12 June 2022 – Logseq Updates 0.7.3, IOS 0.7.2, Namespaces, Reading List, Daily Workflows, Plugins

This week, I am back in my old stomping ground for a few days. After having lived in London for 15+ years, at the back end 2021, my better half and I moved to the UAE. It was all a bit sudden and there were still some Covid restrictions so we could not properly say goodbye. It is nice to be back and we already passed by our old house, saw someone there, internally felt bad and walked on. Later today we will be doing touristy stuff – first up though, is passing by my beloved Craven Cottage (haven’t told the wife yet, so going to be interesting masking it in on the “scenic route”).

It is funny how when you live in a place you always think “I will do that tourist thing one day”, and then never end up doing it. You leave and think “I could have done that” but it is too late. So moral of the story, if you have the opportunity to do something now and act like a tourist for a day, just do it, you will enjoy it.

With my 2 cents (or better yet, pence) out of the way, let’s see What Logseq had in store for us this week.

Releases

The devs released 0.7.2 over the past week, and pushed out a quick follow on release (0.7.3) to fix a couple of bugs. Below are the main takeaways from the releases:

0.7.2

  • Fixed issues
    • Bugs introduced by changing of Copy & Paste logic
    • Block lost when moving, cutting, pasting
    • Handle graph directory change and file inode change (compatibility with some third-party file/folder syncing tool)
    • Performance regression when re-indexing
    • Missing page icon from left sidebar page item
    • Missing app menu under macOS
    • Highlight refs lose association with related pdf
    • Missing katex.min.js in publishing
  • Enhancement
    • Norwegian dummy-notes.md and tutorial.md, Norwegian translation
    • Update mobile.md
    • Korean Translations
    • Predefined view and transform
    • Add edit button to queries
  • User Contributions
    • Norwegian dummy-notes.md and tutorial.md – ingepettersen
    • Norwegian translation – ingepettersen
    • Update mobile.md – simonla
    • Korean Translations – Sunghyun Cho]
    • Reduce shadows on sidebar and inline cards – ranfdev
    • Predefined view and transform – Darwis
    • Add edit button to queries – sawhney17
    • Fix missing katex.min.js in publishing – WsinGithub

0.7.3

  • Fixed issues
    • Regression on Windows and Linux, accessing parent directory, causing high memory consumption
    • Cannot unindent multiple blocks
    • Unable to perform block operations in the document mode
    • Cannot paste plain unordered list in markdown
    • Missing DB persistency when creating new window via file menu
    • Edit button should not be shown in default queries
    • Should show bullet when hovering a block on the document mode
  • Enhancement
    • Support mod+e to copy block embed to the current block
    • Allow query property keys with char _
  • User Contributions
    • Norwegian translations – ingepettersen
    • Support mod+e to copy block embed to current block – sawhney17
    • Edit button showing in default queries – sawhney17

Remember you can download the latest version of Logseq from here.

Logseq News / Events

  • Last week Logseq held a Town Hall community call. It was a very good listen and as soon as the recording is released, I will tweet a link.
  • Please join me in congratulating Aryan on joining Logseq as a summer intern. The amount of work he has put into Logseq, from the community, plugins, code etc etc. this is very well deserved. Here’s to a great summer!

Logseq Mobile

To access the mobile apps, please use the respective link for your OS below.

Logseq iOS app 0.7.2 is out. You can download or update it on App Store.

  • Features
    • Switched outdent and more buttons to make right swipe more natural
    • Disabled the swipe action-bar when editing block. Only allow indent/outdent
    • Resolved conflicts between selection and swipe
    • Remove the indent/outdent button on the mobile toolbar
    • Fixed toolbar not showing when using Apple Pencil or external keyboard on iPad

Logseq Workflows

@Luhman gives us a very good summary of Obsidian vs Logseq. “The biggest factor is whether you prefer working in an outliner or using plain text. Both can do the other, but each is optimized for one of those two options. Most people using Logseq really want an outliner. What I like about outliners is the ability to get context about text by its location in the hierarchy, and to quickly move to different sections of that hierarchy. Plain text is flatter and requires more use of tags and/or folders to get context.

