pay for v3

[colttt]

I saw in #6844 that @NGPixel don't have time, which normally means that he don't have time because of mainjob, family and hobbies..
Question is can we throw some money on NGPixel that he can do a few month off from mainwork and push v3?

[janitory]

https://github.com/sponsors/NGPixel

[colttt]

thanks for this hint, maybe we can get the goal (if every user just spent 5€) we will get this asap.
I also wrote @NGPixel a mail to support him via the company

[scottnursten-22]

Let's try make this a movement... I'll happily give a $5/m personally and $100/m from my co... can we get some steam on this idea??

[colttt]

@scottnursten-22 at first it should be in the Readme in that project, also on the website. This year the project gets 2000 new stars, if everyone just spend 5€ we had already achieve the funding.
You could also try it on reddit, twitter, linkedIn, mastodon or/and it-forum

[colttt]

wrote @NGPixel several times a mail, but get no response :-(

[DAE51D]

I'm in I got five on it... 😉

Seriously it's sad to see this amazing project stalled with 113 PRs waiting to go. Maybe @NGPixel should consider blessing some other developers to take the reigns and move the project forward if he doesn't have time (or desire) to continue. I'm invested in this wiki with so many pages that are integral to my daily use, I'd hate to have to find another one and migrate them all. Another option to consider is leveraging Copilot LLM / AI to review the PRs and other code changes to take some of the load off a human; we use that at my day job all the time (and we're a $6B company).

[DAE51D]

With VSCode and Github's own Copilot (or any number of LLMs out there) it's not that hard at all (in fact, it's quite easy these days).

I am a Sr. Software Engineer and I'm coding things I never thought I could in minutes/hours/days ... everything from the two Vobot apps (see my GH) in Python (and I didn't write a single line of python for either), Home Assistant automations, dashboards, helpers, sensors, etc., System administration on my Linux Ubuntu server, hacking my Unifi systems to install NUT on them, all kinds of things... having Copilot PR review is built in to Github too so that helps with the automation of it.

I'd be down to join a team of like-minded individuals so we can move this forward. Or even if we did it in an interim until the OP comes back online and resumes his rightful spot. I'm not trying to "steal" the mans work, I just want to see the tool I've invested so much time and effort in, not die on the vine and continue being an incredibly useful and productive tool that is center to my homelab. There are many deficiencies to be fixed as well as features to be had (as we all know).

What say ye?

[ajmas]

The challenge is being the solo maintainer and no one else with the same code/architecture philosophy to delegate maintenance to. They are out there, but finding someone you gel with in that way can be challenging. Makes me think of finding the successor to Linus Torvalds.

This project is already under the AGPL, so no need to change the license.

If a fork is done, while NGPixel doesn't have time, we need to consider the implications of what comes after (it could create more unintended work). At the very least someone senior to the project should try to get his blessing to do this. There is a parable of where "good intentions" lead to, so I'd rather be respectful with them.

I wouldn't put too much trust in Copilot, even if it a useful tool. Not being human, it won't under the notions of architectural design intent and philosphy. We should still maintain humans understanding why something was done in a certain way and not simply solving 1 + 1 = 1.1. And anyhow thinking about problems and creating stuff for humans is why we code. Let's not simply commission solutions, without understanding.

[drumadrian]

The challenge is being the solo maintainer and no one else with the same code/architecture philosophy to delegate maintenance to. They are out there, but finding someone you gel with in that way can be challenging. Makes me think of finding the successor to Linus Torvalds.

This project is already under the AGPL, so no need to change the license.

If a fork is done, while NGPixel doesn't have time, we need to consider the implications of what comes after (it could create more unintended work). At the very least someone senior to the project should try to get his blessing to do this. There is a parable of where "good intentions" lead to, so I'd rather be respectful with them.

I wouldn't put too much trust in Copilot, even if it a useful tool. Not being human, it won't under the notions of architectural design intent and philosphy. We should still maintain humans understanding why something was done in a certain way and not simply solving 1 + 1 = 1.1. And anyhow thinking about problems and creating stuff for humans is why we code. Let's not simply commission solutions, without understanding.

Amen 🙏

I agree ☝️

My startup plans to sponsor wiki.js when Sales becomes MRR.

💙

[DAE51D]

My startup plans to sponsor wiki.js when Sales becomes MRR.

That's admirable but TBH does no good if the author doesn't even respond to these threads. You could offer to pay him a full time salary but if it falls on deaf ears, it's all for nothing...

I wouldn't put too much trust in Copilot

I work for a $6B company. Copilot is not only encouraged but mandatory in nearly everything we do now and we're talking hundreds of developers/engineers. It automatically reviews our GHA PRs (but a HIL is required as well to put eyes on the PR and final approval). I have 1000% faith in it and it gets better every month. It is integral to my job and I (as a Sr. Software Engineer of 30+ years) am no longer a code-monkey but a software architect basically for the past year or more. It's a force multiplier in that it's like having an extremely intelligent intern that knows all the semantics and every framework and HOW to get stuff done in code, but just doesn't know WHAT to do. My job is now writing prompts and instruction and skills markdown files for the LLM -- so instead of writing code/syntax I write (markdown) instruction files; then we iterate and fix and debug and collaborate. But I digress... How we review PRs or write features is a future discussion if this idea even gets past the initial discussion stages...

[SchneiderSam]

Imagine a wiki with ai... Sad that wiki.js Dev does not respond to this important topic....

[NGPixel]

Let's clarify a few things.

• I already have a full time job that I enjoy. I'm not looking for Wiki.js to become it. It's simply not sustainable and not a comfortable way of life to depend on donations. If you donate, it's because you enjoy Wiki.js and it's a way to give thanks.

• If the development pace is too slow for your liking, this is open source. Fork the project and add the features you need to it. As long as you follow the AGPL license terms, you don't need my approval.

• No offense to the AI fans here, but it's been shown over and over again that "vibe coding" software is a recipe for disaster and ultimately leads to a buggy, insecure, unmaintainable mess. AI is useful to help you code faster or explore ways to do things. If you don't understand what is being generated and can't vouch for every single line of code it produces, you'll eventually hit a wall of epic proportions.