Plain text can be super versatile (task-tracking and note-taking are great examples), but equally important are the guarantees that your content is truly yours (no lock-in).
With all these plain text posts surfacing regularly, I'd love for the lesser-known org markup (https://orgmode.org) to gain more adoption. It's a real power-house. Its Emacs origin may put some off, but it's plain text, so your content can be ingested/consumed by either regular text editors or any app focusing on specific user-journeys.
I dunno, that kinda seems like throwing a bunch of straw-man arguments and unjustified claims like
>Anybody using a simple text editor is very fast at adding markup for headings, font variations, and so forth. The previous section proved that other markup languages clearly fail in many cases.
Markdown is not quick to add markup with a simple text editor? Really?
No, it has a single tool's support and no spec (though it's in progress last I saw, which is great). Almost everything else compared in that page is superior here, having specs (even if multiple competing specs, e.g. commonmark vs github) and multiple implementations.
You can run pretty much any programming language from within org-mode and have literate lab/reports/configurations. Since org is tightly integrated with Emacs, there's also much more interaction with code in general (Emacs Lisp) which lets you do pretty crazy things. Like run and output SQL queries to your document or have interaction à la Jupyter Notebook.
* Excellent outliner features, similar to those from OmniOutliner for example
* Great support for tables (including mini spreadsheets), links, footnotes / citations, tags / keywords and timestamps
* The ability to build interpreters that create views of one or several documents
The last feature looks weird and exotic, but it is super useful. For example a kanban is just a view that displays e.g. ongoing tasks and forthcoming deadlines harvested from your org files. Actually, Emacs already has this feature implemented as the function org-agenda. But you can build other equivalent views.
(If you know, of course:) How well does org-mode play with Pandoc? Much of my workflow goes from Markdown > LaTeX, and Pandoc works pretty well for that. Thanks!
There's pandoc-mode in Emacs and that lets you fully integrate pandoc with org if needed. But you can also convert org files to LaTeX and PDF directly. That's how I take all my notes.
Orgmode is awesome but it's frustratingly difficult to use outside Emacs, and using Emacs isn't an option or productive for the vast majority of people, far, far too much cognitive overhead. Im hoping for a hybrid that is easier to read as plain markup but has the extensibility of org.
> Orgmode is awesome but it's frustratingly difficult to use outside Emacs
Ah yes. I get the Emacs part putting some folks off. Org markup is super capable and stands on its own merit. I’d love for it to be more independently recognised, spurring a rich and diverse ecosystem of org-powered tools (outside of Emacs).
I've found Obsidian pretty great, but it's not opensource. It has a pretty alive community as well and it's all just text and pictures in the end, so it stays in your control.
I'm a big fan of obsidian and use it for all my notes already but being basically just text, links, and pictures, it's easy to long for the power of org-mode.
THat is true but after trying to get the hang of it for a week I was back to obsidian :) . I think it would take months, and since I don't use emacs as my regular editor it just wasn't worth the cognitive load.
I believe you missed checking out the websites. The two apps are pretty different but do focuses on the philosophy of the simplicity of plaint ext. Interesting Apps.
> the guarantees that your content is truly yours (no lock-in).
Maybe I am a magician of some kind, but my content is always mine, regardless of the file type I use.
For the sake of argument, let's say I have a floppy disk with a Vizawrite file on it, that I typed 30 years ago. There is no lock-in. I still can use that file.
There are at least two problems with proprietary file formats.
First is the app lock-in, which perhaps hasn’t bitten you but not everyone is so fortunate, especially in a scenario the article describes, corporate environments where you can’t install arbitrary software onto the machine you use.
The other problem is corruption. I haven’t experienced this in a few years, but it’s certainly not unheard-of for a binary data file to experience corruption; often the only way to recover it is to perform surgery on it, which not everyone will have the tools and skills to do.
1) You can have corruption in files of any type. Text is no more robust to random bit flips. Pick a file format and there are often tools that can isolate the corruption to a small portion to recover everything else.
