Hey everybody,

I recently downloaded wine on mac (according to this very helpful tutorial: https://www.davidbaumgold.com/tutorials/wine-mac/#part-1:-install-homebrew and the trial version of CT for zettelkasten purposes. I could install it just fine, but I get a weird error when I tried launching CT:

XXX-MBP:ConnectedText XXXr$ wine ConnText.exe
0096:err:process:__wine_kernel_init boot event wait timed out
0096:err:ole:CoGetClassObject class {00000320-0000-0000-c000-000000000046} not registered
0096:err:ole:CoGetClassObject no class object {00000320-0000-0000-c000-000000000046} could be created for context 0x80000001
0096:err:ole:marshal_object couldn't get IPSFactory buffer for interface {00000131-0000-0000-c000-000000000046}
0096:err:ole:CoGetClassObject class {b8da6310-e19b-11d0-933c-00a0c90dcaa9} not registered
0096:err:ole:CoGetClassObject no class object {b8da6310-e19b-11d0-933c-00a0c90dcaa9} could be created for context 0x80000001
0096:err:ole:marshal_object couldn't get IPSFactory buffer for interface {6d5140c1-7436-11ce-8034-00aa006009fa}
0096:err:ole:StdMarshalImpl_MarshalInterface Failed to create ifstub, hres=0x80004002
0096:err:ole:CoMarshalInterface Failed to marshal the interface {6d5140c1-7436-11ce-8034-00aa006009fa}, 80004002
0096:err:ole:get_local_server_stream Failed: 80004002
00a5:fixme:win:EnumDisplayDevicesW ((null),0,0x12df028,0x00000000), stub!

I am really not that tech-savvy and can not really make any sense of that. Does anybody know what the problem is?

Also, a bit more generally: what are your thoughts on using digital zettelkästen? For one, it is not entirely clear to me whether ConnectedText will be developed in the next few years. But equally, I feel like I should be cautious about all programs when we are talking about a timespan of 20-30 years.

Thanks for helping out.

Dr. Andus is the expert on CT and could probably assist with your problem and advice on application for your purpose. Others on this forum also promote CT's virtues and the program has intrigued me. But honestly I just never clicked with it.

Jon

Browza wrote:

Also, a bit more generally: what are your thoughts on using digital
zettelkästen? For one, it is not entirely clear to me whether
ConnectedText will be developed in the next few years. But equally, I
feel like I should be cautious about all programs when we are talking
about a timespan of 20-30 years.

Thanks for helping out.

I'm a regular reader of this forum and decided to weigh in on this as my first post (hi everyone!) because you mentioned zettelkasten specifically. I've been on a long journey of establishing a interconnected, long term notes, having tried everything from dedicated artisanal tools like ConnectedText and Tinderbox, to wiki software like DokuWiki, TrunkNotes, TiddlyWiki, cloud notes apps like Evernote, OneNote, Google Keep, Notion, etc.

It seems that we CRIMPers are always looking for the tool that has the magic combination of functionality and data portability: no one wants to input a large body of data and then be held hostage by a developer, company, or even format, that is no longer active. At the same time, this is 2018 - we should be able to perform advanced, elegant operations with our data that helps us to move beyond what would have been possible even 5 or 10 years ago, right?

The conclusion I've come to is that, at least for myself, there is a tension between feature-rich tools and note longevity. This tension becomes a serious concern over a span of time as long as you've mentioned, 20-30 years. Because I'm erring towards note longevity on this kind of scale myself, I'm willing to sacrifice features, and believe that a plain-text methodology such as the one that Christian Tietze and Sascha Fast have published at zettelkasten.de fits the bill. I didn't want to go this way initially, and found Manfred Kuehn's pushback on the methodology quite persuasive in favor of a tool like ConnectedText. Ultimately, though, I found plaintext to be the easiest way to maintain a digital zettelkasten across Windows, Mac, and iOS, especially now that ConnectedText seems to no longer be under active development.

Hope this helps.

Ultimately, though, I found plaintext to be the easiest
way to maintain a digital zettelkasten across Windows, Mac, and iOS,
especially now that ConnectedText seems to no longer be under active
development.

This is a question we all struggle with. I keep asking myself what 'easiest' means in this context. Does it literally mean the minimum time it takes to manage your info (overhead)? I'm leaning towards using the best software available until it's no longer viable, then doing a 'hard migration' when you need to switch platforms. Which could very well mean copying every single item individually and re-coding the links and metadata. I figure if the difference between the 'best fit' software and plain text is a 20% gain in productivity/usefulness then that's worth several days of hard migration every couple years. I also justify more CRIMPing by telling myself that 'hard migration' is basically a huge review of all your stuff which has a large theoretical value.

If I were to go to a plain text system, I'd take a hard look at tagspaces. Their front-end to a plain text note collection has intrigued me for a while.

I've attempted several times over the years to run ConnectedText in a CrossOver bottle. CrossOver is a hyped up, sort of pricey, version of wine. CT runs OK, but CT's user interface gets ugly when running under wine. Running CT in wine always messes up my CT license.

If you Google the IPSFactory error you'll find numerous threads on the topic. The gist seems to be an InstallShield issue. The advice seems to be "go find a different version of wine". I'm sure that's not helpful.

