Ideas? Bringing community connections into the collaborative space

[kat.ingalls]

I'd love to hear more about your ideas for how to collaborate with others. This is something that's been discussed on the team - we think it'd be hugely beneficial to the design process. Especially, as Ron said in this post, when collaborating with people with complementary skill sets. But you guys call the shots.

There's a lot of great projects being shared on the gallery. And there's several projects where I want to introduce two or more people because they're working on similar projects. For example, there's tons of car projects.

The challenge: how do we bring this public connection (via a Project or Discussion) to a collaborative space (i.e. in Fusion 360)? Privacy, different preferences for communicating (social media vs forums vs gallery vs email), organizing and categorizing different skill sets... these are all challenges we're facing.

Looking for a discussion. Have a concrete idea already? Share it on the IdeaStation!

[lure23]

Kat, 

I don't have a working solution for you on that. But I do have a case that has tried to achieve this, focusing on community based design of an electric sports car.

http://www.scarletmotors.com/member-home

I know some of the people involved, and they know me.

There's multiple lessons to be learned by diving in to their forums, and seeing what happened, what not. My short list would be:

+ they managed to attract a whole lot of people with relevant, wide skill set

- they failed to actually engage that community into making something, concrete, and focused

- they probably failed (so far) as a business case

Scarlet motors said they're open but held quite a lot of details behind the curtains, it seems.

You can also look at case Scarlet Motors as: how could Autodesk products have helped them?

I believe the best general scenario for controlled co-laboration is a multi-tiered one.

There needs to be a "seed", a company or an inner group that has the vision and knows what's in (and what's out). Otherwise, there will be no focus as people start to join the wagon.

Then there needs to be a skill market of sorts. Where the "company" can post what is required (will soon be required) and people can raise their hand like "I'll take that". GrabCAD comes to my mind - they are pretty close to this, right?

Management of distributed work force is another issue. Honesty matters - you have to be able to say if something sucks, or if people don't deliver in time.

Ideally, our tools would not make a difference as to whether we are at the office or on the other side of the world. I've worked in wholly distributed teams, and it can be made to work (hint: we didn't use email - we had multiple chat windows open all the time!).

These are just some thoughts. From the social point of view, the management of the Linux project is an interesting case to study. It follows the thoughts presented above (or has influenced my thoughts - don't know which one is true!)

Interesting to see what others will write, and what tools people are using. Thanks for asking, Kat

[kat.ingalls]

Thanks for taking the time to share your insights, Asko! Tons of great info here. I think we actually approached this from two totally different paradigms - which is great! I was thinking more with respect to how to work together in many small, personal projects. It seems like the insights you had come from a single, community-wide challenge. Is that right? 

1. In terms of "seeding," are you recommending that the problem to solve together should be planted? Or is this something that grows organically out of connections?
2. Should the skill market (for example, mechanical analysis vs. human factors design) be decided on by the 'company,' or is this something that users will identify? For instance, if I know that I'm great at conceptualization, idea exploration, form studies, etc - but have a weakness in making a product manufacturable... Should that be something I identify myself? Or do you think the community 'platform' should identify these things? 

Thanks again for the aweseome insights, Asko!

[lure23]

As I wrote, Kat, I don't have the answers. I'm equally interested to see how good collaboration systems can be created. The potential is incredible.

This has a lot to do with the thread on version control. In Linux development, Linus needed to create a new tool (git) because the existing ones were not up to his needs. I haven't used i.e. Autodesk Vault, but I imagine CAD version control not being at the level that plain text file tools (svn, git, hg) are. In a way, this may be more vital than forums etc. because those can be utilized from non-CAD-specific sources (i.e. bug handling systems are abundand, so are discussion forums or instant messaging infrastructures). But a good version control must be integral with the tool (for CAD - for software development they are separate from compilers).

This probably has the same division as software development "GUI camp" (i.e. all tools under one umbrella - s.a. Eclipse or Visual Studio) and Unix mentality of "one tool does one thing well - but only one thing" (i.e. command line work). Fusion is clearly in the GUI camp. This is fine, but understanding that there are other ways to do it (i.e. notifications could be integrated with existing messaging frameworks s.a. twitter) is also important.

As to your questions:

1. The way I have seen open source (software) projects take foothold is that there is a small group of people (1-3 maybe) who are eager about something. Their skill, progress and want attracts others. So it's a combination of having contacts available but also of "seeding" the vision on what should be done. Pinpointing the itch.

The same can be seen in how Kickstarter projects operate. Awesome products, but someone initially had the idea about them.

2. It really should be the peers. An anonymous system of getting kudos/credits on doing certain things right is what matters. "Company" (or platform) is too busy and too distant to really know your skills. Peers who work with you find it out. There's still certain bias in the current systems. I.e. I find myself to be "endorsed" in LinkedIn for Linux skills by people who absolutely cannot have a knowledge on whether I'm good or bad at that. That's because Linux happens to be the most endorsed skill on my profile, and probably gets suggested the most to my contacts. Then they push "endorse" - endorsing me, but for the wrong skill, actually. This kind of bias can be removed from the system, if the people (LinkedIn) were to notice it's there.

In short, I think I'm more for the "Integrate With Everything" model than the "Try To Provide Everything" (i.e. GUI) model, for Fusion 360. You cannot be the best forum, the best bug tracker, the best version control, and the best cloud CAD - all at one time. But do try! 🙂

[stanh1]

Kat,

Great questions.

I don't think I know the answers but it seems that there might be people around the world that would like to work together on things.

Maybe a bulletin board type list of projects to see if people are interested in any of the projects?

Stan