2) The solution to lock-in is not "let's make everything text", but "let's have open file formats". Many types of data, especially for project-management-like tools, aren't particularly well suited as plain text (or a markdown format). So many extra unnecessary parsing/regex steps have to be jumped through for the program to use it, and the sanitizing functions now have to be moved from code to the user's brain to not break things.
By all means, support open file formats and exporting functionality. But shoehorning -everything- into text sounds like some crazy mix of bikeshedding and worldbuilding. Why waste your time? Especially knowing MD will fall out of favor eventually, and a new format du jour will rise. What will you do then?
I'm pointing out that everyone is comfortable with most files not being human-readable. And that it's a red herring to suggest that text is the longest-lived format. I'm sure my data could outlive text files if I "stored" it on granite blocks, but it's not very useful in that format. Same with plain text; it just isn't a very useful format.
It boils down to a tradeoff of learning your tools while retaining full control vs use apps that abstract away complexity by giving up control.
If your work needs some strict technical requisite (reproducibility for example, or any other of the -ilities in software engineering), you need to have the control to tweak your tools so that you ensure it (Engineering Model, plain text files), but mostly is not needed so you can get away with working in the Office Model (binary files)
The app lock-in doesn't make me lose access to my content. Many proprietary formats are supported by import/export in major apps. There are converter tools. Also, I do not need to switch apps.
The problem with corruption is solved by backups. It's a solved problem.
But can you get your data out of Vizawrite and into some other tool and edit it there? And will Vizawrite get ported onto all the environments you might want to work in? (eg does it have a mobile app?)
The value of plain text with markup is that any of the tools @xenodium mentions above can work with data from any of the others. I use emacs/org to edit my data from BrainTool, I use Orgzly on the same file on my Android and its also part of my LogSeq digital knowledge garden.
Not necessarily a ‘productivity hack’ in that it neither relates to productivity nor is a hack, but in our latest side project my partner and I have a need to share textual data about the various episodes/lessons that we’re creating.
Of course any number of notes systems allow one to do that, and I have my very own system of organisation that helps [0], but we ended up, at my partner’s suggestion, just saving individual .txt files in the folders containing the episodes.
Nothing new here of course, it’s what we all used to do when it was the only option. But it feels like it’s been so long since I just fired up TextEdit, typed the words, and File-SaveAs’d it in to a folder. It’s weirdly refreshing and works really well.
I noted that it was my partner’s suggestion because she’s been learning Markdown, and as an ex-professional-writer (medical/patient) is overjoyed that she doesn’t have to use Word. When we were looking for a solution she said, well why don’t we just save a file there. So simple that I’d missed it.
I'll be first to mention obsidian here. I solved my overloading/losing faith in systems issue with it by using a daily template. At the bottom of which are all the tasks in daily notes, generated by an inline script, another dynamic list of tasks for my freelance work. I could add my #ideas or #content lists too, but they are stored elsewhere. I add the #inv tag to my freelance work and that moves them to the "to invoice" list. When finally checked they disappear, but all still stored on their initial log date (all searchable, of course). So long lists of crud don't get the chance to accumulate, and ideas are easily searchable either in dynamic lists or just using search.
On the same page are mental/physical subheadings so a small bit of journalling takes place. I never look back at it, but it's there.
This beats all previous attempts at organisational systems. Obsidian has a great mobile app. Syncthing keeps them updated. Previous systems were Trello (perma loss of everything archived), big single plaintext file (soon becomes a heaving mess), paper (no search and easy for things to sink/not be surfaced). Taskwarrior, which was great for a long time, but tricky to sync and no additional features like journalling, swipe file creation, flashcards, search, mobile app (sync to cli env on Android was clunky). And surely more I'm forgetting.
Side benefits: all plaintext, whatever happens to obsidian. Graph view of all interconnected ideas. Easy resurfacing of old but relevant ideas through search and backlink sidebar. Could import my attempts at using vimwiki with zero conversion required. Much better UX than Vim. Great place to also write content, use wordcount, sprint, wordcount goal, calendar features etc.
For me the days of Vimwiki and org-mode are now numbered. Even most other note taking software tbh, for anyone OK with syncing their own stuff and figuring out a little bit of basic script syntax for the dynamic lists within the plaintext files.