Browza wrote:

I recently downloaded wine on mac (according to this very helpful
tutorial:
https://www.davidbaumgold.com/tutorials/wine-mac/#part-1:-install-homebrew
and the trial version of CT for zettelkasten purposes. I could install
it just fine, but I get a weird error when I tried launching CT:

Paul Korm wrote:

I've attempted several times over the years to run ConnectedText in a
CrossOver bottle. CrossOver is a hyped up, sort of pricey, version of
wine. CT runs OK, but CT's user interface gets ugly when running under
wine. Running CT in wine always messes up my CT license.

If you Google the IPSFactory error you'll find numerous threads on the
topic. The gist seems to be an InstallShield issue. The advice seems
to be "go find a different version of wine". I'm sure that's not
helpful.

Welp, none of that sounds too promising. But thanks a lot anyways, knowing what not to do gets you closer to knowing what to do.



nblaz wrote:

Because I'm erring towards note longevity on
this kind of scale myself, I'm willing to sacrifice features, and
believe that a plain-text methodology such as the one that Christian
Tietze and Sascha Fast have published at zettelkasten.de fits the bill.

The plain-text-argument is definitely a strong one, as plain-text somehow vanishing is somewhat inconceivable (at least more so than other solutions I have come across).



nathanb wrote:

I figure if the difference between the 'best fit'
software and plain text is a 20% gain in productivity/usefulness then
that's worth several days of hard migration every couple years. I also
justify more CRIMPing by telling myself that 'hard migration' is
basically a huge review of all your stuff which has a large theoretical
value.

That's sort of what Hans Blumenberg did (with a physical zettelkasten) when he moved not from program to program but from project to project. The question for me however, and this is coming from someone whose understanding of technology is sparse, is to what extent hard migration will be possible. I remember that, when I was about 16, I had an incredibly hard time trying to update my grandmother's computer after she did not do it for 5 years, I don't even remember if I succeeded... and there's a difference between something of that extent and substituting all [[ for some other symbol etc.
Then again, if one is somewhat alert to technological change, it should be doable as you said. (especially with plain-text).

I also thought about doing a hybrid of digital/analog notemaking, where I strip down the digital functions to their very core and create the paper notes in a way that would allow me to rebuild a digital version if worst comes to worst...
But maybe I'm just a tad too tired/overthinking.

And as a side-note: It is very nice to see people both so helpful and eager for discussion (especially to/with newcomers as me). It is something that gets seemingly rarer each and every day. Thanks.

Jon Polish wrote:

Dr. Andus is the expert on CT and could probably assist with your
problem and advice on application for your purpose.

Sorry, I'm more of a fanboi than an expert, especially when it comes to this sort of technical problem.

@Browza

Perhaps someone on the CT forum might be able to help you, there are a couple of Mac users there. I have only tried CT in (Android) CrossOver on a Chromebook, and it basically didn't work.

Also, a bit more generally: what are your thoughts on using digital
zettelkästen? For one, it is not entirely clear to me whether
ConnectedText will be developed in the next few years. But equally, I
feel like I should be cautious about all programs when we are talking
about a timespan of 20-30 years.

Once in a while I try to free myself from CT and Windows (most recently have been considering Dokuwiki), but every time I end up sticking with it, as nothing comes close to it.

The lock-in problem is not that bad in CT if you enable its automatic backups, which outputs human readable text files (with the *.cbk extension) to a folder of your choice. So even if CT failed one day to open for whatever reason, the content is safe.

My current Zettelkasten strategy is to stick with a Windows 7 laptop just so I can run CT on it with all the other supporting software (mainly Python 2.7 and Autohotkey) as far into the future as possible.

It would be great if one day I could run CT on a server somehow, so I could access it via a web interface, using a Chromebook.

nathanb wrote:

This is a question we all struggle with. I keep asking myself what
'easiest' means in this context. Does it literally mean the minimum
time it takes to manage your info (overhead)?

This is a great question! I've gotten different answers every time I've asked myself what "easiest" means in the context of a zettelkasten-style note archive.

I've been working with a few contextual parameters that define what "easiest" means to me:

- immediate AND long term data portability: easy access to my data on all of my devices now, while also avoiding being hamstrung by an inactive tool or format with unique or rare features later on down the road
- as frictionless and enjoyable a workflow as possible for data entry and retrieval
- maximizing my time actually using the system, while minimizing the amount of time and effort necessary to maintain the system

I'd be curious to hear more about how other zettelkasten-style notes users define "easiest" in their own contexts!

nathanb wrote:

I'm leaning towards using the best software available until it's no longer
viable, then doing a 'hard migration' when you need to switch platforms.
Which could very well mean copying every single item individually and re-coding the links
and metadata. I figure if the difference between the 'best fit'
software and plain text is a 20% gain in productivity/usefulness then
that's worth several days of hard migration every couple years. I also
justify more CRIMPing by telling myself that 'hard migration' is
basically a huge review of all your stuff which has a large theoretical
value.

I've tried migrating and re-coding my metadata from some of the heavier weight tools I've tried in the past. In the end, I seem to lack the wherewithal to follow through on the process to completion, leaving behind data that I wish was in my current light weight system. I might get around to it eventually, but I doubt it... but might be my own personality quirk :)