mastodon - Link zum Originalbeitrag
Die Medien in diesem Beitrag werden Besuchern nicht angezeigt. Um sie anzusehen, gehe bitte zum Originalbeitrag.

Joplin 3.7 is here! 🚀

AI-powered tools, semantic search, Whiteboards, mobile improvements, better Markdown tables, more reliable syncing - and more.

There’s plenty to explore in the latest release. ✨

Discover everything that’s new: joplinapp.org/news/20260831-re…

What's new in Joplin 3.7 | Joplin

Joplin 3.7 is the release where AI arrives in the app. A new chat panel lets you ask questions about the note you're working on, semantic search finds notes by meaning rather than by keyword, and a...
joplinapp.org
Als Antwort auf mosttoast

mastodon - Link zum Originalbeitrag

Schneckbert 🐌

@mosttoast @oneninesixthree I wish. There is very little (to my knowledge, but I'm happy to learn) with a similar feature set.
And my requirements aren't even very high I think.
E2E, markdown, open source, cross platform and power over how and with what I sync.

AI implementation is an absolute no-go for me, I'm not trying to be the old man and condemn (new) technology, but this is NOT it.

@mosttoast @oneninesixthree
Als Antwort auf Joplin

mastodon - Link zum Originalbeitrag

Blf_tpe

A truly excellent release! Modernized interface, intriguing but useful board, tables in markdown, mcp available, improved sync, all good and brings the app to the next level IMHO 🤩 i feel it has reached a level of polish that makes it on par with the very best.

A bit sad that some see AI and just skip it all only for an optional ability.

Dieser Beitrag wurde bearbeitet. (2 Wochen her)
Als Antwort auf Joplin

mastodon - Link zum Originalbeitrag

Schneckbert 🐌

cursing, strong language, frustration

Sensitiver Inhalt

#enshittification
Dieser Beitrag wurde bearbeitet. (2 Wochen her)
Als Antwort auf Joplin

mastodon - Link zum Originalbeitrag

Schneckbert 🐌

So ... no more donations to #Joplin from me. Opt in or not does not matter; a feature that is not fully removable (by uninstalling an add-on for example).

My notes must not be scanned / harvested by an AI - voluntarily or accidental because of a "bug" ever.

Unfortunately I have no alternative software options, so I have to move to text-editor .md files.

It's pure #enshittification and I have no words how much I hate this #aislop #privacy invasive behavior.

#privacy #enshittification #joplin #aislop
Als Antwort auf Igel31

mastodon - Link zum Originalbeitrag

Schneckbert 🐌

@t28c302 Technology can't be trusted anymore.
Example: Many of us (including me) already knew smart-TV's are essentially spy-hardware.
The physical switch, supposedly turning of the microphones in LG TV's, does not turn it of. *surprise*

Call me paranoid, I don't care. My notes, are my notes and will never touch AI or anyone but me.
(I am not a technology denier, but AI is in 80%+ of cases used for the wrong reasons and unreliable) I do not want it in my notes.

@Igel31
Als Antwort auf Igel31

mastodon - Link zum Originalbeitrag

Schneckbert 🐌

@t28c302 Until they "forget" or ...forget... this feature was turned off. "ups, sorry". No thanks.

I don't have enough time to dive deep into investigating my software constantly.
Unless there a version which doesn't have this at all; I can't use it.
(already uninstalled on my main machine - I can use folders & MD files.
On my mobile device ... I don't know yet. I guess nothing. I'll have to use a physical notebook again then transfer...)

@Igel31
Als Antwort auf Schneckbert 🐌

mastodon - Link zum Originalbeitrag

pmslava

@waldschnecke

I’m a strong privacy advocate too, but I see this differently. AI will increasingly be part of our lives; what matters is how it’s implemented.

This looks like a good example to me: optional tools, support for local models, and useful capabilities without requiring us to hand over our notes. That kind of privacy-conscious approach feels rare these days.

I think the maintainers deserve credit for implementing AI in a way that respects our privacy.

@Schneckbert 🐌
Als Antwort auf pmslava

mastodon - Link zum Originalbeitrag

Schneckbert 🐌

@pmslava AI in a notes app.
Explain what this is good for without a mental handstand.

But regardless, you be you - It's you choice to accept and like this, and mine is not to.

I'm willing to see benefits of AI, but there is almost no current field it is used for good.
Even worse the single only field - medical research - is sufferning because lack of new resources (memory, SSDs). These are literally wasted and "pre-sold" until 2028 to enrich Jensen and a handfull of other billionaires.

