dijkstra, he seems like the sort of person who is only happy if he has something to complain about. #3 started with Haskell, and then went to Prolog before finding Lisp enlightenment. #7 was a CS student at UT when Dijkstra was there. He regrets not taking a course with Dijkstra. Apparently, EWD did teach undergrads. But #7 decided not to since the grade for the course available was based on one oral exam. #5 pointed out that sometimes you don’t know you are in the “good old days” while they are the good old days.

#7 also posted a link to a book by Edgar Graham Daylight, who blogs about EWD at Dijkstra’s Rallying Cry for Generalization. You have to go to the “Recent Posts” list on the right to see the latest post. The post on the front page is the one of the first posts from 2011.

You’re welcome.

I give people numbers since I do not know if they want their names in this write-up. Think of it as the stoner’s version of the Chatham House Rule. I figured that numbers are a little clearer than “someone said this, and someone else said that, and a third person said something else”.

Image of Saint Luke from the Gelati Gospels, a 12th manuscript housed in the Georgian National Museum, assumed allowed under Public Domain.