I use tags and a query to view different groups of tasks
The tasks are generated as jobs come in, i.e. a checkbox line, I assigned ctrl+T to task edit, so I can add a due date. I make sure to add #freelance so it gets picked up by the query below, and displays in a list on my daily template (ctrl+shift+D for me) so whichever daily note I'm in I can see my tasks.
Then I have another for jobs to invoice, adding #inv to the line manually is simple because the tasks plugin lets you click through from the list, and another for "general" todos, which does accumulate, but not in a single file.
I'm a recent Obsidian convert, absolutely love it and its the only note taking app that I've managed to stick to past a few weeks. How do you use custom scripts in it? I'm aware you can make plugins and things for it but I'd love to be able to write up a script or two to update some shortcodes for some nice weekly stats in a few places if possible
I have some examples of weekly stat graphing and Rollups of in ticked todos in completed action notes in this repo: https://github.com/tot0/ObsidianPPV
The week template in templates folder is the most fleshed out now, other time periods are WIP. It leverages customjs plug-in and the js folder in the vault for helpers.
The data model implemented in this vault is August Bradley’s PPV, the reviews cycles it emphasizes really help with regular re-orientation on changing priorities and not letting stuff slip.
There is the full-on dataview plugin, lets you do anything in JS. I just use the tasks plugin query engine, which is plenty for me, but it sounds like you would need dataview. The forum has loads of very well developed examples.
I have been using OneNote after going through a bunch of plaintext tools and I have never been at more ease with dumping info from my head for later retrieval. OneNote makes it as fluent as possible with as little resistance that no other tool offers.
New version of OneNote (metro style) has been bitten by designers though like all good things in recent times. Also, it does not have customizable shortcut keys making formatting code snippets a non-trivial process.
Every page is a white board which no other tool does and that's where it's power is. Plaintext can't beat that.
I love OneNote, it's my primary note taking tool, although I'm eagerly looking for a more open alternative.
I love pasting in screen clips and shots, and being able to extract text from them when I need to.
I also love arbitrary indentation and use that for structuring my notes. This is something that Markdown can't do, and I have yet to find any kind of markup-supporting text format that doesn't treat indented text as pre-formatted.
I dislike that newer versions force syncing with Office 365 and have reduced the file format flexibility of older versions, like saving the notebook on a network share, or exporting to an MHT file if one wants.
I still use the Office 2016 version and am hesitant to try newer versions again. I have numerous notebooks on many different PCs and want to coalesce them at some point, but am not comfortable putting them all on Office 365.
But for day to day note taking, it's the lowest friction tool I've found that comes closest to my preferences.
I've become a heavy user of OneNote recently as well, but I'm always in search of ways to improve my workflow. I have yet to fall into a consistent pattern of usage.
I have a directory with text files and images named like: 2022_03_20_1_short_description.txt or 2022_03_20_2_config_screenshot.png
The names are mostly to group images with related text files when I grep later.
I create the file at the start of each work day, write a few lines of TODO for the task I'm working on, and whenever I copy-paste something that seems relevant - I paste it there and leave it with a few empty lines as separator.
If image is better I screenshot only the relevant part and possibly add some notes with paint and save it in that directory.
When I investigate some error I paste stacktrace there. And when I find solution - I paste the link or the git commit or JIRA task or description.
If I do several unrelated things in the same day I sometimes create several files, but it doesn't matter that much.
It saved me A LOT of time. A quick grep and I have all the context of some error that last happened 8 months ago.
I do something close to this but with even less organization up-front. I label my text file as <day>.txt, like 21.txt, since it is under a month folder.
For images, I take screenshots with the Snag-It software which automatically assigns a time-based filename to the screenshot and stores it in a central directory.
So, at the start, my text files and screenshots have basic filenames which are time-based.
Later-on, I'll create an index for the text files with a Google Doc; an index which is a basic list of the text files with some brief bullet-point comments describing each one.
Actually, for text files, I just describe it with a comma-delimited list of tags/titles per file.
Some may think it primitive but your workflow is fast, efficient, backups up easily (I write a simple DOS batch file which copies all new or updated files to a USB stick) and backs up to the cloud easily as well.
