Hello, former CT users,

What software have you ended up using after CT and how easy or hard was to import your data?

I bought a new Windows 11 laptop, and I'm wondering what software I should switch to, as I'm not sure I can install CT on it or if it's worth it.

Not having the CT forum any more I lost access to lots of useful knowledge about tricks and solutions to CT issues, such as how to move CT painlessly to a new machine.

I've been considering Obsidian, as I like the basic interface (haven't tried any plugins yet), but I'm a bit daunted by the task of having to modify my CT data to be sufficiently useful (readable, given all the CT syntax in there, links to image files etc).

I haven't had to deal with regex for years, not looking forward to having to re-learn all that stuff, or to messing around for days with thousands of text files.

Any advice on how to migrate out of CT and where to, relatively painlessly?

Maybe I should add that my main need is to have a knowledge base, Zettlekasten, if you like, where I can have discrete notes with some annotations (tags, categories, images) and wiki links, and good search function.

Basically a personal wiki, and it's a plus if it also has a nice distraction-free writing environment (which I find Obsidian does).

Obsidian in many respects fits the bill, it's just this importing of the CT data that is holding me back.

Hi !

Paul J. Miller (also a contributor to this forum) switched to InfoQube a few years ago...
https://infoqubeim.com/drupal5/node/4481
https://pauljmiller.wordpress.com/2018/10/23/a-comparative-review-of-four-note-taking-programs/
https://welcometosherwood.wordpress.com/2012/10/24/outlining-in-connectedtext/#:~:text=You%20can%20manually%20link%20the,of%20headings%20within%20individual%20topics

HTH !

Pierre Paul Landry
IQ Designer

I may not be able to help much, but just for the record, when CT stopped working I couldn't find a way to get the data from the CT files. Fortunately I had exported my CT files weekly to html files, so I could use those instead. I needed a program which would allow me to export my data as a website, so I replaced CT with RightNote. The import from html wasn't seamless and I needed to work on the layouts, but at least the data was saved.

Graham

Bear in mind that you can always keep running older programs through virtualization (e.g., running Windows 7 on VirtualBox, VMWare, Parallels, or QEMU/UTM). I run ConnectedText that way when I need it.

If you get CT running in a virtual environment, you could then experiment with export options.

If conversion is needed, that would presumably apply to most other programs that you would be importing into (unless, as per what Graham wrote, you import HTML). Anyway, bear in mind that using RegEx or even running Python scripts is a lot easier now with the help of LLMs.

As for alternatives, Obsidian is a strong contender, but I prefer InfoQube, which I appreciate for its robust outlining features. Also, InfoQube tags, like CT categories, can have multi-parent hierarchies.

Many thanks for all the suggestions.

I still have CT running on my old Windows 10 laptop, so I'm able to export in any of the available formats.

I guess one of my problems is that I stopped using CT on a daily basis a few years ago, as Roam and WorkFlowy fulfilled my daily needs.

This means that I forgot a lot of the CT-related practical knowledge of how to use all its features and lost the muscle memory for the various shortcuts.

I always thought I could just go back to the CT Forum for advice, as it had a huge amount of helpful tricks and scripts, such as advice on how to e.g. reinstall CT on a new machine and replicate one's old setup or even AutoHotkey and Python scripts.

But now that this resource is lost, I'd need to re-learn some of the intricacies of the software.

Also, I have a huge amount of links to specific files on that old laptop, so running a clean and new copy of CT in a virtual environment won't help me much, as the links and files would be gone, unless I also import those files into the virtual environment, which would be a huge hassle probably.

Anyway, I'm ready to move on to a different solution. I don't necessarily need to import everything, and if some of the transclusion no longer works, I can live with that.

Thank you for the IQ suggestion. I am aware that IQ is an incredibly sophisticated software, and I have trialled it a few times, but I struggled to overcome its complexity, and it might be overkill for my needs.

Ultimately all I need is plain text, Markdown, and wiki links, in order to create (or re-create) my own knowledge base in the form of a personal wiki, and being able to find things through search.

Using LLM to help with regex is a great suggestion, many thanks for that. So I might be able to do those regex changes even within CT perhaps.

I guess I can't really get away from having to dedicate a few days to reacquainting myself with CT's features and a few more days of cleaning up the data for export.

What appeals to me about Obsidian is that even if it disappears, the underlying folders and files will be there untouched and no need to export anything.

Thanks again.

