avatar image

Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
  • Saturday May 21, 2022, 9:08 pm
  • Proudly celebrating 15+ years online.
  • Donate now to become a lifetime supporting member of the site and get a non-expiring license key for all of our programs.
  • donate

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - JavaJones [ switch to compact view ]

Pages: [1] 2 3 4 5 6 ... 106next
I have a multiplicity of needs and tools. I'd rather have fewer tools, but nothing does everything I want yet.

I have a daily "life logging" practice that is a combination of work and personal items, including some "quantified self" stuff like when I woke up, how much I weighed, what I ate, etc., as well as work periods and tasks done or things accomplished, documentation of thoughts, feelings, ideas, etc. One of the reasons I got interested in Roam-like linking approaches is because it's a lot less work to connect my daily log to individual pages where I can analyze, etc. So now I can link to e.g. [[Weight]] and I have all my weight entries drawn straight from my journal, I don't have to actually open the page to add that day's weight, the "backlinks" feature shows them all for me just by linking. And that way I can read them in my more narrative journal too, without having to open a bunch of related pages to get the full picture of a day. Best of both worlds. That's just one example. Right now I do all of this in Roam, but plan to move either to Logseq or Obsidian, and maybe Codex long-term.

I also keep track of a lot of life, project, and activity stuff like Dormouse and Wraith. Pages for projects, entrepreneurial ideas, etc. I find the outline structure too rigid for this stuff, so I used to use Quip (and still have a lot of data in it), and have been slowly migrating this stuff to Notion. But not sure I like Notion enough to keep it all there... Depending on how well Anytype shapes up, I might be able to move the database stuff there.

Then there is regular data sets with some reasonable amount of commonalities, i.e. properties, that I want to store, sort, etc. Things like Recipes, Articles or Books to read, etc. These all go in Notion databases at present, they used to be simple lists or header-divided pages in Quip. The databases are better, but still not perfect.

Then there is the actual writing I do, mostly for planned blog posts (I say "planned" because I post them very seldom). This I do in Notion right now, it's the closest analog to a "web presentation" I can write in, and it's faster and better than working in Wordpress directly for me.

Ultimately I'd like as much of this stuff as possible to all be in a single system because interrelation and data re-use is incredibly powerful, time-saving, and useful. As a simple example, being able to reference the data in a property/field of a database from regular text, or link to a database entry in entirety from some other page (Notion can do the latter but not the former), etc. Or even embed a database record, or an entire table (Notion can do the latter, not the former).

That covers most of the non-work stuff. For work I chose and implemented Fibery, but that's getting pretty tangential to the topic of this thread.

If you're willing to have your links, formatting, and other markup be separate from your text, then a stand-off approach like Codex uses may be the way to go. The Codex dev has claimed that he intends to make the stand-off markup exportable, though whether any other app will ever support it fully is questionable. But at least your base text files are untouched. That also means no links in the text files themselves though (e.g. [[link]] ). All that said, Codex is also not locally-based. But parts of it will be open source. So it might be something to keep an eye on...

- Oshyan

Sorry. My bad.

I'll go whip myself.
Not at all. In fact I wish people asked more about it, nobody seemed to care at the time (or now, I guess). Not Markdown-ey enough perhaps. ;D

- Oshyan

Anytype – an offline-first private alternative to Notion:
Yep, I'm in the beta. Mentioned it back on Page 28 :D

- Oshyan

for example, with neuron, there is a nice feature where you can use triple brackets in addition to normal linking (double brackets).  What this accomplishes is a more important kind of "link" than the regular links.  Regular links will work as expected.  These triple bracket links do the same, except on the index page that gets automatically generated, it will create a heirarchy (tree) based on the triple links.  So you can control this, wheras the other links cannot be controlled in such a fashion (normal links won't affect this heirarchy).

Interesting, I hadn't seen that. I like the idea of an ability to add some hierarchy, optionally. But not sure if I'd want to be done with *even more brackets*. 😄 I already find double brackets annoying enough. But yeah, I like the root idea.

- Oshyan

OK guys, after almost two years of following this discussion, I have achieved the setup that accomplishes the goal I originally be able to write books quickly, the way i heard about that luhrman guy.  The key is neuron for me as being able to see the website updated instantly as i type has been the most motivating feature for me.  Amazing stuff.

