why continue with a tool based on completely outdated technologies that struggles to continue implementing functionalities of Obsidian

[tsaphnat]

why continue with a tool based on completely outdated technologies that struggles to continue implementing functionalities of Obsidian

[timcrouse]

Cherry Tree is far from unusable. It is quite useful in its current state.
Here are a couple of key use cases for its use:
-The notes are its own set of self-contained notes. It is very easy to utilize on a case-by-case basis for dedicated project notes. Easy to transfer to other folks or archive for historical purposes.

-It provides text editing in WYWYG format. For folks that do not need markdown, It should be noted that plain text editing is one of the top requests in the Obsidian community. To date, there has been no effort to implement plain text editing as a core function of Obsidian. Folks have attempted to create a plug-in for this functionality but they seem to come at a cost of only partially working and or crippling other functionality to mimic plain text editing.

-Cherry Tree is portable out of the box. Something that requires tweaking to happen in the Obsidian package.

Best Regards
-Tim C.

[ventolinmx]

@tsaphnat and @adminka-root suspicious campaign against Cherrytree should raise concerns about Osbidian. Obsidian can change its business model anytime and start charging users making everybody that moves there to start paying. They are worried that there are other alternatives because this makes less users captive and dependent on one provider. Obsidian is not open source software.

[mkyral]

When I feel that Cherrytree is not work for me anymore, I will look for replacement. But not sooner.

It is OK to ask to implement some missing feature that is used in other SW, but it is not OK to come here and tell us (and developer) that we should not use Cherrytree and use something else. You can tell us why did you switch to other SW, but forcing others to switch is rude.

[He-Zu]

For me, cherrytree is the perfect note-taking program. I have already tested many note-taking programs. Zim, Obsidian, Joplin, Laverna, etc. None of the programs are satisfactory for me. Far too overloaded with functions that I don't need. I also don't like the look of the programs. Markdown is superfluous for me. For my purposes, Cherrytree is still the very best note program. Clear and easy to use. Best regards

[ventolinmx]

You clearly don't understand open source software development. Cherrytree is good and useful for thousands of users. If for whatever personal issue you hold a particular grudge against it, the question really is: Why do YOU continue to use it? Stop harassing people here, nobody owes you a thing.

[datavectors]

My Philosophy

As one who started my engineering career in 1955 I have seen many initiatives come and go over the years. But some do stick with me and CherryTree falls into that category of favourites. Thank you, Giuspen.

CherryTree format is not a closed format but in common with other prime tools such as Adobe the format is XML. Indeed the entire structure of any CherryTree document can be viewed in an XML editor such as XMLCopyEditor. Elements (nodes) can be parsed by Python scripting (XMLElementTree).

One benefit of CherryTree not often discussed is the integration of codeboxes where different languages can be run. Rather like Jupyter Notebooks.

I dual boot Ubuntu 22.04 and Windows 10 (less so) .

I find that I have many unstructured CherryTree documents scattered throughout my dev desktop and I have decided to bring forward earlier penned ideas to build a repository. Not only is this needed for CherryTree but also other corpora created by different tools.

I have advanced to the point of running eXist-db on jetty in my Ubuntu desktop (It could be Windows). CherryTree documents can be uploaded through eXist-db REST API. in fact if any feature might be added to CherryTree it would be to export nodes as collections to eXist-db. Meawhile this can easily be done by scripting (see actiona below).

Now at this point I admit that I turn to the new kid on the block Chat-GPT. Although I track the mushrooming models.

See the Gartner hype cycle.

https://www.gartner.com/en/articles/what-s-new-in-artificial-intelligence-from-the-2023-gartner-hype-cycle

My desktop agent I consult is Phind.com and where I bounce off my thoughts. Often I have to challenge and correct the advice but it makes development work easier. I did learn that Obsidian only exports Markdown. This makes it a wrong choice for other formats such as XML, SVG, JSON and other advanced types and so I fall back to eXist-db.

Remember that one ultimate goal of collecting this ocean of data is to present it to helpers such as LLM's.

I recently added Obsidian to my dev petri dish as I experiment. But also I have a cohort of other tools to come together as an orchestra of instruments.

These are some of the current players in my orchestra:

CherryTree, Zettlr, Obsidian, eXist-db, Sublime Text, zerotrust data vault, RStudio, D3.js, Krusader,

I spent some time a few years back buried in Atom editor (Electron) until GitHub decided to sunset it. Lesson learned. Initiatives can fail for multiple reasons. Microsoft/GitHub ditched Atom in favour of VS Code.