Have you checked the Internet Archive (archive.org) to see if the Connected Text forum was archived there?

Paul Korm wrote:

Have you checked the Internet Archive (archive.org) to see if the
Connected Text forum was archived there?

Thanks, unfortunately it didn't work, I couldn't log in.

I can't really help, other than to recommend that if you end up NOT choosing Obsidian, you might want to look at Heptabase.

Steve Z.

P.S. It feels like a blast from the past to see your name on the forum, Dr. Andus.

Yes, very nice to see your post, Dr. Andus.

It was one of your posts that caught my eye via Google and brought me here. Any way you could be lured back?

Stephen Zeoli wrote:

I can't really help, other than to recommend that if you end up NOT
choosing Obsidian, you might want to look at Heptabase.

Steve Z.

P.S. It feels like a blast from the past to see your name on the forum,
Dr. Andus.

Thank you for the friendly messages... I never really went away, I was just too busy with other things to post. It's been nice be able to follow the adventures of long-time members here.

The nature of my work took me away from note-taking, outlining and writing for a few years, and for the type of work I was doing, WorkFlowy and RoamResearch were just fine (well, more than fine, I love both and use them every day).

But now I'm pivoting back to more extensive reading, note-taking and writing, and having just upgraded to a Windows 11 laptop, I also need to review which software and what data I can or should transfer from my old machine.

Thank you for all the suggestions. I've been checking out InfoQube, Obsidian, and Heptabase.

For now I'm gravitating towards Obsidian, as I like the interface, the distraction-free writing experience, and the fact that the underlying data can be backed up independently from the software or even read by multiple other software, and on top of that it's free or relatively affordable with the sync.

I do like the Markdown approach, the relative simplicity of it.

Amontillado wrote:

Yes, very nice to see your post, Dr. Andus.

It was one of your posts that caught my eye via Google and brought me
here. Any way you could be lured back?

Stephen Zeoli wrote:
I can't really help, other than to recommend that if you end up NOT
>choosing Obsidian, you might want to look at Heptabase.
>
>Steve Z.
>
>P.S. It feels like a blast from the past to see your name on the forum,
>Dr. Andus.

I moved my notes from ConnectedText to Obsidian. I used two methods for this. Both require that you have access to ConnectedText to export files. I used my old Windows 10 laptop for that.

1) The easier, more recent method
- Export the ConnectedText notes as HTML pages
- Use the official Obsidian importer to import the HTML files
- Then use Visual Studio Code Regex find and replace on the folder of markdown files to fix the internal links: search for \[(.+?)\]\(.+?\.HTML\) and replace with [[$1]]. There's also an Obsidian plugin to do regex across files, but I like the clarity of the VS Code interface for seeing the matches and the replacements that it will do
- Use VS Code Regex find and replace to remove all references to external.gif and ico_pdf.gif. Replace them with blank
- If you want, use Microsoft Power Rename to retitle the files from lowercase to Title Case. https://learn.microsoft.com/en-us/windows/powertoys/powerrename
- Obsidian will remove invalid characters in your note titles like ?: with _. You will need to do more Regex find and replaces to edit the links and file titles so they match up again

2) The harder, older method
- I wrote a Python program that uses Regex to convert CT markup to markdown
- This works on a folder of ConnectedText notes that have been exported as text files
- If anyone is interested, I can find a way to share it, e.g. via Google Drive?

Hope this is useful.

edit: Obsidian will replace invalid characters like ?: with _

Cyganet wrote:

I moved my notes from ConnectedText to Obsidian. I used two methods for
this. Both require that you have access to ConnectedText to export
files. I used my old Windows 10 laptop for that.

1) The easier, more recent method
- Export the ConnectedText notes as HTML pages
- Use the official Obsidian importer to import the HTML files
- Then use Visual Studio Code Regex find and replace on the folder of
markdown files to fix the internal links: search for
\[(.+?)\]\(.+?\.HTML\) and replace with [[$1]]. There's also an Obsidian
plugin to do regex across files, but I like the clarity of the VS Code
interface for seeing the matches and the replacements that it will do
- Use VS Code Regex find and replace to remove all references to
external.gif and ico_pdf.gif. Replace them with blank
- If you want, use Microsoft Power Rename to retitle the files from
lowercase to Title Case.
https://learn.microsoft.com/en-us/windows/powertoys/powerrename
- Obsidian will remove invalid characters in your note titles like ?:
with _. You will need to do more Regex find and replaces to edit the
links and file titles so they match up again

2) The harder, older method
- I wrote a Python program that uses Regex to convert CT markup to
markdown
- This works on a folder of ConnectedText notes that have been exported
as text files
- If anyone is interested, I can find a way to share it, e.g. via Google
Drive?