Grepping to find files by date is easy and if a more detailed document had to be generated, you have pandoc or Libreoffice to produce PDFs using those source data files.
When I see this all I can think of is Emacs and org-mode, harness all that power and that existing ecosystem. At a quick glance seems to be a lot of what this does.
The entire system is very elementary compared to org-mode. This is good for some (people that wanna try this approach since it's simpler and mostly works fine), bad for others (people accustomed to all features org provides).
My thoughts exactly. Org-mode just works and has so much functionality that I haven't even discovered yet. I am currently using it for notes and TODOs.
So, gotta jump in here and say - I'm going thru the "Let's figure out Emacs, Org-Mode, Org-Roam - get my todo, calendar, files, notes all in one... somewhat coherent system". I've got my Emacs cheatsheets, and all that. I'm on the journey. But I can't help but think - some one really needs to do the UNIX->Next->MacOS of this. That is, figure out a solution that works, standardize it, make it easy.
I saw a guy made an Raspberry-Pi, with Emacs, & an e-ink display. When someone makes a "personal productivity device" - some sort of E-Ink Palm Pilot + Blackberry keyboard + Easy mode Emacs-org/notes/todo/calendar/rolodex, that syncs to your laptop, phone, refrigerator calendar... that's going to be huge.
The reason I am going through this journey is because I am treating it as a life-investment. Getting your personal productivity / information systems in order, getting that problem solved... for LIFE. It would be a huge win. So despite the pain, I'm sold on the benefits, have many people who have told me about the impact of org-mode etc on their life. So, I'm taking that at face value, trying to actually apply this advice to my own life (instead of reading about it, agreeing, and then taking 0 action on it). The irony of not prioritizing setting up a reliable TODO system is not lost on me... It's a daily laugh, as I Google search yet another Emacs package manager, or diagnose random Lisp errors - this programming language I have never used before, but this important tool requires me to know in order to... configure a note template. What.
I know there are historical reasons for the various keyboard shortcuts. But all these "arcane" (I know, "different") keyboard shortcuts for "Yank", "Buffers". Emacs keeps on opening windows. I can barely remember how to even save a file & quit. Don't worry - got Doom Emacs! I'm getting it, but still a whole 'nother set of "non standard" keyboard shortcuts.
I genuinely appreciate the various bloggers and Youtube videos out there, where people explain their system, walk you through installing & configuring stuff.
But I can't help to think. This system is so powerful. So many smart people working on it, making plug-ins for it, creating systems for it. Where is the smart person who can tie it together as a holistic system, make it EASY, and still be open source. Am I missing something? I'm thinking "Green" mode on the SLR camera. It works. If you want to tweak, you can experiment with aperture/shutter priority as you learn when/why to use it.
Ok, thanks for hearing out my Emacs rant. (as I go smoke a J, and go back to learning Emacs)
Whatever system I have (tried to) used over time, I've run into the following challenges:
1. how to deal with old tasks (are they relevant anymore? should they be removed?).
2. connecting tasks to their related source and needed information to act
For the former, the accumulation of old stuff has consistently led to less and less usage of the system and as a result to a continuously decreasing trust in that system. Adopting a new solution has given the feeling of a better system with better chances to success; but it has mainly been about having a (temporarily) new clean inbox.
Having a data model/system to implement regular reflection/cleanup/alignment on goals I find helps with cruft build up. August Bradley’s PPV has review cycles weekly/monthly/quarterly/yearly which cover higher levels of abstraction as the time period gets larger. I’m building an implementation of it in obsidian, august does his in Notion (which I started with but started to see a cap on extensibility), it’s a WIP but I’d love other to make it less work for me :P
https://github.com/tot0/ObsidianPPV
@luxpir mentioned Obsidian in another comment [1] and it might be what you are looking for, in particular with a community plugin called Obsidian Dataview [2]. Dataview uses JavaScript as a query language and can query the Obsidian "vault" and create different views.
Another tool to look at is Logseq [3], which is essentially an outliner (supports both MarkDown and Org syntax), but supports a rather simplistic TODO management system. The benefit here is exactly what you describe—as you work in Logseq, you can create Todos, thereby connecting the task to the related source and any context that surrounds the todo. Furthermore, Logseq also allows for queries [4], allowing you to query your "knowledge graph", which you can embed in other pages.