If you want to hear of my thoughts on why I think Logseq is the best app out there, please do check this post out.


Want to have your sidebar appear on hover? Well @Aryan, gives us this magic.

All you need to do is add the below to your custom.css file:

div#left-sidebar:hover {
    transform: translate(0%, 0%);
    transition: translate, 1s;
    opacity: 1;
}

div#left-sidebar {
    transform: translate(-99%, 0%);
    opacity: 0.01;
}

A little bit of a selfless plug. I recently implemented a reading list in Logseq, in great part thanks to the inspiration of Felipe. The reading list is based on a template which includes the below. I then run a query to display it in a structured format.

  • Meta
    Type:: [[Books]]
    Status:: [[ ]]
    Cover:: ![Image]( ){:height 172, :width 98}
    Title::
    Author::
    Link:: [Goodreads]()
    Start::
    End::
    Year::

If, like me, you have seen a recent rise in the use of Folgezettel and wondering what this is and what / how is this related to Zettelcasten, then here is an explanation: “A Folgezettel is a specific and intended connection between Zettels, realized through their position in the archive. You put a Zettel in a specific place and end up with a Folgezettel, which is nothing more than a connection between the preceding note and this new Zettel as a Folgezettel”


@ADHD Optimzed asks: So I’m going through a load of educational podcast episodes on neurology and practical use in daily life which I am intending to summarise and reduce to personal cheatsheets for personal use. I havent listened to them all yet and I think don’t yet need to create pages so I am shortening the titles to the topics covered before I really get into it. I’m already finding that the topics in the titles alone repeat themselves and have sub-topics.

  • eg How nutrition affects mood
  • eg How hormones affects mood
  • are “#mood-nutrition” and “#mood-hormones” at the moment
  • or those of you who have been doing this a while is there a way to do what I guess would be nested tags or are [[heirarchical/pages]] more functional long-term?

I added the word-frequency plugin but I haven’t used it yet I don’t know how useful it will be to help me separate what bits are “how-tos” and what is “background info”

@Alex writes: You look at it from a hierarchical point of view, instead you should turn things around: not where (or how) do I store things, but how do I want to find things.. I don’t know what you need so I’ll describe what I do:

  • Directly link things that are directly related with [[links like this, that are obviously unique]]
  • But if your main notes are atomic enough then you can still reuse them because they use [[links like this, that are obviously unique]]
  • Things that are ill-defined, but that should be findable have #[[shorter tags]] or even just #tags (I like to differentiate them by using # for vague groupings vs [[precise]] links
  • You can use #tags to find all items, but more useful are queries:
    {{query (and [[tags]] [[shorter tags]] (not [[other tags]]))}}

This setup will make direct links obvious, but also linked associations simple to create. At the same time you don’t have to decide beforehand where to store things.

Fantastic question, and fantastic response provided shortly after. The Logseq community is really a place you can rely on.


Thinking of automating your daily workflow? Then @Aliphys may have what you are looking for here.

@Aliphys provides us with a template of the important things they want to reflect on, and tackle each day, and with the additional of the SmartBlocks plugin, you can see the data from a different point of view.


Say I’ve got a giant doc I’m porting from org-roam to logseq. I’d like to break it out into some smaller documents (I think) because the original was ~500 lines long. So the way I’d usually approach this is to work through and cut/paste the lines into another doc, but that breaks block refs/embeds. Is there anybody who has migrated things from bigger single documents into smaller logseq pages who has recommendations on that?

Response: Try the block to page plugins. it’s very easy to convert the bullets into pages and divide up the document that way

Great back and forth between @doulos05 and @Sawhney.


Wanting to keep Logseq always on top (as asked by Ntotao), then you might want to try this.

Alien Dude 21: I use a utility called Deskpins. It is free and it adds a little pin icon in the system tray when you run it. You drag and drop the pin icon on the application you want to stay on top. I use it to pin on to smaller ppt slides in read mode when I carry out online training so viewers can follow the agenda. I haven’t tried it with Logseq but it is worth a go.