Hope this is useful.

An interesting journey. Since you’re gravitating to Obsidian for longer, more data-intensive work, can I recommend the plugin “Various Complements”, which is by far the fastest and most convenient way of mining the data in an Obsidian vault. While you’re typing, it automatically scans your markdown files for name matches and pulls them up in a dialog box just below the line you’re typing. You can then create a link to a specific note by selecting the relevant entry in the dialog box. It’s very unobtrusive, but if you’re looking for links with previous work, amazingly useful. It’s also useful for inserting contact details (I’ve transferred my contact management into Obsidian for precisely this reason: another plugin, “@Symbol Linking”, focuses exclusively on a contacts folder, if you’ve got one).

Cheers!
Bill

Dr Andus wrote:

Thank you for the friendly messages... I never really went away, I was
just too busy with other things to post. It's been nice be able to
follow the adventures of long-time members here.

The nature of my work took me away from note-taking, outlining and
writing for a few years, and for the type of work I was doing, WorkFlowy
and RoamResearch were just fine (well, more than fine, I love both and
use them every day).

But now I'm pivoting back to more extensive reading, note-taking and
writing, and having just upgraded to a Windows 11 laptop, I also need to
review which software and what data I can or should transfer from my old
machine.

Thank you for all the suggestions. I've been checking out InfoQube,
Obsidian, and Heptabase.

For now I'm gravitating towards Obsidian, as I like the interface, the
distraction-free writing experience, and the fact that the underlying
data can be backed up independently from the software or even read by
multiple other software, and on top of that it's free or relatively
affordable with the sync.

I do like the Markdown approach, the relative simplicity of it.

Hi Cyganet, thank you so much for taking the time to spell all this out, this is fantastic advice!

May I ask in what respects is 2) the harder method?

It somehow sounds simpler, but then I don't know much about Python, I've only been using scripts that others wrote for me.


Cyganet wrote:

I moved my notes from ConnectedText to Obsidian. I used two methods for
this. Both require that you have access to ConnectedText to export
files. I used my old Windows 10 laptop for that.

1) The easier, more recent method
- Export the ConnectedText notes as HTML pages
- Use the official Obsidian importer to import the HTML files
- Then use Visual Studio Code Regex find and replace on the folder of
markdown files to fix the internal links: search for
\[(.+?)\]\(.+?\.HTML\) and replace with [[$1]]. There's also an Obsidian
plugin to do regex across files, but I like the clarity of the VS Code
interface for seeing the matches and the replacements that it will do
- Use VS Code Regex find and replace to remove all references to
external.gif and ico_pdf.gif. Replace them with blank
- If you want, use Microsoft Power Rename to retitle the files from
lowercase to Title Case.
https://learn.microsoft.com/en-us/windows/powertoys/powerrename
- Obsidian will remove invalid characters in your note titles like ?:
with _. You will need to do more Regex find and replaces to edit the
links and file titles so they match up again

2) The harder, older method
- I wrote a Python program that uses Regex to convert CT markup to
markdown
- This works on a folder of ConnectedText notes that have been exported
as text files
- If anyone is interested, I can find a way to share it, e.g. via Google
Drive?

Hope this is useful.

Thanks, Bill,sounds interesting, made a note of that.

For now I'm not focusing on plugins yet, as I'm trying to think through what data to import and whether to use separate folders or not (or even separate vaults), for instance to separate personal diary entries from Zettlekasten-type reading notes and work-related stuff...

Once I've set things up, I'll take a closer look at community plugins.

MadaboutDana wrote:

An interesting journey. Since you’re gravitating to Obsidian for
longer, more data-intensive work, can I recommend the plugin
“Various Complements”, which is by far the fastest and most
convenient way of mining the data in an Obsidian vault. While
you’re typing, it automatically scans your markdown files for name
matches and pulls them up in a dialog box just below the line
you’re typing. You can then create a link to a specific note by
selecting the relevant entry in the dialog box. It’s very
unobtrusive, but if you’re looking for links with previous work,
amazingly useful. It’s also useful for inserting contact details
(I’ve transferred my contact management into Obsidian for
precisely this reason: another plugin, “@Symbol Linking”,
focuses exclusively on a contacts folder, if you’ve got one).