Both Obsidian and Logseq store your files locally, so they can be easily version-controlled (In fact, Logseq routines commits your files for you).
I'll just add that the tasks plugin in obsidian has its own scripting syntax, much simpler/plain English, and is lighter weight than the entire dataview alternative. I actually set my system up to replicate logseq and taskwarrior within the more mature and faster running obsidian project. Ran them in parallel for a while on the same files.
Main difference for me was that Obsidian can be an outliner or general markdown editor. Logseq very focused on the former, and atomising your notes, but in reality this is rarely useful. In my case at least, especially when writing long form content. Search is fast and complete in obsidian, so there's very little benefit to recalling and embedding single lines from entire vault, it's just a resource sink in my case.
in emacs org-mode you can archive some tasks, text - it will move it to another text file ex. archive.txt with the timestamp or archiving it. Beside this keep your file in git. so you delete text & make explicit commit message that something was removed.
Some people that were interested in their productivity that I read seemed to have systems that regularly cleaned their todo lists for old tasks. Stephen King [1] advises aspiring writers not to note their writing prompts, as the best ideas should pop into your head regularly, or nullprogram.com [2] programming in /tmp/.
So maybe the way to go is to have a logrotate-like on your input file.
[1] On writing -- Stephen King, although I read it a very long time ago and would not be able to point to a specific page.
[2] I searched the web but can't find the exact citation, and there is no search feature on the website.
Most things accumulate and need pruning. Some people would like to do manual pruning but I favour a time-based approach of keeping everything in monthly folders.
Monthly folders so basically with every new month, some things are not carried-over but are still there and can be referenced if necessary.
So basically, a new month helps with reducing the accumulation of old stuff which bogs down a system.
Yeah, I've been thinking about how to tackle this and I am convinced any personal note taking app should have an archive all and start over button.
A clean sheet is intimidating when you don't know where to start but is a hard requirement downstream once your chosen abstraction starts to break under the weight.
Interesting, I also keep coming back to text files for productivity, because of the speed and simplicity. Some of the points the author mentions though like due dates, a journal/schedule and version history still aren’t really solved very well I think. Not to mention collaborating in a team that way (not a requirement for the author I think, but something I could use myself).
I actually tried simply using VSCode for a while with just a plain text "todo.txt", "schedule.txt" and "projects.txt", but that didn’t scale very well (also didn't really get things like autocomplete to work properly).
If anyone’s interested, we decided to see if we can actually solve this ourselves and are building an editor/IDE from scratch now, but specifically for "todo.txt" (https://thymer.com).
This is very detailed and well documented and is similar to how I work in my web based app, Nolific. Specifically, I keep notes in plain-text and write drafts in markdown. I don’t keep my todo in the system, but it would be relatively easy to do so. Once I’m happy with a draft, I copy/paste it into an MD file on my blog. All my notes are easily searchable.
I built a browser based solution because I spend most of my day in the browser anyway, so I just keep Nolific pinned in a tab.
It’s free and open source if you wanna check it out.
The rather-name-just-search approach seems less hassle but lack of version control may be an issue. Perhaps automatic checkpoints could be made on large changes or when leaving the page.
I bet I can explain orgmode in 10 lines. This is literally the only orgmode functionality I use, and have used, for years.
1. You specify subheadings, or “outline format,” with asterisks at the start of a line, with 1 asterisk (*) per level.
2. Use no asterisk for ultimate high level. If you have no asterisks at all at the beginning of a line in your text file, it’s literally a plain text file, not really org mode at all.
3. One asterisk (*) at the start of a line for top level beneath that.
4. Two asterisks (**) at the start of a line for the level beneath that - call it sub top level.
5. Three asterisks(***) at the start of a line for the level beneath sub top level…
6. You get it: add asterisks at the start of a line, one more than is present in the asterisk level above yours, to go “deeper” into subsections. If you want to “reset” and start at top level again, use 1 asterisk or *. To reset to sub top level, use two asterisks or **…
7. Now, to navigate between these levels, press tab to expand, and tab again to close. Typically your view will only be of “top level” headings.