Also check out these free alternatives (by Cannibalox) – Microsoft Powertoys (what I currently use as it offers other nice qol improvements) – AutoHotkey has an ‘Alwaysontop’ one-liner command


Idea to have whole company knowledge in one graph (link for full details). So I’ve been using LogSeq at work for research purposes for the past few months and it’s helped me a lot. Today I was thinking about that all this information is only accessible to me and if anyone else wants information, they would have to ask me first and I would copy some blocks into a document. This got me thinking: we already have a shared network drive where everything is synced automatically, so why not have a LogSeq graph inside this drive?

Response (one of many): I don’t think the workflow logseq drives is right for this. Logseq has no features for managing at scale. My view is that ‘digital gardening’ workflows are largely a personal endeavour, not a collaborative one. Before you know it, we’ll end up with pull requests or similar to allow edits to be made on a topic.


Creating a Travel Book? Great question XoTrm. Say you want to document places that you have been too, or food you ate?

@Jangus Won responds:

  1. Take a restaurant I want to input into the system
  2. Create properties that are relevant to what the restaurant means to me. I’d make the propertyValues:: [[pages]] (i.e. city:: [[New York City]])
  3. Repeat for a couple restaurants (3-5)
  4. Extract the properties (i.e. city:: , cuisine:: , etc.) that I use the most and create a template out of them so that I can add the properties immediately after thinking of a restaurant

Why do this? This is more of a philosophical point, but what you’re asking is the essence of bottom-up vs top-down organizational thinking. Bottom-up thinking tries to counteract the problems of top-down thinking which are:

  • Encountering friction every time you have a restaurant you’d like to input into your system
  • You know exactly what the restaurant means to you but now you have to put that thought on hold and go through your system hierarchically, asking whether or not the restaurant fits in this or that category, increasing friction and scattering your attention
  • Asking your categories if they’re right also introduces the problem that restaurants (and most any data) fit into multiple categories (i.e. is Olive Garden fast-food or Italian?)
  • With the properties way, structure emerges naturally as you input more data.

Namespaces to Create a Dynamic Index in Logseq? @Josh with another great video:


Plugins on Mobile? Not yet possible, but on the Town Hall meeting call it was mentioned that the devs are making progress in implementing this. More news to follow.


Although, not posted in the past week, I think the message posted by @Smithy will resonate with a good few users. “I come from a background of years of taking notes in OneNote 2007, where everything is laid out in nicely presented folders and sub-folders (I stuck with the older version because it wasn’t cloud-integrated). This new way of note-taking, writing everything out in the daily journal and deciding afterwards where it goes, or where it connects to, is mind-blowingly powerful compared to the old way of taking notes, but boy is it tricky to get started.”

@PaulRudy provides a brilliant response.

  • In Logseq, however, you can have namespaces, which (to my understanding), means hierarchical tags. So your tags could include, for example, #system/nervoussystem/musculoskeletal#system/circulatory and so on. /nervousmusculoskeletal, and circulatory in that case are namespaces within #system. You could go on to #system/circulatory/venous#system/circulatory/arterial, and so on. This is one aspect of hierarchical organization that can allow you to separate from folder hierarchy.
  • Another important point in reference to being less attached to discrete pages: With block hierarchy, any parent block is able to perform functions of both a discrete page, as well as a discrete folder. To think about this, look at a normal folder with text files in it. If you re-imagine that folder as a parent block, and ever file in it as a child block, and if there happen to be subfolders, they too become children, with child-blocks of their own. This is what blocks allow you to do in Logseq. And blocks can be large-document length, of course, but they’re perhaps most useful as bite-sized pieces.

Definitely worth a read to see the response (and question) in full here.


@Dario recently updated the Logseq Mastery Course. There is a lot of amazing content (over 21 000 words, 7 hours of video, and nearly 40 Excalidraw diagrams). I would consider myself to be a decent Logseq user but still found quality content in this course. Dario is also a very active participant in the community and posts a lot of YouTube videos for free. Even if you do not take the course, he is definitely one to follow.