My philosophy now is to keep an open mind as tools come into view. Take the view that some will fail. Adopt a multi tool philosophy.

There are other classes of tools. A simple tool such as ripgrep-all can rip through a CherryTree (or entire desktop) searching for objects which contain terms such as "Die Hard" (your mention).

Another invaluable search tool is Recoll.

And for serious analysis of corpora we can draw upon AntConc for concordance analysis.

But standing apart from these are what I class as UI automation tools to orchestrate the multiple players in the desktop and the cloud. Including big guns such as Azure, Heoku, Docker etc.

One venerable tool which dates back to circa 2000 Macromedia days (actionscript) is Actiona. I use this as desktop client to orchestrate processes at command line level.

I hope that this jumble of ideas helps. Another kid on the block is vector databases. But study the Gartner hype chart and we see the need to develop prompt engineering tools to get more sensible answers from AI tools.

Klaatu barada nikto!

https://en.wikipedia.org/wiki/Klaatu_(The_Day_the_Earth_Stood_Still)

[Hoffenbar]

Obsidian isn't completely free, only for personal use.

[adminka-root]

As I have mentioned before, the only advantage of Cherry Tree is that it is completely free. Does this automatically make it the best tool? Everyone has their own answer to this question.

Personally, I don't see any issues with Obsidian not being an open-source solution. Why? Firstly, Obsidian heavily relies on community-developed plugins (unlike Cherry Tree!). If Obsidian were to become a paid product, it would greatly upset people, leading to a decline in the creation and improvement of plugins. Consequently, Obsidian would lose a significant portion of its potential. Such a model is not beneficial for the creators of Obsidian. Secondly, if this were to happen, notes could easily migrate to another similar product, possibly an open-source one, with just a simple cursor movement.

However, as I mentioned, in this matter, everyone may have their own opinion, which is subjective and cannot be disputed. It's a subjective factor. As for objective factors, I haven't seen a single argument in favor of Cherry Tree. The true power of personal databases is revealed in using front matter with defined note classes, in filtering notes for better search results. The very idea of Cherry Tree, the "tree-like structure," is detrimental to the end user. It goes unnoticed when there are few notes, but the more notes there are, the more pronounced the time costs become: when finding something you need becomes easier not within your notes but rather simply using Google or ChatGPT.

P.S.:
Alright, Mr. mkyral, this is specifically for you. The original question was directed towards the developer. Your intervention is a manifestation of impolite behavior. The same can be said about your attempt to silence someone afterwards. If you don't like something, feel free to leave. If you are interested, you can participate in the discussion. The choice is yours.

[mkyral]

P.S.: Alright, Mr. mkyral, this is specifically for you. The original question was directed towards the developer. Your intervention is a manifestation of impolite behavior. The same can be said about your attempt to silence someone afterwards. If you don't like something, feel free to leave. If you are interested, you can participate in the discussion. The choice is yours.

I'm user of CherryTree (and open source developer as well) and I don't like possibility that development will stop. Obsidian maybe looks modern, but it is just yet another Electron app that means yet another Chromium running on background. It is not optimal.

Also, my first comment was a reply to a comment that was deleted meantime. So my comment is out of context now.

[giuspen]

I do understand the frustration of requests being ignored but it really is impossible for me to analyse all the requests and at the same time focus on the current feature or bugfix I'm working on. I'm sorry. It obviously helps to bring issues to my attention when multiple users manifest interest on it.

[MJimitater]

@tsaphnat how much are you getting paid? ;)

[timcrouse]

I only get paid in time because this app is SOO AWSOME no markdown mayhem slowing me down, simple text :)

…

________________________________ From: MJimitater ***@***.***> Sent: Monday, January 8, 2024 3:48 PM To: giuspen/cherrytree ***@***.***> Cc: timcrouse ***@***.***>; Comment ***@***.***> Subject: Re: [giuspen/cherrytree] why continue with a tool based on completely outdated technologies that struggles to continue implementing functionalities of Obsidian (Issue #2420) @tsaphnat<https://github.com/tsaphnat> how much are you getting paid? ;) — Reply to this email directly, view it on GitHub<#2420 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AGW34IIEBZXGXLQ4CIXICL3YNRLTJAVCNFSM6AAAAABBK6ZBC2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQOBRG44TOOBSGA>. You are receiving this because you commented.Message ID: ***@***.***>

[mofo1]

CherryTree has been around for years and years, and I've used it for years and years starting in the early 0.20-something when it was a python-built package. I've watched other software come and go and CT is still standing and reliable. And now its based on C++. I use Obsidian upon occasion too but I strongly prefer CT. If I'm being perfectly candid, CT has been around for years whereas Obsidian has been around for about the length of a fart. Thanks, but I'll stick with CT because I know that I can depend on it. Personally, I also love sqlite too.