I seem to be using Obsidian mostly for writing.  Truly amazing setup, i have yearned for this for maybe 15+ years, you can see my posts here lolll.

That's amazing, congrats! Obsidian seems like a good bet to me in this space, just generally speaking. I don't agree with all their choices and am waiting for some functionality to come natively or in plugins, but compared to the mess that is Roam, and the proprietary nature on top of that, yeah...

- Oshyan

In our excitement at reunion, let us not forgot Codex.

I watched a few of the videos. It all seems a bit stream of consciousness from what I’ve seen so far. Is there a design document for it that's available to read? Maybe I haven’t watched the right video - or enough of them yet. But I’m having a little trouble getting my head around the central paradigm and design of the thing.

That's fair. In that case perhaps best to wait until the beta is available or something. Just know that it exists and, if you're interested in the potential that interlinking your knowledge has, then Codex may end up with one of the stronger implementations of it in the end. Particularly of interest to academics, which I think is his main intended audience, but from what I've seen it may actually make a great personal knowledge management and research tool as well. At least that's *my* hope and interest in it.

Will let you know when there is more to see...

- Oshyan

In our excitement at reunion, let us not forgot Codex. ;) Here are some sample vids to whet your appetite.

- Oshyan

40hz! Great to see you here again. :)

Thought everyone might want to take a look at The Codex Editor project, if you haven't already:
And check out his YouTube channel where he demos it:
This guy's work is insane IMO. *This* is the system I'm holding my breath for...

- Oshyan

Roam vs. Notion:

You see who that is by, right?  Of course if the creator of Roam talks about Notion, it will be in a deprecating manner.  Always check the source...!  He has major beef with Notion because they didn't want to collaborate, crowing over things that were preceded in other applications.

Exactly. And this is a big reason Conor does not impress me as a founder. He's petty, vain, impulsive, and egotistical. Look at the way he interacts on Twitter and elsewhere vs. someone like Phil Libin (co-founder of Evernote).

Notion's community engagement isn't much better, depending on your perspective. It's a lot more personable, but at lot less "real" at the same time. There is lots of "Good idea! We'll add it to our request list" and whatnot. But virtually zero understanding of whether that means any damn thing, what the priority of any given feature is, or even what the broad strokes vision is of what Notion is trying to accomplish and how they're actually going to continue achieving it. But I do like their design. ;)

- Oshyan

Notion officially launched their backlink features today. Still needs some work, but I think it might be enough to let me put more of my stuff in Notion vs. Roam/Obsidian/Logseq/whatever. In particular things that work well with databases like regular reviews, recipes, media lists (books to read/read with notes, TV to watch/watching notes, etc.).

- Oshyan

Logseq also has a public roadmap and API is on the list and being worked on. I don't know how/why but despite being very new, it seems to be further along in some key areas I care about vs. others (even vs. Obsidian in some respects).

Even areas I *don't* personally care about but which might drive adoption (and thus help ensure longer-term success) are getting attention, like localization. Does Obsidian have that yet? Or Roam? (I don't think so) Logseq does! Let's not even get started on Notion's localization odyssey. ;D

Maybe I just like the dev and his user engagement so far:

- Oshyan

Logseq is a local-first, non-linear, outliner notebook for organizing and sharing your personal knowledge base:

My first thought was "Ugh, another entrant in this crowded space? It never ends!". But actually Logseq may be the Roam alternative I've been looking for! They've already solved block *and* page embeds as well as syncing/backup (Github). In that sense they're ahead of Obsidian. Very interesting indeed.

-  Oshyan

I'm really interested in specific benefits of direct text editing, but so far I am hearing *that* it is useful and important to you, but not much about *why*. That's OK, it is enough that it *is* important to you, clearly.

As far as general advice about how to choose a tool(set) in the current landscape, I maintain my position that "database vs. text" is a false dichotomy. It is definitely more useful IMO to state specific roots of concern that *may* arise from those two paradigms, such as "data ownership" (not a given with a DB, e.g. if it's using an open DB format and/or all content is markdown and it has a robust exporter). Which is part of why I want to know these specific advantages and use cases for external tools operating on text files.

It may simply come down simply to it "feels right". Or is rooted in a more abstract value or moral stance such as unwillingness (perhaps reasonable) to trust a company with your data under any circumstances. That kind of foundational limit will pretty severely restrict your choices, but that's OK if you understand and accept the trade-offs, if that specific value is important enough to you.

