RepRap Development Index

(I'm posting this here in General because it touches on several aspects of the reprap project - but it's mainly about development)

(TL;DR: I think the Development Index of RepRap needs work but i'm not sure what or how. Opinions sought.)

From reading a few forum posts I think I am not alone in feeling that something it missing from the project. As I make the transition from newbie to seasoned newbie my focus is moving towards playing around with new ideas for reprap. Whilst I enjoy the occassional flash of inspiration I am realistic enough to know that others have probably gone through the same thought process, and there has maybe been some work carried out on the idea. It's at this point I feel there is something lacking. I've been trying to think about this over the last few weeks and haven't been able to crystalise my thoughts, so I am writing this asking for the opinions of others in the hope I/we can work something out.

These are kind of questions that I have when thinking of ideas and tinkering:

• What ideas have people had?
• What ideas have already been tried?
• Who is working on what?
• What has been developed?
• Are the ideas I can contribute to or learn from?

To get a feel for the current state of affairs I wrote a script to analyse pages on the wiki tagged as development. I threw the results into a page here (Beware! It's a big page, and it is static, that is, it was generated on May 18th and all time references are relative to this date!) It shows pages by Category, Status and Time of Last Update, and because it is a script I could quickly specify criteria to determine the pages in "Active Deveopment" (definition is at the top of the section), and, being generous, there are about 161 of 336 pages "active". [As a side note: If people find this kind of page useful then I will happily put in a bit of effort to create a bot, or other means of making it dynamically update.]

There are of course many other ideas and designs around - on blogs, thingiverse, github, etc - and google can help find some of them if you use the right terms, but sometimes one wants to browse rather than search. Even with this "analysis", it's difficult to get a feel for what is current and relevant.

Some other thoughts:

• The use of Categories on the wiki does not really aid organisation at the moment and it seems that it takes motivated individuals, who can afford the time, to maintain anything useful. People tend to forget to update their pages (or forget about them altogether in some cases) and then they contribute only to the confusion.
• Nevertheless, perhaps organising development in terms of the print-lifecycle (Model -> Translation -> Firmware -> Electronics -> Frame/Support -> Axis -> Extruder -> Hot End -> Bed -> Material) may help?
• Updating other peoples pages seems intrusive, even though the spirit of wiki's is that we all can edit. Adding a comment to someone elses Thing on Thingiverse seems more comfortable, for example.
• To use the wiki as the central place for tracking development would require someone to take the role of organiser/caretaker/point of contact.
• I remember seeing a blog post or two from Spacexula giving a sumamry of recent wiki activity which were great. These kind of things a bit more regularly would be a boon for the development eco-system I think. (and i'm not suggesting work for Neil here )
• There's not the same sense of community that one gets when working in, for example, github or thingiverse. It's missing the aspects of collaboration that make it easy to be a part of. If someone posts a new thing on Thingiverse I get an update in my RSS reader, I can then comment on the item, make a copy or contribute a derivative back. It's all easy and there's a sense of being involved. In contrast, if I want to find a new development within the wiki I have to actively look amongst the development pages, and the index with it's list of pages tells me nothing useful to help me. The "Talk" page is as good as useless in terms of promoting conversation, and unless I update the page with links going out, there is no way to navigate to similar pages or ideas.

Perhaps we need another tool, page, app or something to bring together the various pages and artifacts that make up "development"? Because of my day job I tend to think that a ticketing/task tracking tool like Jira is missing, i.e. something where people can submit ideas, bugs etc, and track them. But then again, my next thought is always that having something so "project management" oriented would probably suck the fun out of it all. Who needs yet another to-do list hanging over them? However having just a disparate set of pages in a wiki doesn't seem to be doing it either.

I feel that something is needed, i'm just not sure what it is, and so I was hoping people could reply to this post with their opinions. Maybe we can come up with something together? Or maybe I just need to be told to shut up This post is a bit rambling, but hopefully it brings across roughly what i'm trying to say.


Cheers,
Gary

---

------------------------------------------
garyhodgson.com/reprap | reprap.development-tracker.info | thingtracker.net

thank you for putting into words what I have thought for some time, although I to do not have the answer.
I have never felt that a wiki is the best platform for development, something more like a cms, or google blogs, but hosted within reprap maybe, I searched for the ideal software for collabarative work some time ago, but with no success, the nearest thing I found was smf.

---

Random Precision

Writing down my thoughts seems to have unclogged my brain a little.

I think the problem can be broken into two parts: how best to manage the development of ideas and projects for reprap; and how best to discover and track these developments.

For the former, I am coming to the realisation that something like github satisfies many of the criteria I would look for when hosting a development project. Although it's primary focus is to manage software projects the infrastructure it provides fits well with non-code based projects too. For those not familiar with it here are some pertinent points:

[*] Version Control, naturally, using git or Subversion, with all the benefits that come with it: branching, merging, tagging etc.
[*] Ability to fork the project to make your own modifications, and then submit "pull requests" to have these merged back into the parent project
[*] Issue tracker
[*] Network Graph to show who has forked and which branches exist
[*] Can explicitly add collaborators to a project to build a team, or use the informal "fork and pull request" approach
[*] Dedicated Html Site, if required, using Pages, e.g. this page is simply a branch called gh-pages in this project
[*] Wiki which can be open to the public or restricted to collaborators, e.g. [github.com]
[*] Ability to watch a project and be notified via RSS of updates to it's files, issues and wikipages
[*] Free for public projects
[*] If a file called "README" is checked into the project then this is shown on the landing page. A variety of markups are supported


The Prusa Mendel is in github, as are quite few of the firmware projects, and there is even a mirror of Skeinforge too. So it obviously works quite well, but there is still the problem of discovery, tracking and organising, which is largely what my initial post refers too I think. Also, the problem of determining what is relevant still exists - though the ability to directly contact the project owner helps here I think.

---

------------------------------------------
garyhodgson.com/reprap | reprap.development-tracker.info | thingtracker.net

Are you still interested in helping to develop a solution for this? A While ago I tried to develop a website I was going to call, "Open Source Everything". I wanted to be able to track open source projects and sort them by categories as well as obstacles that needed to be overcome. I would like to make it easier for others to jump into open source development and avoid overlap in projects. I think it would be nice to also have a recruitment option so that a project can post an obstacle they have encountered without being sidetracked from their main goal and maybe even offer a reward for helping overcome it. Multiple projects in any category who encounter the same obstacle would be alerted of possible developments towards a solution. Obstacles could be marked, "In development" or "Solved" and even given a rating of importance. Licenses or patents for solutions can be displayed as well as credits. All information should be displayed with a simple, easy to navigate, user interface that can benefit any one or type of project. The site itself can be an open source project. Maintenance and information sorting can be user submitted like Wikipedia by voting up and down changes. I tried start this on my own with HTML, CSS, and JavaScript using a color coded graphical tree view but it quickly became to complex for me to handle on my own.