[ventolinmx]

It would be impossible to work on a structured tree on Acrobat or Word. That's why we use Cherrytree in the first place!

[ventolinmx]

You just got reported and blocked.

[mkyral]

If you want to use different SW, just use it. We don't need to know it.

[adminka-root]

I'm user of CherryTree (and open source developer as well) and I don't like possibility that development will stop. Obsidian maybe looks modern, but it is just yet another Electron app that means yet another Chromium running on background. It is not optimal.

My old cherry tree database is 47 MB in size and consists mostly of text. You know what's not optimal? It's not optimal to perform a search on such a database every time: it takes a really long time. It's not optimal to save it. It's not optimal to subject it to version control. Even simply closing the program takes about a minute.

And most importantly: a large number of unresolved issues on GitHub. It can be concluded that the author develops for themselves and when there is time. This is just his hobby, which he cannot spend much time on. Therefore, it is unlikely that anything will change for the better over the years... And if so, the inevitable question arises: why continue development?

Just because he likes it, because it's his hobby. But at least, a warning could be introduced or a discussion created stating that his product is not suitable for serious tasks and has some drawbacks. A comparative feature table could be created to help inexperienced users decide whether Cherry Tree is suitable for long-term use. Or the same problems may arise, as I experienced, with search of notes and migration to another product.

[adminka-root]

You clearly don't understand open source software development. Cherrytree is good and useful for thousands of users. If for whatever personal issue you hold a particular grudge against it, the question really is: Why do YOU continue to use it? Stop harassing people here, nobody owes you a thing.

Each of your statements seems more like a political slogan than a well-grounded position. Perhaps you are deliberately turning the dialogue into a battle. Or you are simply not able to evaluate things like a man from the point of view of logic, and not from the point of view of emotions and impressions. I suggest you continue to use a pager instead of a smartphone because "it’s useful".

[ventolinmx]

Yea, right. I can't evaluate things like a man of logic but your logic is like this: "I couldn't do this thing with Cherrytree so development should stop and everyone should move to this other thing I want them to use."

You just got reported and blocked.

[adminka-root]

Yea, right. I can't evaluate things like a man of logic but your logic is like this: "I couldn't do this thing with Cherrytree so development should stop and everyone should move to this other thing I want them to use."

You partially understood what I wanted to convey, but you paraphrased it too categorically. Indeed, my life philosophy revolves around the idea of "if you can't do something well, don't do it at all." At least when it comes to people who will use the results of this work. On the other hand, I also understand that the program is part of a hobby, so it cannot be abandoned. The person enjoys writing it, but has little time. That's why I suggested not to stop development but to evaluate the program in terms of other solutions so that the end user can make an informed choice.

You just got reported and blocked.

No need to be upset. I didn't mean anything bad.

[mkyral]

It is opensource, if you don't like current status of development, you can help, sent pull request or even fork Cherrytree and continue development in direction you want. If it will be successful, users will come to you. And they will have their own requests ;-)

I's sad, you have issues with your big DB. Is it reported? Was it analysed?

[adminka-root]

I's sad, you have issues with your big DB. Is it reported? Was it analysed?

Thanks for understanding. No, I did not make a report related to the problems of a large database, since a report without the database itself is useless/not repeatable, and the database is personal, only for me.

But I wrote other reports (suggestions), to which the developer did not react in any way (#2083, #2084).

if you don't like current status of development, you can help, sent pull request or even fork Cherrytree

1. No time. 2. I became against the idea of a tree structure for storing notes. For example, if you have a "Movies --> Drama/Action/Comedy/thriller/etc" tree and the movie note "Die Hard", you cannot uniquely place it in the tree without creating a duplicate. Because this film is both an action movie and a thriller.

