Abandoned Software and Lock-in
https://bicycleforyourmind.com/abandoned_software_and_lock_in

I agree with pretty much all of that. All 3 of my apps store user data in XML. XML is far from perfect, but it is well known standard and text based.

Nice article. Thanks.

macosxguru wrote:

I tried to address your questions in:

Abandoned Software and Lock-in
https://bicycleforyourmind.com/abandoned_software_and_lock_in
#macOS #software

Hard to argue - well said.

I use Curio and would like to use it more. I believe the latest version, which I can't run on my antique Macs, has wider support for external files, putting it closer to applications like Devonthink.

Exports from Curio include Markdown export of the entire project (the whole Curio document). I generally do everything in Curio in Markdown, so that gets a lot of what I created. It's a lifeboat.

True future proof is either plain text or guessable XML. That's my thinking, anyway.

macosxguru wrote:

I tried to address your questions in:

Abandoned Software and Lock-in
https://bicycleforyourmind.com/abandoned_software_and_lock_in
#macOS #software

Andy Brice wrote:

If you take a Windows release of my software from 20 years ago, there is
a good chance it will run on the latest Windows OS.

Indeed - my EccoPro from around 1997 still runs fine when I happen to fire it up, although some things (workgroups, mapquest, etc) may be broken.

macosxguru wrote:

I tried to address your questions in:

Abandoned Software and Lock-in
https://bicycleforyourmind.com/abandoned_software_and_lock_in
#macOS #software

Interesting article, thank you.

But I have a few reservations about it :

I think it comes down to how people tend to think and work with data. You say "I love outlines. I live in outlines." and "For me, with my focus on writing, the solution is “focus on the format.” Text formats are my focus." , and that's a blessing , because you're less hampered by the issue of visual thinking. But I ( and certainly a lot of other people ) tend to think spatially, even in my head. I literally see thoughts floating in my head, and i try to reorganize them spatially.

This is why software like Curio or Tinderbox is so invaluable to me. Curio is like this giant whiteboard that can accept almot any type of media you throw at it, and allows you to reorganize all of it spatially , and even connect them ( in a more limited way than Tinderbox for example ).

It also depends on the type of work you do. I happen to have projects that deal with a lot of multimedia material: images, videos, graphs/diagrams, PDFs, as well as simple text. This is obviously not easy to save into a standard format like plain text or Markdown. And the main value of Curio isn't just that it can absorb multimedia stuff, but the fact that it can -organize- and -map- them spatially in a way that creates more sense and meaning from groups various of multimedia data, and makes new insights possible.

Tinderbox is the same way, athough just limited to text. The value isn't in the day the data is stored, but in the way it's organized and mapped and linked to each other. It's the -linking-, the mapping and organizing that creates more sense in places where it didn't exist previously, new insights and therefore new data. Tinderbox is basically mainly a linking machine ( and also a semi-programmable search and retrieve machine), it's almost its main purpose of existence.

Spatial software ( MindMaps, Curio, Tinderbox, OneNote, Obisdian -when using visual graphs mode-, etc) serves a type of work that is simply either impossible or very hard to do with software that presents data sequentially like outliners or databases. Because the value isn't in how the data is stored, but how it is linked AND presented all at one visually.

Using universal formats , as you concluded in your article, may solve the portability of stored data, in the sense of transfering one storage to another storage. But how do we transport the main value of the work done with these type of spatial apps : The deep linking and complex organizing ( the kind that goes beyond the simple grouping in folders ans subforlders) of data, sometimes of multimedia type ?

I don't know what is the solution, and maybe one should simply accept that there isn't one . Maybe a small consolation would be what Stephen Zeoli wrote earlier, about restarting the work all over again : "So maybe there is some solace in the fact that you’ve already gotten good value from the now-abandoned application. And if you need to rebuild the map in a new app, it is a new opportunity to re-examine your efforts and perhaps develop new insights."

I think it comes down to how people tend to think and work with data. You say “I love outlines. I live in outlines.” and “For me, with my focus on writing, the solution is “focus on the format.” Text formats are my focus.” , and that’s a blessing , because you’re less hampered by the issue of visual thinking. But I ( and certainly a lot of other people ) tend to think spatially, even in my head. I literally see thoughts floating in my head, and i try to reorganize them spatially.

You are absolutely right. I was explicit that I don't do mind-maps, not really. I am not a spatial thinker. So, my solution is different from yours.

Using universal formats , as you concluded in your article, may solve the portability of stored data, in the sense of transfering one storage to another storage. But how do we transport the main value of the work done with these type of spatial apps : The deep linking and complex organizing ( the kind that goes beyond the simple grouping in folders ans subforlders) of data, sometimes of multimedia type ?

I don't have an answer to this question. I can only suggest that you export the data you have in those programs to some other format, but I am not sure that is always possible and definitely not comprehensive or capable of retaining the functionality of the program. The upside is that you might be able to save some of the contents of your work.

I don’t know what is the solution, and maybe one should simply accept that there isn’t one . Maybe a small consolation would be what Stephen Zeoli wrote earlier, about restarting the work all over again : “So maybe there is some solace in the fact that you’ve already gotten good value from the now-abandoned application. And if you need to rebuild the map in a new app, it is a new opportunity to re-examine your efforts and perhaps develop new insights.”