8. You can use orgmode in Emacs (assumed default, as that's where it originated) or you can just as easily use it in vim after installing an extension. In vim, where I typically use it, press zR to expand everything, if you’d rather default to a view where everything is expanded and you contract stuff to make it more navigable.
I don’t use any org functionality outside this, just this, and it’s been a godsend.
It can be as much or as little as you want, really. If you are completely new to Emacs and org-mode, you can take just a little time to learn a few shortcut keys and the basic formatting, and you are off. And, yes, then you can spend forever learning more and making it the way you want. But critically, you don't have to and can be up and running as quick as anything else.
Hum, meh, personally I've jumped the ship from Unix to Emacs few years ago: in a month Emacs became mine most used application, in another or two my windows manager (EXWM) and if few more months I decide to devastate my decades-old hyper-curated home taxonomy to put all my files in a cache-like tree handled via org-attach, accessing them mostly via search&narrow before with linkmarks + manual org-mode, after with org-roam. I have almost anything in Emacs form emails to personal finance, files, agenda, ... everything integrated in ways no other tool I know of can give...
The learning curve exists of course, but it's not that hard if you have a bit of IT background and in any case it pay back so much that's absolutely worth the initial "capex"...
Not sure if you're referring to Emacs or org mode.
Org mode has an extremely shallow learning curve. I literally watched Carsten Dominik's Google Tech Talk and was immediately productive. Every time I wanted to do more, the information was easy to find.
I’m the opposite. I love emacs and have been using it since the previous millennium, but no matter how often I try, I just can’t get comfortable with org-mode. I spend far more time looking up how to do stuff than I do writing, and I never seem to be able to remember anything.
The editor understands the text files in a systematic way. I can jot down a date for a todo as I’m taking a note in my daily journal, and as long org-agenda can look across all my text files and show it in my agenda.
Plain text sucks. It’s so hard to represent context, time, location and state.
I have got to the point I just use the reminders app in apple ecosystem. Notes go in the notes app. Things get deleted when done or disinterest kicks in. Shortcuts fills up regular tasks for me.
I quickly craft tiny ad hoc DSLs as needed. staying in text lets me control my own data, minimize spending, always work offline, have easy version control and backups, maximum multiplatform support, and maximum futureproofing. not perfect but I usually gain more than I win
My gains are sync across all my devices instantly including the one strapped to my wrist, voice control, location awareness, context awareness, prompting, scripted automation, attachments and full tagging and categorisation system and I can share tasks with people including colleagues and family and check status.
I'll take that set of features over your self-imposed compromises.
I closed over 23,000 tasks so far with it while managing a full time job and a family.
With plaintext you can do what you need. For example, with todo.txt context are just prefixed by `@`. You can also add special tag with `key:value`. So for example you can add `location:paris` and retrieve it with whatever search tool.
Has anybody had success with plain text working in teams in business? I prefer it myself but teaching everyone I work with markdown, org-mode, git etc seems like a big undertaking. For example, we have a specification in development and that spec lives in Microsoft word. Plain text would make tracking the changes and collaborating on it much easier IMO, but word has it’s own abilities to do that. And with plain text you give up the wysiwyg formatting which people are accustomed to. Perhaps writing specifications in confluence would be better than plain text or word - who knows. It’s likely only because of my bias that I think plain text is a reasonable choice.
Plain-text means just something that can be manipulated easily, witch is damn good. But tools to properly and comfortably manipulate text are needed. For me the solution is Emacs that let me combine almost anything, access almost anything in a snap. It's a bit buggy sometimes, not really crafted for such usage, but even with such issues is still far above all other tools I know (and being a sysadmin I have see many software) to a point that I've developed the idea that after Xerox and LispM we humans completely lost the way of IT development just to lock users out of the real potential IT gives...