[datavectors]

To adminka-root who cites Obsidian it is not a case of CherryTree vs. Obsidian. Harness both of them and others to join the orchestra. Some years ago here I advocated using eXist-db as repo of CherryTree documents and I am only now revisiting this idea, exploring how the various tools can play together in an orchestrated work flow. CherryTree does not need any major development. So I see eXist-db as my repo where I can sync with CherryTree, Obsidian et al. Throw pandoc into the recipe and we can flow between different environments. eXist-db is open source and battle hardened. Just upload CherryTree documents as XML through REST API (my current experiment). And on the example of "Die Hard" cited think of context vectors.

[adminka-root]

To datavectors.

Using machine learning technologies to discover connections between note content and search queries is indeed a great idea. Thank you!

You mention that you use both Cherry Tree and Obsidian simultaneously. Could you provide cases where one tool is more convenient than the other, please?

Overall, I see that you have developed your own conscious and mature approach to note-taking and utilization. I would be interested in delving deeper into it. Are there any resources where you elaborate more on your method of organizing data?

[adminka-root]

I do understand the frustration of requests being ignored but it really is impossible for me to analyse all the requests and at the same time focus on the current feature or bugfix I'm working on. I'm sorry. It obviously helps to bring issues to my attention when multiple users manifest interest on it.

I understand that you understand that people may not like it when their requests are ignored. Many developers design the application code in a way that allows for the easy creation of any user plugins. Moreover, popular plugins can be enhanced and added to the core distribution. This benefits both users and the application creator, eliminating the need to develop new functionality from scratch.

[adminka-root]

To datavectors.

Thank you for the detailed response. The first thing I noticed was your age. You are over 70 years old. As it is known, one of the problems for people who constantly work on a computer is the strain on the eyes. For example, I dread the idea that someday I will have to quit working as a programmer and delve into another field. It's impressive that you have maintained your productivity! Keep it up!

I was not aware of the existence of Phind.com and AntConc. Trying out these tools is undoubtedly a great idea. AntConc, if I understand correctly, does not work with XML. Do you use AntConc to search for data from eXist-db, performing a prior conversion to txt? If so, I assume that you have some index to determine what has changed in the XML since the last conversion to txt? Or is this not necessary, as the conversion process happens instantly?

[datavectors]

Actually started as apprentice at age 16 so you can do the maths.
In those days it was analogue computing.
Must explain that I am having difficulty in managing mouse which has gone berserk, in Ubuntu. Still trying to debug that. So I will keep this short and might writeup separately in a datavault. Or in Windows.

eXist-db has its own search engine. CherryTree uploaded to eXist-db instance as collections can be searched using the eXist-db features. Try installing text collections such as Shakespeare works (after installing eXist-db look in Package Manager) to test the search interface.

Now AntConc is in quite a different class of corpus linguistics. It is just one application in a theatre of linguistics applications.

https://www.laurenceanthony.net/

Adding a note here. CherryTree can export documents as text. File > Export > Export as Plain Text. Then feed a text corpus to AntConc.

On your point about monitoring changes to documents. Inspect CherryTree document in XMLCopyEditor and you see datetime stamps per node. Of course this has to be parsed. Python is good for this. Or Meld will do the job.

You did ask about Obsidian and I found this thread:

https://www.reddit.com/r/ObsidianMD/comments/vc35ey

Finally. Add Zotero to toolset.

[thdoan]

I didn't understand anything @tsaphnat said except for the title lol, but they're not wrong that CherryTree struggles with some fundamentals like performance. I've used Delphi-based TreePad Lite and KeyNote NF extensively in the past, and both of them run rings around CherryTree. Heck, I recently tried Obsidian and even though it's using a webapp framework and gobbles up way more memory, Obsidian feels a lot faster than CherryTree as well.

Here's a typical example of CherryTree's slugglish response:
https://screenrec.com/share/FLhDuA8GYd

In 2024 apps should not render this slowly. This behavior is consistent throughout the app experience. I'm not sure if this performance issue can be fixed apart from something drastic like moving to QT, but I don't see that happening anytime soon.

[silverwings15]

yeah, Cherrytree has unfortunately not seen much development the past few years and is now way behind other modern competitors like Notion, Obsidian, RemNote... that said, i still respect how much work giuspen has done on this project by himself

[thdoan]

Yes, it's one of the more impressive Python desktop applications (along with Zim).