That might be the best we can do.

macosxguru

The new Obsidian Bases feature might hold some promise. Right now, you can have as many views of your data as you want, provided all you want are table views.

Imagine adding Canvas views to Bases. Documents that appear on the fly would either appear at the bottom of the Canvas or maybe in a list of notes not yet positioned.

I remain a die-hard Devonthink enthusiast, though. This morning I need to correlate, link, and organize about 50 PDF documents. I thought about doing that in Obsidian with Bases but I need to tie tags and properties to the PDFs. I don't think Obsidian can do that with a non-Markdown file.

Well, there's an obvious answer. Create a Markdown file for each PDF and embed the PDF in the Markdown.

Or, as I like to say, three lines of Python for a boatload of PDFs.

I believe this issue is far less dire than it was a generation ago.

Take Tinderbox (https://www.eastgate.com/Tinderbox/ which I design. It’s huge, opinionated, and exotic. Its documents are XML files. If Tinderbox disappeared, you could extract all your data with modest difficulty — difficulty that is decreasing every year. (You could get much of your data out by opening your Tinderbox documents in BBEdit!)

Same thing for Scrivener, Pages, Keynote: if you need to, you can get your data out. If LOTS of people need to (for moderate quantities of “lots”) you can make it easy to get your data out.

In addition, a generation ago emulation was pretty much a pipe dream. Now, it’s fine and even ubiquitous.

Now, this doesn’t mean you’re entirely safe from annoyance and inconvenience if your favorite application vanishes. It might take some work. But this is no longer something to inspire 7 pages of worry in a week.

Let's hope Tinderbox lasts a long, long time.

One of the things your app has going for it is a remarkable user community, intellectual, polite, and very helpful. I've never made it to one of your meetups, but I often catch them after the fact on Youtube. Great food for thought and a fine showcase for your products.

Much respect for your work.

Looks like there are some cool new features, too. I re-upped my license and plan to dive back in.

All the best!

eastgate wrote:

I believe this issue is far less dire than it was a generation ago.

Take Tinderbox (https://www.eastgate.com/Tinderbox/ which I design.
It’s huge, opinionated, and exotic. Its documents are XML files.
If Tinderbox disappeared, you could extract all your data with modest
difficulty — difficulty that is decreasing every year. (You could
get much of your data out by opening your Tinderbox documents in
BBEdit!)

Same thing for Scrivener, Pages, Keynote: if you need to, you can get
your data out. If LOTS of people need to (for moderate quantities of
“lots”) you can make it easy to get your data out.

In addition, a generation ago emulation was pretty much a pipe dream.
Now, it’s fine and even ubiquitous.

Now, this doesn’t mean you’re entirely safe from annoyance
and inconvenience if your favorite application vanishes. It might take
some work. But this is no longer something to inspire 7 pages of worry
in a week.

So I encountered a real world example of how work can be stranded in an app. I work for a publisher. We are updating our web store to show which of our other titles are related to each title. You, know, "If you like this book, you may be interested in these." Anyway, I chose to do this in Workflowy, because after I imported the list of our titles, I could easily nest the related titles under each book. That worked fine. But then, when I wanted to export the data and import it in a usable way into Excel, I just couldn't. I dropped the list into Dynalist, and had the same problem. I dropped it into OmniOutliner. No help. Ultimately, I put it into Craft, an app I already use to share documents with my colleagues, and I think this will work.

I am not looking for a solution here -- some of you might have helped me figure it out, but that's not my question. My question is, which app do you trust the most with keeping your information and gives you confidence you can get it out and into other apps? I know this is a fluid question, which depends on the type of information being stored (narrative text vs. outlines vs. structured data), so I don't expect an exact answer. I'm just trying to learn generally which apps this group is confident in.

Thanks in advance for those daring to venture into this conundrum.

Steve Z.

My personal list of note-taking app requirements to ensure longevity of stored information:

- Only use open-source software. Period. Even if the main developer stops working on the software, you can find and pay someone to add features and fix bugs because the source code is available. You could also improve it if you know or learn how to code.

- Only use apps that save data in open formats. Bonus points for tried-and-true text-based formats such as XML, plain ASCII text, markdown. I'm also OK with SQLite databases because it's open-source, portable, and there's a plethora of ways to extract the data from it.

- No web-based tech which immediately rules out Obsidian. Any program that uses Electron will be a memory hog and be at the mercy of the javascript ecosystem (Node.js, Chromium, etc.), OS compatibility, and could be a security risk (any vulnerability in Chromium or Node can potentially be exploited). Not to mention, if something catastrophic ever happened in the world, running Obsidian or Logseq on low-end hardware would be near impossible.

- Desktop first! While having a mobile connection to my notes is convenient, it's not as important as having a solid desktop experience with the ability to intuitively export data, import data, and organize it as I see fit. Bonus points if it's cross-compatible with all of the major desktop operating systems.