In the end whatever helps you be productive is great. It may not help others choose tools as a heuristic for evaluation and decision, but it's certainly a valid stance.

@superboyac, have you played with mind mapping tools? I would imagine so, but don't they work exactly like "Scapple"? Only some are way more mature than it sounds like it is currently... I would go crazy without Search. ;D

- Oshyan

Speaking of online vs. offline and Anytype, I recently got onto the private beta for it and it's surprisingly good already. It's pretty much feature-parity with Notion for *pages* functionality, but has no usable database functions as yet. It's clear though that they will get there as there is a hidden DB that indexes *all of your pages* already. And that's exciting, if they keep it, because it means all pages will be able to have meta-data, be sortable, filterable, etc. in DB views. This is unlike Notion, where there is a hard distinction between Pages (stand-alone) and Pages in a DB.

Happy to answer questions anyone has on Anytype in present state, too.

- Oshyan

My recommendation for users is to make a decision about database or files (or any combination according to preferred workflow), pick the app(s) that seem to suit best for now and then just use it(them). Check alternatives only when hitting an issue. They're all going to change and develop dramatically (some will vanish) and it will be easier to compare in a few years. Everyone is aware of what the others are doing,  so the whole herd will add desirable features in a lagged sync.

While I agree with your overall point, again I think this DB or files thing may be a false dichotomy, at least for many people. I've yet to see a compelling real-world example of specific, practical workflows that would necessitate "files" that have no DB component. I think you may feel strongly that your own workflows demonstrate this, so I'd love to hear some examples of how you intend to (or already do) work this way and what significant advantages it brings you (vs. for example an Obsidian plugin that does the same thing as some external tool you use directly on files).

In other words I think there are other ways to achieve/solve the feature/functionality desires you have that make you want to work with "files", while still having nothing to do with a files vs. DB distinction. Whether you want to accept whatever sacrifices that might entail, such as having to pay for an Obsidian plugin for example, is another matter.

I do however think there is a very clear and important distinction between cloud/SaaS (Roam, Notion) and desktop/offline, perhaps with optional cloud sync (Obsidian, Anytype). *That* to me seems the more important choice to make since all of these systems pretty much import and export various flavors of markdown.

- Oshyan

Notion now has bi-directional links/references. It's pretty basic at present, but hopefully means they will put some more attention into this type of functionality.

- Oshyan

But I can't think of a way to get convenient, reliable transclude atomicity in a set of every changing plaintext files without a paying a price re (2). Seems like something has to give. Bet: transclusion will prove so useful that Markdown syntax will be extended for it. Hopefully in a standardized way. NetCommonMark (Net as in networked notes) on the horizon? Once standardized other note apps would have similar transclude views so there would be less of an lock-in effect to the Obsidian editor.

I can't think of one either. IF you allow arbitrary programs to access *and* edit the data, then all bets are off, period. Unless, of course, those programs all support the same standards. And then it's questionable what value using different programs would have.

Atomicity and transclusion is completely compatible with file-based systems though, as far as I can see. It may have a performance overhead, but there is nothing stopping a tool like Obsidian from having a "working database" or sidecar XML files or whatever it needs to support those raw text files to have extra features like block references and transclusions, etc. There is a repeated distinction made here between "database" and "file based" in relation to things like references, network views, and other things, and I just don't see them being that related. Having everything in a DB may make it easier or faster to do certain things like transclusion (since in theory all text is dynamically generated from the DB in all cases). But as far as I can see a reasonable version of similar functionality can be created using offline, text-based systems with augmentation. If I'm wrong someone please explain it to me. :D

- Oshyan

Obsidian can't export to markdown because it doesn't contain the notes in the first place. The files have an independent existence and can be edited using any program at any time.

I don't understand the point of that distinction. They don't need to be exported because they're *already* markdown, aren't they?

- Oshyan

The boom in interest in both Roam and Obsidian, and various third-party stuff, is both good and bad. Good in that great things will probably come out of it. But bad in that things are still so much up in the air that we don't know how much these tools will change 6 months from now. So I still find it hard commit to either ATM, since which one (or other) I will prefer in the not so distant future probably depends on features not yet there or not yet refined. I'm kind of stuck in test drive mode 8)

That's one advantage that Obsidian has over Roam; there's not much in the way of commitment since it sits on top of text files.  I'm not currently using it though I maintain the same folder structure as when I did- so I can pick it back up if I desire at a later time.