Full disclose: Logseq Mastery is a paid course and I am in no way affiliated. I simply believe in showcasing quality product so that you can make your own mind up.


You may have noticed a few bugs being introduced or old bugs resurfacing following releases. @Zyrohex asks: Question, is the dev team working towards doing more edge-to-edge testing prior to pushing out releases? I’ve been using the app for about a year now, and a struggle/pain point is old bugs resurfacing, or little things that would seem obvious, break in next release.

Response: End to End testing is a bit focus and lots of tests have been implemented / are being implemented to ensure we get quality updates.


Interview with AI direct in Logseq? Amazing!


How to sync Logseq via Git across devices and platforms? Check this brilliant article from Martyna. Goes through the process step by step with very clear instructions and images to help.


This week, I leave this section with this tweet / video from Jeffrey on having a chaotic mind. So much of your video resonated with me.


Logseq Feature Requests

You can check out the full list of Feature Requests here.

  • Reorder blocks inside query results – I have a query which gets all blocks with the tag #priorityA. Now I want to reorder the blocks inside the results of the query: I want to move the important blocks to the top with the shortcut Alt + Shift + Up and the less important blocks down with Alt + Shift + Down. Sadly, moving blocks vertically is currently not possible inside search results.
  • Improve the sync user experience – The normal way to sync Logseq content is through Cloud Storage Services (like OneDrive). However, the experience is not quite intuitive. As a user, I expect that settings and plugins are also sync along with contents.
  • Enforcing property value types – I would like to see the ability to define and enforce property types as required, and then the type, like a SQL database. I think more people would be able to move from Notion if this was implemented, because the risk of creating an error in Logseq with a missing or incorrect property value is preventing a migration to this great project for me. For example, in my intended CRM/project management use-case, if I enter: project_invoiced:: true instead of: project_invoiced:: true
  • Offline enabled multiplayer without cloud syncing dependency – I noticed that you’re currently working on both sync and e2e encryption for Logseq. That’s amazing thanks! I was wondering what the foundation of that was going to be? I ran across Automerge 2 recently and thought it sounded amazing for this use-case. It would be amazing to have a multiplayer editing experience across devices without being dependent on any sort of centralized storage syncing service.
  • Org mode column view support – Is it possible to have in Logsec an implementation of Org-mode column view? If #+COLUMNS property is defined, it would be possible to switch with hotkey to column view. If it is not so easy to implement value editing in the column view, maybe it will be enough to have read-only mode?
  • Have nodes in graph view stay put unless moved – When you move one node, it often moves several other nodes in the process. I wish all other nodes would stay put while moving a node. Also, I wish the graph view didn’t reset every time I returned to it. I want the ability to save the placements of the nodes. Implementing both of these would help a lot in keeping pages organized.
  • Independence of graph name from folder structure and reduced clutter – I believe that basing graph names on its parent directory name while there is no way to override this behavior by supplementing your own custom name is a very flawed concept. It’s easy to imagine that some people might want to organize their projects with structure like HOME/projects/PROJECT_NAME/. However, doing so will clutter project directory with 3 new folders created by Logseq, which decreases clarity and slows down navigation if user keeps some other files or documents related to this project in the same folder.

And still my favourite(s):

Logseq Plugins

  • Summarizer – v 0.0.4 – cut size down to 42mb by deleting the parts of the python standard library that were not used – use pip to manage python dependencies so we don’t need to commit them now – issue #4, option to nest generated summaries under source content – submitted to marketplace
  • interval-hints – Configurable Logseq plugin to annotate deadline and scheduled times with a countdown or interval.
  • blocknav – Simple plugin to facilitate top-level block navigation
  • go-now – A quick plugin to go to your journal page and append a block to it and also scroll to that point
  • gpt3-openai – A plugin for GPT-3 AI assisted note taking in Logseq. Allows users to generate human-like text using GPT-3 within the LogSeq editor. Write a GPT-3 command in a block, then run the open gpt3 command via the slash or block menu. The plugin will generate a GPT-3 response using the OpenAI API and insert it beneath the block. Genius from Brian.

Many thanks as always for reading this latest edition of Logseq Weekly.