@pmslava
Dieser Beitrag wurde bearbeitet. (1 Woche her)
Als Antwort auf Schneckbert 🐌

mastodon - Link zum Originalbeitrag

pmslava

@waldschnecke

One concrete benefit: finding a note by meaning when you can’t remember the exact words you used.

I’m not trying to become an AI advocate, and you’re free to reject it. My point is that AI will be around regardless of whether we use it. Harmful uses will happen without us; useful, privacy-respecting ones need people who care to build and support them. That’s why I think distinguishing good implementations from bad ones matters. Rejecting both won’t leave us with better options.

@Schneckbert 🐌
Als Antwort auf Schneckbert 🐌

gotosocial - Link zum Originalbeitrag

Airikr

I use Joplin, but only for local sync with Syncthing (the best sync). Please link to an article regarding this. I want to read more about it.

PrivacyNotes should be a good replacement.

  • End-to-end encryption
  • Android app as well as softwares to mainly Linux, and a web interface
  • Created within EU (Switzerland)
  • Open source

I found it yesterday and haven't tried it fully, but it looks promising. The only downside for me is that it is not self-hostable.

PrivacyNotes: Your Encrypted Vault for Notes, Tasks & Journals

Note down your thoughts, tasks, and journal - encrypted, always. End-to-end encrypted with zero tracking, hosted in Switzerland.
PrivacyNotes
Dieser Beitrag wurde bearbeitet. (1 Woche her)
Als Antwort auf Joplin

mastodon - Link zum Originalbeitrag

Thomas K.

Unfortunately your “table improvements“ fscked up some documents and the keyboard (IOS) continues to hide and show again and again when editing them. Thx but no thanks.
TBH Joplin becomes more and more the App I did not want. I loved it when I could just edit the plain text, render the markdown for viewing and sync between devices. That’s it! Worsinnovation at its best.
Dieser Beitrag wurde bearbeitet. (1 Woche her)
Als Antwort auf Tizlit

friendica (DFRN) - Link zum Originalbeitrag

Tobias Ernst

@Tizlit @Joplin It is optional and you can point it at your own local AI (or conversly, can use it as a prompt and knowledge storage connected to your local AI via MCP). This is a fair way to integrate AI into Joplin ***without*** compromising data integrity for those who reject AI (and even for those who use it). Let's find a balanced view, some of use need it for work to stay competitive.
@Joplin @Tizlit
Als Antwort auf MrGrumpyMonkey

friendica (DFRN) - Link zum Originalbeitrag

Tobias Ernst

@MrGrumpyMonkey @Joplin Users who start exploring local ai and keep getting told they need Obsidian for memory. Now we can stay with Joplin. Or: Students who keep theor notes in Joplin and need to quickly recall where exactly they "already heard/read/noted" something.

If you don't like it just keep it in tje default disabled state.I petsonally love thr new possibilities because they are implented clean and transpatently and unobtrusively, unlike Copilot and others.

@Joplin @MrGrumpyMonkey

don't like this

Als Antwort auf Joplin

mastodon - Link zum Originalbeitrag

dantealice

I jumped from Evernote to Joplin and have really enjoyed it but...opt-in or no I WILL NOT be complacent to 'AI' (LLM) features and the harm they do in my life. It's destroying our planet and making everything and everyone worse. I don't care if it *seems* inevitable, so did a lot of obsolete things throughout history.

We lived without "LLMs" before and we can do it again.

I don't know where my notes will be stored next, but they won't be with Joplin.

Als Antwort auf dantealice

friendica (DFRN) - Link zum Originalbeitrag

Tobias Ernst

@dantealice @Joplin

I fully understand and to a degree share your resentment. But I think what is really destroying our planet and shoveling wealth to a few extremely rich is the public cloud-based AI. They create cheap subscriptio model user investor captial until lock-in occurs and then they will milk everybody who has relied on it. Just like any new technology humankind invents, it is first exploited by a malevolent few - despite the fact that it could carry benefits for the many. This is also what Microsoft does with Copilot, it is designed to create vendor lock-in and forced upon existing Microsoft users with nearly no way to opt out.

On the positive side, currently there is a lot of progress in the OSS model scene and in downsizing. You can now run a fully competent model on any computer with a recent graphics card, and desktop AI capabilities will continue to grow at very modest power ratings. I have a DGX Spark that reasons sufficiently well for my personal use cases at a mere 135 watts and I am sure we will see further optimization.