I'm not clear that Roam is that much more of a commitment than Obsidian in that they are both based on Markdown and exportable to same, right? So yes, you might lose the proprietary features, the interlinking, transclusion, etc., that makes Roam cool, but you don't lose what you've written, or at least you don't have to (export it). Am I wrong on that?

- Oshyan

That's an interesting take on the situation.
I share concerns about the professionalism of Roam's development. The offline enticement to believers through a PWA, risking data loss if cache is deleted before sync with the online database, just seemed the wrong way for the program to work.
But I hope it does well because a database has advantages that files don't - even if I prefer files for my own use.
Are the advantages you're talking about really a distinction between "files" and "databases", or are they separate? I would suggest that in fact you can probably do all or most of what Roam does with "files", or at the least by adding a database *in addition to* the files (i.e. a database that manages the files/interconnections). Which is to say I don't think Roam has any kind of monopoly technologically on the benefits it purports, and Obsidian could replicate all or most with the will and the time put into dev. Correct me if I'm wrong though.

One is that he's just substantially expanded the team; I have no experience of development teams, but in other areas it's not unusual for it to take time to become productive rather than a drain on existing resources.
Is there any evidence that he did, aside from the customer relations person I think he mentioned?

Hopefully they predicted this and had stuff like the pomodoro already up their sleeves to give an illusion of movement.
Does that really track with your knowledge of how Conor has been operating though? That he would somehow develop something in advance but not release it just so he could then release it later to make it seem like there is progress happening during an otherwise slow period? Doesn't sound like it matches with what I've seen, at any rate.

The second is that a huge number of cultists are desperate to fiddle personally with the program. Mostly, it seems to me, because they have a drive to fiddle (enhanced by lockdowns) rather than a particular need.
This is true and I have no problem with them doing so, in fact I think it's great. What I was pointing out here was that the RoamMonkey features were, for the most part, *really effing useful* and *should be in the core*. In fact Roam added a new feature a mere few days after that RoamMonkey vid was released which does something similar to the Template feature. It is more powerful than RoamMonkey templates, but unsurprisingly (because it's Roam and Conor) it's *harder and slower to use*. Anyway my point is that this is not someone fiddling just to fiddle, this is a smart person (RoamMonkey author) seeing *important* things missing from Roam and spending his time *as an amateur* to develop them externally.

This keeps them attached. Other programs like Trello have benefited hugely from third party enhancement. And I notice the building excitement in Obsidian over the near-term API release.
Trello has an API, as you mention. Obsidian will also have an API. These are reasonable programs to develop for because they give you an appropriate channel to do so. Roam does not, yet. It lets you embed live-read CSS *and* javascript in its fricking pages! That's an insane risk, if I know anything about web tech security. And judging by my own experiences of massive slowdown in experimenting with CSS theming in Roam while the system tried to interpret what I was writing apparently in realtime, I think it's a crappy way to extend a system anyway. Even the RoamMonkey dev admits the fragility of some of the things he's built.

So it's great people have the enthusiasm and interest to build stuff for Roam. But I *don't* agree with Roam dev's decision to allow live interpretation of web languages within Roam DBs largely to enable this kind of hackery. I get that they probably wanted to encourage hackery and didn't want to have to wait for - or spend time on - a proper API at this stage. But I think it's irresponsible. I think there are a huge number of "cultists desperate to fiddle with Roam" because Conor is encouraging that, but I don't think he's doing so in a good way for a professional app.

Let me say one last thing which hopefully clarifies where I'm coming from. If this were a free - and especially open source - app; if this were something people could host on their own machines and expose themselves to the risks by individual choice; then I would have less concern with the "lack of professionalism" that I think Conor is exhibiting. But no, they're charging what is fairly clearly a premium for a hacky, messy system that, yes, is super cool, but also is developed largely at the whim of a potential egomaniac, and the community and his relationship to it is just feeding that. IMHO of course. ;)

I get why people are excited to be a part of it, that very messiness has some exciting aspects. But people who have lost data have a more realistic view of it, and there is the potential that more data will be lost because sloppy devs and irresponsible management lead to things like that. If that does happen, hopefully people will see that it was questionable to put that much faith into the "roam cult" and Roam and its founders, because at the end of the day people are excited about doing real work, better work, and to do that you ideally need a reliable tool. Cool, new, and exciting only get you so far.

