What are some of the smaller or bigger features that would make Memex incredibly useful to you? Dreaming big is encouraged 
Mindmapping. I don’t know how exactly this would be implemented, but I would love to see some way of creating a visually linked mindmap between the various pages, notes, tags, etc… Various mindmap/knowledge systems exist - MindMeister, Xmind, TheBrain - but they are either clunky, TOO advanced, or not really sufficiently integrated into my workflow.
A dream piece of software would be a combination of OneNote/Notion/Evernote, Pocket, Feedly, email, and a mindmap to connect it all somehow. A way to collect your inputs (email, RSS, saved articles/webpages), organize them (with tags, notes, etc…) and then make new connections and retrieve ideas.
Memex is clearly going to become a great replacement for Pocket, but I’m not sure how it addresses the other stuff yet.
2 Likes
Integration. I don’t want Memex to be a browser feature I forget about. I want this annotation layer to be everywhere. So I can search and summon Memex links and annotations instantly, drop them on other documents and connect them together. I want to be able to pull permitted annotations into content management systems easily.
I know some of these are facilitated by the architecture (query storex and the open annotations api) so i’m not saying they’re lacking or not in the memex vision – I’m just emphasising that to me the key 10x feature is effortless ubiquity.
1 Like
Thanks Archer for the input 
Which tools are you using at them moment where you’d like to do this kind of integration?
I am also down to jump on a 30min call with you if you like and talk through this.
On the annotation request: Yes this is where we aim to be going. Now you can already share your notes, pages and collections, and have the TextExporter to copy paste them into your writing workflows.
Next up is a Readwise integration, and more collaboration features, like Hypothesis style discussions.
1 Like
Yeah we realised this too in the past few months. Memex place is to handle and distribute your knowledge input, so it can more easily be brought into your writing and sharing workflows in apps like Notion, Roam or your own blog.
I am also with you very interested to understand your workflows more, so if you want, lets have a 30min call to talk through this.
1 Like
I just checked in on Memex the other day after a few months away from it and was happy to see that you’ve abandoned the full history search - I emphasized this in other comments (here and here) that we don’t need a searchable backup of the entire web - Google does that. We need a searchable, organized, integrated backup of what is actually important to us - which might only be 1% of what we actually browse. Its unfortunate that so much time and resources were lost on it, but live and learn! You’re more focused now, so that’s a worthwhile lesson.
I wrote about my tools and “workflow” (if you can apply such a term to such a mess) a bit in this thread.
I’m still happy with TickTick as my task manager, Calibre as my ebook/pdf organizer (and now epub annotator, a new feature that I helped beta test over the past month). You can’t escape Office (or a similar suite), and it covers my cloud storage needs.
What has changed since then is OneNote, as I’ve discovered the new wave of “networked notetaking” tools that are coming out - Roam, Obsidian, VS Code Foam, Logseq etc…, which are starting to do a competent job with the “mindmapping” aspect that I mentioned above. This is what I’ve been dreaming of for 6+ years (and even tried to make a very crude version of using wordpress and d3js a while back). I’ve already exported my entire 1 million+ words of notes from OneNote to Markdown files and am slowly going through and “refactoring” them into something sensible/usable. I’m doing this with a combination of Foam and Obsidian - they both offer different complementary features - but will likely end up using Foam long-term for many reasons.
While thrilled with all of this, what I still see missing is the input/information gathering part of the process, and this seems to be where Memex can fit in nicely. I still somewhat use Pocket to save articles, but it’s not really a part of my process at all since it isn’t locally stored, doesn’t have annotation etc… I just save articles and they never get integrated into my process.
Some people solve this by copying right into their notetaking tool, but a) Markdown based tools are just too limited and b) there is too much friction to do all the copy and pasting from web anyway - I never did it with OneNote, despite it being quite capable at storing anything. I’d rather have it done in the background for me (once I choose for it to be done, rather than storing everything).
Memex seems like the perfect solution here, as it will do everything Pocket can do, but since it is locally stored it can be better integrated with non-web applications. @Archer mentioned this, but I’m less interested in pulling into Memex than making the collected information actually useful by pulling from Memex - perhaps with a sort of “memex://” protocol that can link to specific page, annotation, collection, etc…
This would allow me to locally store websites (ideally viewable, or even just stored, in “readable” mode - stripping out the website structure and leaving just the article content), annotate them, organize them with folders/collections and tags, search them, and then pull them into my formal writing tool as needed.
I suppose it would be nice if Memex could also store, annotate, etc… for PDFs and even other ebook formats - Calibre’s new ePub annotation is great, but lacks (and likely will never have) this “linkability”. But this is less of a priority for me than Web integration/collection/annotation -I’d get that nailed down before you continue working on PDF management. People are mostly satisfied with their PDF/Document workflows - local folders, various PDF readers/annotators, etc… Whereas there is no satisfactory tool for web collection/annotation.
