The was discussion about this problem in the Machinekit chat room. The truth is that the community support in Machinekit is and was not good (being tactful) and with changes in development it only got worse.
The documentation is bad and obsolete, the site itself is pretty chaotic and finding information is almost Sisyphean task. Everybody recognises this. The changes happening in development on Github were not manually propagated here to Google groups. Moreover, the links at top are again long obsolete in its step-by-step tutorial value (but the overall explanation and goals are still more or less valid). This is a lot more murky than to say "The site needs replacement" - because Machinekit always recognised the need for the user to be proactive and keep a finger on the pulse of development (and encouraged to do some development themselves). No version of Machinekit (meaning Machinekit organization's project here and in whole text) is production ready in the sense of guaranteeing stable version and high level of support. With the understanding that if somebody wants to or need to, he should produce his own stable branch (with implementing upstream changes as he sees fit). That was the status quo for companies using Machinekit in their commercial offerings. However, the community forum is good for something like user issue sharing, basic supporting questions and proud presentations of own use of Machinekit. Just people should not forget that there is also the Github issue tracker specific to each project used for developer talk (which does not mean that only developers or contributors can comment on presented issues). This out of the way, let's talk about how to reinvigorate the Machinekit community (because otherwise all this is useless). The website with repository README page are first point of contacts with new users. Usually what I need to know about new OSS is WHAT is it, WHY does it exist, HOW can I use it and WHEN should I use it. Current status of Machinekit site did not change much from the time Machinekit was forked. It is Jekyll based site with own theme build from Machinekit-docs repository with use of Github Actions service and deployed to Machinekit.github.io repository as Github Pages. I think this is a nice setup. (Not the Jekyll part, I don't care if Hugo, Gridsome, Docusaurus or whatever else is used, but the fact that it is basically static site which can be hosted pretty much anywhere and quickly replicated in case of problem by anybody.) The build is based on a very precarious Docker image (which is based on Debian Jessie and cannot be rebuilt anymore). And some parts are broken and no longer building. In other word, bad. So, the path with the least resistance as far as I can see it is to choose some simple theme with documentation support, change colours to ochre and teal, put logo at top and write some basic documentation answering the above presented question. While letting the current site live somewhere online for users interested in archaeology to study. Problem is, it is still going to require many man-hours to accomplish, but it is needed for survival. The other point is this forum. To tell the truth, I hate mail lists. But many people love them. Removing the obsolete links is connected to the website. But maybe the community would be better server with modern interface with common functions known from other places like mentions, responsive design for reading on mobile devices (not nice with current setup), SSO. Tools like Discourse, nodeBB or Flarum are able to somewhat function over emails with Discourse being the most advanced in this regard. So the importation of messages from this group would be possible and hosting it on 1 GB small server should be enough for the size of this community. The questing stays if it was a positive move for the community and not just task for the task itself. Comments welcome. Cern. (alias @cerna on Github) Dne středa 25. listopadu 2020 v 20:46:14 UTC+1 uživatel sliptonic napsal: > I'm speaking out of frustration, disappointment, and a touch of anger. > You've been warned. > > I had such high hopes for this project. I've been using > linuxcnc/machinekit for 13 years. Like most people, I have a love-hate > relationship with it. So much promise and so much frustration all > together. When the fork happened, I was one who was cheering. Forks used > to be a bad thing but git changed that. Forks mean new ideas can be tested > without disturbing the mainline. "Finally", I thought, "New approaches > and solutions. Hope". > > Ha! What a joke. This project has been a let down. > > The technical side has been great. Really some amazing ideas and > progress. > The individual people are some of the nicest around. Helpful, hopeful, > and smart. > > But the project? Absolute disaster. Every attempt at helpful feedback > has been met with words and no action. The C4 concept might work in other > projects but has produced no fruit here at all. Worse, it's prevented > normal participation by increasing the friction. > > Documentation is almost non-existent. Where it exists, it's often > contradictory or just wrong. There are no reference projects and hardly any > videos. The newsgroup is almost silent and the gitter channel/matrix room > are ghost towns where nobody answers. > > The project started with a high value for using git (remember that fork > thing?). Today neither the machinekit-hal or machinekit-cnc repos have any > branches or tags besides master. Seriously, HOW THE HELL do I find the > last known working code!? > > It looks like someone started a major development effort, broke things, > and then walked away and I can't find any discussion about that > initiative. Why is it so hard to understand what the hell is going on? > > Look, if the project is dead, say that. Lead it out to the pasture and > put a bullet in its head and let's get back to making LinuxCNC better. > > If core contributors don't have time right now to continue working, can we > please take the time to mothball the thing properly? > > But if somebody thinks there's still value here (and I seriously hope you > do) then for the love of God can we please do this right? Can we document > what works, consolidate our communication, and lower the barriers to > contribution? > > Does anyone still care? > > Ok, I feel better now. > > > -- website: http://www.machinekit.io blog: http://blog.machinekit.io github: https://github.com/machinekit --- You received this message because you are subscribed to the Google Groups "Machinekit" group. To unsubscribe from this group and stop receiving emails from it, send an email to machinekit+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/machinekit/c2a05607-e3e8-4cf0-8cbd-a1f92dcaa988n%40googlegroups.com.