I'll take Obsidian's humble, experienced team approach any day.

- Oshyan

Roam Monkey:

Here is the maddening dichotomy of Roam, Roamcult, and Conor laid bare: Things like this - which, by the way, looks really useful - are a testament to the flexibility of the system, but also an indictment of the development priorities of the team lead.

It is on the one hand nice that Roam is flexible and extensible enough that such things can be done within it, despite them not having access to the code. That is really cool, in a way. It allows the community to help cover things that the core devs cannot, and without open sourcing, since not all developers/companies and profit models are amenable to doing so. It does present some security risks to allow this kind of stuff though, even with the warnings.

On the other hand, the security risks are arguably the smaller concern here. A majority of the features in RoamMonkey are highly useful to most people. They are, in fact, largely features that should be a core part of Roam! Maybe they will be some day, but here's what frustrates me so much. These features were achievable by a self-professed amateur JavaScript developer using only external access. Surely this could be done as well or better and faster by the internal developers. And yet, in the same several months that this guy has been developing these highly useful features that almost everyone would use, Conor and his partner(s) have added a Pomodoro timer, Mermaid diagrams, and other arguably niche and certainly not "first priority" stuff. This is behavior that users SHOULD NOT accept from a developer of an app they are paying for (and, I would add, paying a premium over many comparable products).

Roam is doing cool stuff, but I refuse to support Conor's self-indulgent approach to development priorities. It's not going to work out well in the long run.

- Oshyan

Thanks for the responses so far!

Clever and good UI design can still be simple, even minimalist. In fact cutting down a design to the bare minimum necessary is a huge skill in itself. So examples of good minimalist designs that *hide underlying complexity* (either in functionality or in the concepts/knowledge underlying the app functions) are absolutely useful! Making a simple UI for a simple app is less impressive and interesting, of course. ;)

The first one i think of is a software i use every day:
Great ui, everything is under your fingers, just before you need it. It's a programme to design in 3d, easy to use, light, great intuitive ui

This is exactly what I was looking for as my interest is actually specifically in 3D as well! I didn't want to focus too much in that area so as not to skew feedback. :D There are some nice aspects to that app, but if you have any more detail you can provide, specific things you like in the UI (especially vs. other well-known apps with similar functionality), that would be great. If not no worries.

Hi JavaJones,

if you are adventurous I will suggest the following,
1. Use mouse hover action
2. Bigger buttons or images in place of buttons
3. Change color / image on click or mouse hover
4. Show window in slide and close in fade ways
5. Play some small ding dong
6. All messages, even error ones should be funny, like you are saying to your pals.

But as Mouser said, keep an option 'Disable Animation/CPU hog etc.', as user will very soon want a simple cleaner interface to get the job done fast.

Thanks for those ideas. I agree, some of that is good. I especially like panels that are available to slide in/out as-needed, dynamic state change, and prominent but unobtrusive visual indication of status, changes, etc.

Speaking of interesting UI ideas, concepts, approaches, etc. your last comment about having an option for simplification made me think of this:

Any more ideas or examples? Thanks!

- Oshyan

So much 4wheeling! Awesome. They look like fun places. I love that open fir with the "bar" (or spent bottles?) on the "mantle". :D

- Oshyan

Hi folks! I'm in the process of conceptualizing a new graphics program and I'm looking for ideas and inspiration in user interface design. I would love to see your favorite examples of excellent, well designed, intuitive, even "clever" User Interfaces in applications (not websites). Aesthetics matter too, but I'm mostly interested in functional design.

It would be especially helpful to see examples of applications with complex functionality that is exposed through an intuitive UI. 3D programs, music programs, etc. Creative applications, content creation, that kind of thing.

Ideally examples would be from desktop applications, but mobile apps (phone/tablet) are useful tool. Basically any UI that has ever made you go "wow, I wouldn't have thought of that" or use words like "inspired", "ingenious", or "slick" to describe it. The kind of UIs that are immediately intuitive and easy to use, appealing to look at, and a pleasure to interact with. With everything where it "should" be and no actions taking more steps than seem necessary.

Examples of UIs with some nice ideas and "touches", but that are not necessarily a cohesively awesome UI all around also helpful. But I'm really looking for excellent all-around design examples.

Thanks much!

- Oshyan

Pages: [1] 2 3 4 5 6 ... 106next