Joplin's features for connecting to AI can connect to your local AI. They are not locked in to any particular cloud-based provider. And they are fully transparent as to what they do, how they do it, and how to turn permissions or the entire feature set off or on.

As such, Joplin can become a tool in the movement towards responsible use of AI to the benefit of many instead of the irresponsible exploitation of the many by a few.

Therefore, I actually support this move by the Joplin developers, in spite of my rejection of the cloud-based AI race. And I think in the future all note-taking apps will have baked in AI and Joplin will be the only one which gives you full control on how to use or not use it. So Joplin is still even your best guess in the long run.

Please reconsider moving away from Joplin just because they have provided a new interface for those who want or need to use it.

@Joplin @dantealice
Als Antwort auf Joplin

mastodon - Link zum Originalbeitrag

Schneckbert 🐌

In case anyone else is looking for options without AI, I ended up at "Marknote" apps.kde.org/marknote/

It lacks a LOT of features (like e2e, cross platform...) but I'll only use it locally on my machine, which is already encrypted.
so far all other note taking apps I looked at have some AI options; which is absolutely mind-boggling to me.
Notes are private, why would you allow an LLM scraping this.

Marknote

Write down your thoughts
KDE Applications
Als Antwort auf Schneckbert 🐌

friendica (DFRN) - Link zum Originalbeitrag

Tobias Ernst

@Schneckbert 🐌 Simple: Just as you say, I would never let a public LLM-based autonomous agent like Claude etc. scrape my private notes. The story changes when I run the LLM myself in a simple prompt-response controlled observable manner, give my locally deployed LLM a sub-notebook where it can store memories, or use my locally deployed LLM to analse my private notes. All these scnearios are fully observable and can be genuinely useful. No data leaves my device or my LAN. I really don't get it why y'all abandon Joplin for this. You can simply disable everything (it is even diasbled by the default) and will have the same useful Joplin you always had, plus virtually no bloat as Joplin itself does not ship with any pre-bundled model. The situation is fundamentally different ot Firefox vs. Waterfox (Firefox actually comes with models that you cannot opt to not download, you can just disable them). Even from an ideological point of view, the way in which Joplin now enabled AI interaction is the right way to do it. They do not force anything on anybody, and anything that you can enable optionally is fully transparent. This does not mean Joplin turned to the evil side!
@Schneckbert 🐌
Als Antwort auf Tobias Ernst

mastodon - Link zum Originalbeitrag

Schneckbert 🐌

You're free to have your opinion of course. I can not allow any LLM near my notes. I understand that you segregate and you run it local, but everybody seemed to have forgottent hat these "models" on stolen data, slave labor and nothing other than "un-ethical megacorporation greed". If @joplinapp wants to be asociated with that; so be it.
And again; no; I'm not a technology denier, I work in network security.

Read this, better pull up your pants first.
noyb.eu/de/ai-eu-member-states…

KI: EU-Mitgliedstaaten planen „digitale Enteignung“ der Europäer im Interesse von KI-Unternehmen

In einem geleakten Dokument schlägt die irische Ratspräsidentschaft den EU-Mitgliedstaaten vor, dass das Gewinnstreben von KI-Unternehmen Vorrang vor den Grundrechten der Europäer haben soll.
noyb.eu
@Joplin
Dieser Beitrag wurde bearbeitet. (13 Stunden her)
Als Antwort auf Schneckbert 🐌

friendica (DFRN) - Link zum Originalbeitrag

Tobias Ernst

@Schneckbert 🐌 I am really with you on this one: That would be a very dumb move.

But I have to repeat myself, I use open source models. Even if their training data were scraped illegally and/or against my moral principles, still the company who did that has no material benefit from me using the OSS model they released. Maybe the OSS models are "appetizers" they release for free in hope to make me want to use their paid subscription models which are "even better" but that won't happen. Incidentally, there are European non-commercial efforts such as #OpenEuroLLM which train on legal training data only . Once their model will work sufficiently well, I will switch over.

On the other hand, I cannot afford to ignore and reject the technology outright because others in my business have started using it and I have the choice to use it as well or get left behind. In this scenario, I prefer advocating for responsible use (including not paying the tech brows, knowing what you are doing, and keeping the human in charge) over outright rejection. I am not alone, and Joplin tries to cater for both kinds of responsible users: Those of your type who outright reject it (they get the default setting), and those of my type who try to use it responsibly.

Rejecting the software for giving users the choice feels kind of overdoing it if you ask me.

#OpenEuroLLM @Schneckbert 🐌
⇧