Cheers!
Bill

Dr Andus wrote:

Hi Cyganet, thank you so much for taking the time to spell all this out,
this is fantastic advice!

May I ask in what respects is 2) the harder method?

It somehow sounds simpler, but then I don't know much about Python, I've
only been using scripts that others wrote for me.

Hi Dr Andus,

The first method requires no specialized skill on the part of the user, just a willingness to follow a series of steps, and not much can go wrong. Worst case, you end up with a set of notes with busted links and you fix them one by one.

The second method requires the user to know how to run a Python program and if it doesn't work for their data, how to edit the program. It also took me several levels of learning to create it: learn Python, learn Regex, write and debug the program, ensure all the Python packages are intalled, run the, then fix all the edge cases that the program didn't cover. It took ages. These days, an LLM can probably make that a lot easier.

I have put a copy of the Python program here: https://drive.google.com/file/d/10aXga9H0Do85Cej6_CmB9Q_BgTOgel7c/view?usp=sharing I hope the link allows you to download a copy to your computer.
You can run it using a Python IDE. I used Thonny and I would expect VS Code to work as well. I just ran a quick test on some of my files and it still works.

Thank you for the explanation and for sharing your file, though you've managed to convince me to go with the first method :-)

Cyganet wrote:

The first method requires no specialized skill on the part of the user,
just a willingness to follow a series of steps, and not much can go
wrong. Worst case, you end up with a set of notes with busted links and
you fix them one by one.

The second method requires the user to know how to run a Python program
and if it doesn't work for their data, how to edit the program.

The good news is that both methods are non-destructive, so anyone can try either one and see what works. Good luck!

Hi Dr. Angus,
It's great to hear from you! I really enjoyed your blog and posts on the CT forum.

I've switched to Obsidian for note-taking and information storage, and Workflowy for outlining. I used Notepad++ to make syntax changes. It took me some time since I'm not very good in technical stuff. Now I'm finally well-settled in obsidian environment.

It was reading Dr Andus' posts here more than a decade ago that led me to ConnectedText. It was a lovely piece of software backed by a superb community. Even after development ceased (a sad story) it was useful on Windows for a while. I migrated many of my notes to DEVONthink, where their usefulness aged out. When Obsidian showed up in private beta April 2020 I started my first vault, which is still my daily note hub, despite numerous CRIMPing side trips to other apps.

I always thought of ConnectedText as the closest thing to Tinderbox on Windows... better than Tinderbox as the learning curve wasn't so steep. I believe CT was the inspiration for Roam and other outline and graph apps. Excellent software.

Hello Tomasz and Paul,

Thank you for the kind words, it's good to hear from you and see your posts, too!

Glad to hear that Obsidian works for you as a CT replacement.

Yes, the first time I heard about Obsidian, it also reminded me of CT, but when it came out, I was mainly working online, on Chromebooks, and didn't want to be tied down to a Windows machine.

It is with a heavy heart that I have to bid farewell to ConnectedText.

I'm keeping my Windows 10 laptop alive as long as I can only to serve as my personal CT museum...

Indeed, if there is a museum for the history of software, it should be there.

Dr. A:
I migrated to "Obsidian App" and am more than impressed.

Very modifiable.

Check YT videos, esp. by a creator called "Danny Hatcher" https://www.youtube.com/watch?v=SAfBpltPx0w

I stopped using CT years ago, but try to export your Topics/Entries to a common format.

Either as a Group and/or Individual Topics.

Perhaps Export your CT entries to most simple text level.

There are already some plugins that facilitate importation of some common backups.

But, consider the power of ChatGPT and/or Grok to

(A) Learn / recall "Obsidian App" use of YAML in *.md files (your target format)

then for whatever backup/export file type you use to exfil your CT content

(A) ask the AI to take either the xml or other output CT file and TRANSLATE IT to YAML / *.md

Try this on various export types OR worst case, feed the AI the raw text of the output file and ask ChatGPT/Grok to transform to yaml/md

you can tag and categorize each note for connectivity.

You may have to re-tag and re-categ your notes. (Or see if the AI can discern and translate CT's tags as defined in it exports.)

"Obsidian App" is free / open design / LOCAL and not on Amazon Servers.

You can adapt to it. I promise.

as you may have seen, "Obsidian App" has many ZK plugins.