Harbour and AI: Renaissance or stagnation? A call to the community

[marcos...@gmail.com]

Hello everyone,

Harbour is a great language, but it is not well known by many people. Many still consider it simply as a tool for updating programs written in Clipper.

The number of people who use it is very small. We have a very stable community, but it is small. It does not grow, and there are no large companies backing it.

I believe we are at a critical moment. With the arrival of AI, this is the time to overcome the difficulties we have faced or it could be the end and the definitive stagnation of the language.

I have tested many AI models, and very few of them manage to create error-free programs. This is due to a lack of knowledge about the language's syntax, since it is a niche language, there is little code on GitHub for the models to train on properly.

For my part, I will put all my effort into ensuring that Harbour and its graphic library, MiniGUI Extended (which is the one I use), can position themselves better.

To achieve this, I have started doing the following:

I have created repositories on GitHub with documentation and examples for Harbour.

How can you help?

1. Give stars to the repositories to increase their visibility.
2. Create your own repositories with technical information and complete system code.
3. Follow or create social media accounts dedicated to publishing content about Harbour.
4. Generate new content or follow those who are already creating it.
5. Create libraries that solve common problems when developing systems, so more people can focus on solving their specific needs, and the Harbour MiniGUI Extended ecosystem can grow.
6. Achieve, by all possible means, that Harbour is recognized as a modern language capable of competing with any other.
7. Comment on the language's maintenance forum with new features you believe are important to implement.

Some GitHub accounts and blogs that create content:

GitHub  
https://github.com/marcosjarrin/Harbour_oop_manual  
https://github.com/marcosjarrin/harbour-tech-manual-and-examples  
https://github.com/VictorCasajuanaMas/HB_ORM  
https://github.com/FiveTechSoft/HarbourBuilder  
https://github.com/ivanilmarcelino/designer  
https://mod-harbour.github.io/mod_harbour.v2/

Blog  
https://miniguiextended.blogspot.com/

Instagram  
Account dedicated to Harbour:  
https://www.instagram.com/juanperez95123/

Groups  
https://groups.google.com/g/minigui-forum  
https://groups.google.com/g/harbour-users  
https://groups.google.com/g/harbour-devel

YouTube  
https://www.youtube.com/@IvanilMarcelino  
https://www.youtube.com/@BadaSystemIA

Facebook  
BadaSystem:  
https://www.facebook.com/profile.php?id=100045542240165

Nexo Tecnológico:  
https://www.facebook.com/groups/1652593052612383

Harbour MiniGUI:  
https://www.facebook.com/groups/366493540133705

Harbour 2 PHP – DevTalks & AI:  
https://www.facebook.com/groups/863412360879659

We are facing a challenge as a community. Either we take advantage of this moment, and with the help of AI, we boost the language we use, or we fall into oblivion.

Sincerely,

Marcos Jarrín

[hmpaquito]

Gracias D. Marcos !

[Bob G3OOU Gmail]

Good Morning Marcos

Well said. I will add a new section on Software to my technical web site which is mostly aimed at hardware engineering (RF Communications) to try and increase the visibility of Harbour products.

Best wishes 

Bob

Bob F Burns
C Eng, FIET, MSE. G3OOU
Retired Electronics and Software Engineer
Admin web site: http://www.g3oou.co.uk
Technical web site: https://www.qsl.net/g3oou/

--
You received this message because you are subscribed to the Google Groups "Harbour Users" group.
Unsubscribe: harbour-user...@googlegroups.com
Web: https://groups.google.com/group/harbour-users
---
You received this message because you are subscribed to the Google Groups "Harbour Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to harbour-user...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/harbour-users/ca1615f3-faa0-4703-b959-8d80a3d78706n%40googlegroups.com.

[Auge & Ohr]

hi, 

on Youtube is a Fivewin Video from Antonio Linares about AI

https://www.youtube.com/watch?v=F--iXdyNLmw

Jim

[antonio....@gmail.com]

https://github.com/FiveTechSoft/hbide

[miro....@gmail.com]

https://github.com/mirogeorg/HarbourHarvestedResources

[Concentra Informática]

Hi.

This isn't exactly what is being discussed in this thread, but I believe
it is related to the subject and relevant.

In my opinion, Harbour needs a stable version.

As it stands, it looks like an abandoned tool. The last stable version
is MANY years old.
It might not make a difference to regular users, but to outsiders, it
looks stagnant.

I’ve tried discussing this in the past (more than once), but the
conversation always ended with the point that managing versions was a
huge undertaking, and those with the technical skills to do it didn't
have enough free time for the task.

Antonio, if you’re reading this: I believe that, right now, AI can solve
this time issue very efficiently.

Unfortunately, I don't have the technical expertise for this task
myself. Even with AI, it needs someone with deeper Harbour knowledge to
oversee what the AI ​​is doing.

At last, I know this isn't the right forum for this question—it should
probably be in the Harbour developers' forum—but I felt the opportunity
was relevant.

Thanks to everyone for your time.
Maurício Faria

[antonio....@gmail.com]

Hi Maurício,

Thank you for raising this — your concern is valid, and I think many of us share it.

You are right that, from the outside, Harbour can look stagnant: the last official stable release is 3.0.0 (July 2011), while the development tree has been at 3.2.0dev for many years. That gap creates a perception problem, even for people who use the project daily and know it is actively maintained.

However, I would not describe Harbour as abandoned. If you look at the current repository (github.com/harbour/core (https://github.com/harbour/core)), you will find:

• Regular commits (including fixes as recent as May 2026)
• CI builds on Linux, Windows, macOS and additional VM environments
• A maintained ChangeLog.txt with detailed entries
• Packaging infrastructure already in place (package/mpkg_src.sh, HB_BUILD_PKG=yes, nightly Windows binaries on SourceForge)
• An active contributor base (Czerpak, Szakats, Czajczynski, and others)

So the bottleneck has never been lack of development — it has been the release process itself: deciding what qualifies as "stable", testing across all supported platforms/compilers, preparing binaries, writing release notes, updating SourceForge, and coordinating it all with a small group of volunteers with limited time. That is why past discussions ended at "it is too much work."

On your point about AI: I agree it can reduce the mechanical burden significantly — generating release notes from ChangeLog.txt, preparing packaging scripts, drafting checklists, even helping triage regressions. But you are also right that it cannot replace Harbour expertise. Someone who understands the compiler, RDDs, GTs, cross-platform quirks, and backward-compatibility guarantees still needs to review and sign off. AI accelerates the process; it does not remove the need for a maintainer.

For practical purposes today, users who want something newer than 3.0.0 can use:
• The nightly Windows builds (updated daily from master)
• A fresh clone of the Git repository

But I fully agree that this is not the same as having a proper 3.2.0 stable (or similar) that outsiders can point to.

If there is interest in reviving the release discussion, the right place is indeed the Harbour developers' mailing list. A realistic first step would not be a perfect release, but a scoped one: pick a snapshot of master that passes CI, flip HB_VER_STATUS from "dev" to "" (or "rc1"), run the existing packaging scripts, and publish. AI can help with everything around that — changelog curation, release notes, regression checklists — while core maintainers focus on the technical sign-off.

Thank you again for bringing this up. The perception matters, and so does having a clear, downloadable stable version.

Best regards,
Antonio