After trying many many programs over the years (and still searching), I landed on CherryTree for the foreseeable future. It saves data as XML or in an SQLite database and notes can be exported as plain text or HTML. It's actively developed and open-source. Several years ago the developer rewrote it in C++ (from Python) which improved performance greatly and C++ is a well known programming language. It's simple to outline, organize, format, and find notes. Similar to Obsidian, you can quickly jump to notes by typing the name of the note/node. It's not perfect and lacks some functionality such as easy web clipping, mind mapping, etc. but I can live with that knowing my notes will be around and accessible for a long time.

Stephen Zeoli wrote:

when I wanted to export the data and import it in a usable way
into Excel, I just couldn't. I dropped the list into Dynalist, and had
the same problem. I dropped it into OmniOutliner. No help.

Surprised. Workflowy's OPML should have let OO export to .csv or .xlsx, giving you a nice nested spreadsheet in Excel.

>which app do you trust the most with keeping your
>information and gives you confidence you can get
>it out and into other apps?

"Information" casts a wide net, so I'll restrict my answer to documents. I want apps I can trust for portability, security and no lock-in. So, any textfile-based app would fit, of course, especially one whose files are distinct and either exist as textfiles outside a database or can be easily found within an app's database. For the most privacy and file transparency I suppose the more hardcore options would feature open source apps like Obsidian, Logseq and Joplin.

But text files can too disadvantageous for much use, for many reasons due to its lack of rich features and manual overhead in managing files.

I've never had a problem with a commercial app with mode full-featured file formats as long as they have had good export to open formats like Markdown, JSON, csv, pdf, plaintext - and that includes Microsoft's and Apple's apps. That would also include apps like Notion, Zettlr, Bear and Standard Notes.

Bonus points for apps which support APIs or integrations like Zapier, or services which let you easily download your entire archive.

As far as OPML goes OmniOutliner has proven itself over decades even if the app has been marginally supported in recent years and is in maintenance mode. It's powerful and bommbproof and I have tens of thousands of words in it even though I've increasingly found it clunky to use and I came to dislike writing in it in recent years. I have even more issues with other outliner apps, though. (I like Dynalist but not its pricing, and anyway it too is in maintence mode.)

Despite those preferences about apps I most trust my main writing app Ulysses, with *hundreds* of thousands of words in it, uses a proprietary database, and although one can use it with external files that are local or cloud-based, I don't, instead relying on the app's internal, cloud-synced database (which would be a bear to retrieve files from) to make greatest use of rich formatting and metadata capabilities from it custom Markdown flavor. But it's never been a problem. It is very robust, and I have other priorities like requiring iOS/macOS cross platform sync.

So I don't act on this particular concern. Most people don't, and most don't have to.

Future proofing is something I thought a lot about yesterday. I discovered a couple of new-to-me functions in vim. I stared into the rabbit hole and the rabbit hole stared back. I spent all evening exploring things stuff I'd never seen in vim before.

Vim is ugly. The different modes, Visual (block), Visual (stream), Insert, Replace, Command, Binary, and Normal, might as well be designed to confound the unwary.

At the same time, it's awesome. I may even grow a neck beard, it's that great.

Anyway, plain text is about as stable as cuneiform. A reasonable second choice is ability to export to accepted and open file formats. As long as an app will do either of those, I'm happy.

Yep, agreed – and that's why I now use Obsidian exclusively for all my PKM needs. Everything saved out in text (markdown), excellent tools for web clipping (Obsidian's own, plus plugins like e.g. Slurp – and others), totally cross-platform (works very well on mobile), and thanks to constellation of plugins, hugely flexible.

And it looks nice (unlike VIM)!!!

Coupled with the outstanding plugin "Various Complements" (a ridiculous name for something so powerful), which (a) suggests words as you're typing but far more importantly, (b) suggests links to notes in your Vault as you're typing – in a pleasant, unobtrusive way – Obsidian is also the perfect Zettelkasten solution. Again, there are other plugins which offer "classical" Zettelkasten functionality (numbered/dated/timed notes, etc.), but I prefer the enormous flexibility provided by Various Complements.

There is even a growing number of non-Obsidian-proprietary syncing solutions.

And if Obsidian eventually dies, well, that'd be tragic, but all my data would still be accessible. Giving me a good excuse to turn to another favourite app (desktop only, alas): Octarine, which I use to edit my personal notes on our Nextcloud system.

Cheers!
Bill

Pretty much completely agree about Obsidian. One thing I wish for in Obsidian is global community plugins. It would be nice not to have duplicate copies of plugins in each vault.

I have projects that are sufficiently siloed that one vault wouldn't make sense. Currently, Devonthink is my go-to. I understand as soon as I can update my ancient Macs the new Devonthink has some nice features, such as node maps.

Part of my enthusiasm for vimwiki comes from being able to use it in spite of overly jack-booted policies at work.

It's not that I disagree with jackboots in cyber security, but being told I couldn't have vim on my workstation was a disappointment. So, I run vimwiki on a department Linux system, completely legit in our policies.

But on a workstation? Oh, horrors, no vim! Use Notepad++, I was told.

Then this morning I see my workstation was flagged for a Notepad++ privilege escalation vulnerability.

Some days...