I've been a hardcore text-and-CLI-first-maybe-only guy for decades, as a workflow optimization rule. I coined my own private little name for it too: CLIFMO. I have my own development process and architecture strategy all associated with that name/abbrev. (Which I might write down and share one day, haha.)
one thing I love about having a plaintext bias when working is that many opportunities arise to design ad hoc DSLs, as I need them. when they begin to feel like a net win
its also fairly easy to start blurring the lines between text notes and working code, esp with shell scripts and langs like Python
I like the todo.txt concept and have given it a go for a while, most recently using the really nice frontend UI "Sleek". I ended up moving away due to the limited functionality of todo.txt as a whole. I needed to be able to add simple notes to my todos at times and it's just not possible or easy to accomplish this basic need with todo.txt.
People arguing about proprietary formats being supported by other applications and backups plus checksums being a solution against bitrot miss the major point:
Simplicity.
If "simple" text is all you need to fulfill a certain goal, just stop there.
Any system should be as complex as absolutely necessary, not a shred more.
And yes, plain text is good enough for a very many usecases.
If in the end I have to show ß or Ø or È it doesn't matter how it's encoded, but I would certainly argue for a unicode standard to be better than some proprietary format.
KISS doesn't mean that you can't have nice features, it means to have nice features with meaningful complexity
I believe plaintext/markdown as a format is a bit overblown, though I am glad there is a wealth of tools and tips around it.
Just as an example as a Hebrew speaker who regularly uses math - MD editors that support both RTL and math rendering are difficult to find. And that's squarely inside the domain of a "knowledge worker".
Emacs supports both RTL and math rendering (and not just in preview but inline too). I think this kind of flexibility is part of what is driving the popularity of plaintext… there are dozens of modern web apps for everything, but if you need an idiosyncratic combination of features, old-school tools are sometimes one of the better options.
> Windows, in my opinion, is far behind Mac OS X, iOS, and Android, in having thoughtfully designed and efficient software—both in general, and in particular for writing, organization, and task management.
Jesus, I thought it was just me and my long-time Apple bias (as a person who grudgingly uses Windows a lot of the time)
How come we don’t have plaintext symbols for the formatting basics: bold, italic, color?
Maybe we could expand Unicode based on lessons learned from Markdown/typesetting & the formatting characters could be added as non-printing characters. Visibility could be toggled with other whitespace characters.
David Allen is a leader of The Movement of Spiritual Inner Awareness. Between cargo cult and actual cults I have a low opinion of programmers in general. Programmers think everything needs a system or an abstraction. I guess when all you have is a hammer everything is a nail.
The guy who wrote the article is Michael Descy. The article mentions GTD a few times, THAT is what David Allen is responsible for, but the article also mentioned the Autofocus system, which was a new concept to me and which I am now quite fascinated by.
> cults... programmers
I've been coding for 20+ years and have yet to see any correlation between programmers and any cult (cargo or otherwise... ok maybe cargo SOMEtimes, but certainly nothing else)
GTD seminars are used to induct members into the cult. It's analogous to Scientology and their ideas on productivity. The audience for GTD is mostly programmers due to corporations sponsoring the seminars. Everything programmers do is ideological. Programming is also hard which is why coders are susceptible to scams. Uncle Bob and OOP principles are very cult-like as well although the methodology has proven itself suboptimal and convoluted.
Any thoughts on the Autofocus method? It seems to have had a heyday I missed about 10-11 years ago with apps made very specific to it kind of being end-of-lifed as of 5+ years ago (sadly), but it looks useful (and not at all culty)
With all these plain text posts surfacing regularly, I'd love for the lesser-known org markup (https://orgmode.org) to gain more adoption. It's a real power-house. Its Emacs origin may put some off, but it's plain text, so your content can be ingested/consumed by either regular text editors or any app focusing on specific user-journeys.
I built two org-powered apps for iOS myself:
https://plainorg.com
https://flathabits.com
There are other great ones out there:
https://beorg.app
https://braintool.org
https://easyorgmode.com
https://logseq.com
https://organice.200ok.ch
https://orgro.org
http://orgzly.com
Thanks to Karl Voit for driving org markup awareness outside of Emacs via Orgdown https://gitlab.com/publicvoit/orgdown. He's also got a great post showcasing org strengths at https://karl-voit.at/2017/09/23/orgmode-as-markup-only