With all of this, I could have all my “discovery” stored in Memex, and (perhaps with global keyboard shortcuts) use Obsidian/Foam (or Word, or any other app) to pull relevant info into my documents where I will formalize it all. I have zero conception of how the software actually can or does work, but since you were initially building a local, searchable backup of everything that we browse on the net, it seems like this should be possible!
Anyway, I’m happy to chat whenever - let me know when works for you. I am in UTC-6 timezone, so perhaps 8 hours behind you?
Edit: nevermind - I see you posted a link to your scheduling tool. Very slick. I’m signed up for tomorrow. Chat then!
1 Like
Automation via IFTTT
Automation via IOS Shortcuts
Option to relocate the offline storage - I don’t want to take up a lot of valuable SSD space on boot drive
1 Like
I would really benefit from a more featureful search for tags in tag filter portion of the Memex dashboard search bar.
Right now, when filtering by tag, when you enter a string in the tag search bar it appears the search only matches tags that begin with the string you entered. It would be extremely useful if the tag filter search returned tags that contained your search string anywhere within the tag. Regular expressions would be even more useful, but not critical.
As an example use case, my workflow is to add as many tags as I think might be useful, so I can recall the page I’m tagging from many different mental contexts. I also use nested tagging. As a (somewhat convoluted) example, a page about a neat Jupyter notebooks feature might be tagged with any (or all, if I can remember to) of the following:
- programming.python.jupyter
- jupyter
- datascience.tools.jupyter
- programming.python.jupyter.features
If I type “jupyter” into the tag filter, I would like the tag search to pull up all of these tags so I can select them all in the event that I can’t remember which exact tag I placed on a page about Jupyter notebooks that I’m trying to pull up. It’s likely that my tagging approach is a bit of a mess, but I think this would be a useful feature regardless.
1 Like
Hi all,
in most close-reading techniques I’ve seen so far highlights fall into roughly half a dozen categories and it is suggested to either use different colors for these categories or different symbols.
these categories actually express how what you read fits into your frame. what supports your own view, what expands it, what contradicts it asf.
so additionally to the tags I’d love to be able to differentiate at least the following types of highlights, I usually use five colors for these:
- inspiring novel idea
- central claim
- passage I’d quote
- passage I disagree with
- passage I don’t understand
long-term benefit of this would be that you could show not only which passages have been highlighted most by a community but roughly gain an orientation as to why, i.e. which passages an entire community of people doesn’t agree with or understand etc.
we can’t do that with tags because they are personal and different.
kudos,
Mario
2 Likes
Being new to the community, and being not very privacy conscious I suppose, I would have imagined something named memex to be a full-fledged index of all my online activities, including audio and video, without the need for highlighting or bookmarking but instead with the feature of “smart” indexing based on my history of searches and interests in general – the equivalent basically of a full-text searchable perpetual screen recording across devices with automatic content classification and ranking. Not too much to ask, right?
button to schedule when you are about to read the content of a webpage/article and e.g. set up a livestream on yt, twitch etc. for that date. thus others could join to read the source together.
it did used to track all of your browsing history automatically.
But it was removed. I use something else to screenshot every 30 seconds (ManicTime).
Hey thanks, that does sound good, especially if it is available for mobile and capable of automatically parsing the saved information into a full text searchable storyline.
If we had an API, would Zapier support such a workflow?
Btw we are about to release the collaboration features, where you can send invite links to other people so you can annotate pages and collections together!
If that’s an interesting one for you I’d love to learn ab it more about your use cases.
1 Like
if you had an API it would immensly help the “interoperability” you were talking about.
most non-techies I showed the memex to were put off that they’d have to use a plugin. it apparently is not a contradiction to be an academic and yet totally disinterested in technology. shrug.
we are a team of junior-developers supported by a senior dev and a designer who in the context of a mentorship wanted to help us keep up with newest technologies and best-practices.
we like the memex-idea a lot and thought about building a UI that could use clean-data editor such as https://editorjs.io/ to publish papers/essays and a potential memex API to add annotations without using the plugin.
drop me a line if you like and are open to the idea,
Mario
Thanks for the good feedback. Really helpful!
most non-techies I showed the memex to were put off that they’d have to use a plugin.
Yeah I know. The extension infrastructure is not ideal. We are now moving towards making the memex service cloud first and with offline support.
Offline first has too many drawbacks and complexities for our small team and is also not user friendly.
It was necessary for sensitive information like full-text history search but not now anymore.
RE editor: Actually one of the things I started planning out is the editor, and also eyeing at editor.js.
Also API we started conversation recently. Good timing. 
Yeah let’s chat. You can check for a time here: calendly.com/worldbrain
Happy to take 1h for this.
1 Like
great! will discuss it with the team this Thursday and come back to you! kudos
Might I suggest alternatives to Zapier that some people self-host?
1 Like
Of course! Very much!
1 Like
I might have to post on the n8n.io - about their API’s. But posting this for now as an an alternative.