Sensorica Next Generation Infrastructure
Lynn Foster
Public
Seen by 228
Sensorica is working on fundraising for OVN infrastructure. There has been much discussion in the Sensorica email list on this. This Loomio discussion is to have a more structured conversation on what the fundraising is for, and what is the agreed upon structure for any project resulting from the fundraising.
NOTE: This discussion is NOT about the fundraising itself, that is happening elsewhere.
General proposal on the table for the possible infrastructure project, for discussion:
* The project team will include all the people who have been working on the infrastructure in any significant way, if they want to join.
* The team (new and old people) will all agree they can work with each other productively before final formation of the team.
* The team will only include people who have proven commitment to the OVN or other next-economy forms, not in it just for the money.
* The team will manage itself. The team will set up its own process methodology and its own value equation.
* The team will make technical and architectural decisions for the good of the project, with input from supportive people who have relevant technical knowledge.
* The role of the R&D and outreach people at Sensorica will be as valued collaborators, and focus on Sensorica requirements and preferences, feedback, and testing.
The other part of the discussion is what possible paths to take, and there have been some thoughts put forward on that. Many of these are not mutually exclusive, so could be worked on in parallel, more a matter of how any funding would be divided up between the finally chosen options.
Some possibilities that have been mentioned already, and I'm sure people can put more forward:
* Strengthen and clean up the existing model layer (where the logic lives), and create an API for that.
* A complete new UI, or multiple new UIs, to include web and mobile interfaces. Or partial new pieces of UI, and keep some of the existing mundane parts. New UI relates to the previous one, because a new model API and new UI would work together.
* Start completely over, build from scratch. (This one is mutually exclusive I guess, except the requirements developed in the NRP could be used as a starting point. Or not.)
* Rethink the base technology. This includes for example using blockchain. Might be other things people are thinking of in this arena. (Note using Ethereum or similar would be at a different layer than the NRP logic itself.)
* Rethink some or all of the model / operations logic. (For example we had started rethinking Exchanges to be much more generic and user defined, inheriting less from the existing world.)
* Emergency maintenance only on NRP, otherwise stop work in favor of other efforts.
* There are some invisible parts of NRP that need improvement: One click installation; new login code; get current with the software libraries we are using; that kind of thing.
* Contribute to the work already going on to develop interoperability between networks: standard vocabulary and protocols, etc. This is needed whether between 2 instances of NRP where there are resource flows between networks in different instances of the software (like Sensorica and the media collective), or 2 instances in totally different software, making it totally flexible what any network chooses to use, as long as each software platform can speak the language to each other.
NOTE: The above is a gathering of earlier thoughts, and is somewhat of a mish-mash.
The goal is for people to come up with specific proposals that everyone can discuss and vote on.
Hoped for result:
1. Agreement on governance and project principles.
2. Agreement on what specific initiatives the funding will support, preferably weighted.
Frederic Durville Mon 5 Oct 2015 7:25PM
@bobhaugen what do you need for the iterative R&D? Or what do you have in mind?
FD
Bob Haugen Mon 5 Oct 2015 7:31PM
@fredericdurville - we haven't supported it well. And when I say well, I mean, pretty much not at all. And we don't have enough experience with it to have any gut feel to go by. I was a manufacturing and supply chain guy, and I have gut feels for that kind of thing. I tried to get an iterative R&D collaboration going with some people in our neighborhood, but they did not understand collaborative R&D, so each person just went off on their own and seldom communicated.
So what we want is a conversation about iterative, collaborative R&D where some people who are doing it talk to us about everything they are doing while they are doing it. It would be sorta annoying. The best thing would be to visit Montreal again and watch and participate, but we can't afford it. Which is why I was trying to get something going nearby.
I know that's not much to go on, but I love unsolved problems and I don't think one is solved yet. Or at least I haven't seen anything that does it well.
Does that explain anything? Clear as mud?
Lynn Foster Mon 5 Oct 2015 8:06PM
@fredericdurville Here are a couple more specific observations, and I'm sure there are many more:
* People like to design and maybe prototype and test more than one option at the same time, and then evaluate which one to go with.
* People tend to iterate the same design - prototype cycle for some unknown number of times. Like, "let's do the same thing but try with this change".
Our current recipes don't support this type of thing well.
Bob Haugen Mon 5 Oct 2015 8:12PM
What Lynn wrote was a lot more practical than what I wrote...;-)
Frederic Durville Tue 6 Oct 2015 12:05AM
Well, I do know about R&D.... I have been involved in R&D almost my entire life...
I would be happy to work with you and see what we can accomplish.
How should we start? Would you like me to start outlining a typical R&D process?
In addition to R&D, there is also what is normally referred to as "engineering", and there is sometimes overlap and confusion between the two processes.
Normally R&D applies to a new product or a new process or new technology. Sometimes R&D is driven by a specific market demand / application, but sometimes R&D is just plain investigation without any specific objective other than understanding.
Engineering normally refers to refinement / change / improvement of an existing product, which may either replace the older product or generate a new product.
A typical R&D process leading to a new product has several "steps", usually starting with a "concept idea". This is typically just a thought on paper. This concept idea then needs to be conceptually or theoretically investigated, and experimentally confirmed. This is a "proof of concept" experimental demonstration. This "proof of concept" is usually a relatively simple experiment. This may require acquiring specific material or equipment, or simply re-using existing material and/or equipment in a new configuration.
Once the "proof of concept" has been successfully completed, the next phase is a "refinement" phase, leading to a first concept prototype. Typically more theoretical analysis and modeling is performed to better understand how the device will function and perform, and some estimation of manufacturability is also performed. This first prototype should provide a complete functionality as desired for a complete and actual device. However, the actual implementation of the prototype in shape and form may be vastly different from the final device. This is just to confirm all the functionality expected from the final device. Several prototypes may be built to be tested in various environments or different places.
At that point, a market study should also be undertaken, to define the market and define the requirements for the device.
The next step is then to build and test a pre-production prototype (or a few). This prototype should be relatively close to the final commercial device. At this point, all manufacturing requirements and specifications should also be worked out. The fabrication of the pre-production prototypes should validate all manufacturing processes. Design adjustments may be made to accommodate or improve manufacturability, or accommodate the customer or the application.
At that point, the new device should be ready to be manufactured and introduced on the market.
From thereon, any improvement or modification should fall under engineering.
Is this a good start?
FD
Bob Haugen Tue 6 Oct 2015 12:09AM
@fredericdurville great start! Thank you very much.
Frederic Durville Tue 6 Oct 2015 12:19AM
@lynnfoster
"* People like to design and maybe prototype and test more than one option at the same time, and then evaluate which one to go with.
* People tend to iterate the same design - prototype cycle for some unknown number of times. Like, “let’s do the same thing but try with this change”."
Yes, indeed. Let's take the example of the soil moisture sensor. The initial concept idea is to use an optical beam and measure how it is reflected with different level of humidity.
So there are several "problems" that need to be solved:
1) how to emit the beam of light.
2) how to deliver the beam of light to the desired location.
3) how to capture the reflected beam of light.
4) how to measure and analyze the reflected beam of light.
5) finally, how to correlate the measurement of the beam of light with the moisture content.
Each of those individual problem can be solved with different solution.
1) can be solved by using either a LED or a laser (or maybe even other sources).
2) can be solved with various types of optical fiber, or by placing the light source directly at the location where we want to do the measurement.
3) can be solved by using a mirror, or simply the natural reflection / scattering from the sample to be analysed.
and so forth and so on.... each possible solution may have its own advantage or disadvantages. So one can do some analysis (pros / cons evaluation) for some of them, and do some actual experimental measurements for some others....
A lot of time the R&D process is completely by-passed, and then it is pure luck if it leads to a successful product.....
I hope all this is helpful.
So please, just let me know how I can help you.
FD
Bob Haugen Thu 8 Oct 2015 10:16PM
@fredericdurville - this is just a ping to let you know that things are moving a little slowly, but your outlines are very useful, and we will use them. We're trying to figure out how to organize these projects.
Frederic Durville Fri 9 Oct 2015 2:05AM
Glad to know that I was able to help...
Ping me again if you need anything else...
FD
Bob Haugen · Mon 5 Oct 2015 5:59PM
@fabioballi - that's a good trade. We're thinking with @mariafrangos about how to organize the UI/UX improvements. More to come in that Loomio thread that Maria started. I'm thinking about how to approach work coordination, and will be asking the Breathing Games crew for suggestions.
I'm looking for somebody who can lead the charge on iterative R&D, and I'll try to follow.